| [2010/07/21 00:01:11] @ Log started by gepetto | ||
| [2010/07/21 00:01:11] <mattock> hi all... it seems I'm having a Class namespace issue I don't fully understand. Any clues why this is not working as expected: http://pastie.org/1053280? I'm using Puppet 0.24.5 on Debian stable (Lenny). | ||
| [2010/07/21 00:01:20] <mattock> oops | ||
| [2010/07/21 00:01:27] <mattock> http://pastie.org/1053280 | ||
| [2010/07/21 00:01:39] <Djelibeybi> mattock: First, upgrade to 0.25.5 or the freshly released 2.6. :) | ||
| [2010/07/21 00:02:03] <mattock> djelibeybi: any other ideas? :) | ||
| [2010/07/21 00:02:38] <mattock> I mean, I'd like to know if is this a bug or a feature/misconfiguration | ||
| [2010/07/21 00:02:55] <ohadlevy> mattock: show your vixie class | ||
| [2010/07/21 00:02:55] <Djelibeybi> mattock: is vixie being included somewhere else? I suspect 0.24.5's autoimport might be at fault | ||
| [2010/07/21 00:03:03] <Djelibeybi> Not even sure if there was autoimport back in 0.24.5 | ||
| [2010/07/21 00:03:20] @ Quit: warreng: Quit: Leaving. | ||
| [2010/07/21 00:03:20] <ohadlevy> Djelibeybi: yet should work | ||
| [2010/07/21 00:05:05] <mattock> vixie class here: http://pastie.org/1053283 | ||
| [2010/07/21 00:05:38] <ohadlevy> mattock: all in the same file? | ||
| [2010/07/21 00:05:41] <mattock> djelibeybi: nope, vixie is not (yet) included from any other class | ||
| [2010/07/21 00:05:45] <mattock> yep, all in the same file | ||
| [2010/07/21 00:05:59] <Djelibeybi> mattock: put class "vixie" first | ||
| [2010/07/21 00:06:06] <Djelibeybi> and move the other two out to their own files | ||
| [2010/07/21 00:06:14] <Djelibeybi> i.e. install.pp holds vixie::install | ||
| [2010/07/21 00:06:22] <Djelibeybi> and config.pp holds vixie::config | ||
| [2010/07/21 00:06:25] <mattock> djelibeybi: ok, I'll try that | ||
| [2010/07/21 00:07:18] @ Quit: itguru: Ping timeout: 258 seconds | ||
| [2010/07/21 00:07:24] <ohadlevy> mattock: yeah, what Djelibeybi said | ||
| [2010/07/21 00:08:10] <ohadlevy> mattock: I am also not sure if your vixie:config will ever do anything | ||
| [2010/07/21 00:08:34] @ TREllis joined channel #puppet | ||
| [2010/07/21 00:08:50] <mattock> ohadlevy: it does, but I can double check that | ||
| [2010/07/21 00:09:03] <mattock> splitting to separate files did not help, the error message is the same | ||
| [2010/07/21 00:10:09] <Djelibeybi> mattock: what does your init.pp for rsnapshot look like? | ||
| [2010/07/21 00:10:15] <Djelibeybi> it may need to be split the same way | ||
| [2010/07/21 00:10:38] <rmt> Thanks, guys. Ciao. :-) | ||
| [2010/07/21 00:10:43] @ Quit: rmt: Quit: Leaving | ||
| [2010/07/21 00:11:17] <mattock> djelibeybi: ...just a sec | ||
| [2010/07/21 00:12:24] <Djelibeybi> mattock: I am somewhat guessing at this point. btw. :) | ||
| [2010/07/21 00:12:44] @ lkoranda joined channel #puppet | ||
| [2010/07/21 00:13:23] @ jmccune joined channel #puppet | ||
| [2010/07/21 00:14:48] <mattock> djelibeybi: full rsnapshot init.pp: http://pastie.org/1053288 | ||
| [2010/07/21 00:15:08] <mattock> and full vixie init.pp: http://pastie.org/1053283 (now split into separate files) | ||
| [2010/07/21 00:15:27] <mattock> still the same error message, and still using "include vixie" works properly | ||
| [2010/07/21 00:15:35] <Djelibeybi> mattock: again, init.pp should start with class "modulename" | ||
| [2010/07/21 00:15:37] @ MattyM joined channel #puppet | ||
| [2010/07/21 00:15:45] <Djelibeybi> and modulename::subclass should be in subclass.pp | ||
| [2010/07/21 00:15:46] <ohadlevy> mattock: you need to include the vixie somewhere, the autoload magic is just to find the actual files defining the classes | ||
| [2010/07/21 00:16:00] <Djelibeybi> mattock: and what ohad said. :) | ||
| [2010/07/21 00:17:26] <mattock> hmm, strange... I've used this type of puppet manifests (with subclasses) with great success :). So I need to use "include <classname>" anyways? | ||
| [2010/07/21 00:17:43] <mattock> init.pp with several (sub)classes | ||
| [2010/07/21 00:18:00] <ohadlevy> mattock: something must include it, otherwise puppet will never know | ||
| [2010/07/21 00:18:10] <ohadlevy> the autoload is to avoid the import statements | ||
| [2010/07/21 00:18:26] <mattock> ohadlevy: ok, now things start to make sense :) | ||
| [2010/07/21 00:18:58] <mattock> and by autoload you mean "require => resource["name"]" ? | ||
| [2010/07/21 00:20:23] <ohadlevy> mattock: no, autoload means that it knows that an install class can be found in the rsnapshot/manifests/install.pp file without importing that file | ||
| [2010/07/21 00:21:33] <Djelibeybi> mattock: autoload means you can do an "include modulename" or "include modulename::subclass" without doing an import | ||
| [2010/07/21 00:22:57] <mattock> ok, cool | ||
| [2010/07/21 00:23:42] <mattock> djelibeybi, ohadlevy: thanks so much for your help! this thing distracted me pretty badly from what I was really doing :) | ||
| [2010/07/21 00:23:47] @ Quit: gebi: Ping timeout: 258 seconds | ||
| [2010/07/21 00:24:43] <mattock> btw. is there a public place where I could write some puppet tips / recipes / notes? they might be useful to somebody else, too | ||
| [2010/07/21 00:24:59] <mattock> and I have trouble finding the relevant stuff from my manifest comments | ||
| [2010/07/21 00:25:08] <Djelibeybi> mattock: the wiki! :) | ||
| [2010/07/21 00:25:24] <mattock> ok, that's what I thought :) | ||
| [2010/07/21 00:25:27] <mattock> just checking | ||
| [2010/07/21 00:25:34] <mattock> thanks! | ||
| [2010/07/21 00:26:40] @ Quit: kolla: Remote host closed the connection | ||
| [2010/07/21 00:28:40] <Djelibeybi> mattock: also, the new Puppet Forge is a good place for recipes | ||
| [2010/07/21 00:29:23] <mattock> ok, I'll check out the Forge | ||
| [2010/07/21 00:31:27] @ Quit: jmccune: Ping timeout: 258 seconds | ||
| [2010/07/21 00:31:31] <eric0> talk about your mixed metaphors. | ||
| [2010/07/21 00:31:48] @ devdas joined channel #puppet | ||
| [2010/07/21 00:35:18] @ benoit__ joined channel #puppet | ||
| [2010/07/21 00:37:35] @ mclarke joined channel #puppet | ||
| [2010/07/21 00:38:18] @ erlingre joined channel #puppet | ||
| [2010/07/21 00:42:17] @ ckauhaus joined channel #puppet | ||
| [2010/07/21 00:42:32] @ spawnyd joined channel #puppet | ||
| [2010/07/21 00:44:57] @ toi joined channel #puppet | ||
| [2010/07/21 00:45:20] @ Quit: gniks: Quit: Leaving. | ||
| [2010/07/21 00:45:35] @ gniks joined channel #puppet | ||
| [2010/07/21 00:47:57] @ alban2 joined channel #puppet | ||
| [2010/07/21 00:50:18] @ Quit: alcy: Ping timeout: 265 seconds | ||
| [2010/07/21 00:52:27] @ Quit: Djelibeybi: Quit: Leaving | ||
| [2010/07/21 00:54:19] @ alcy joined channel #puppet | ||
| [2010/07/21 01:03:45] @ Quit: gniks: Quit: Leaving. | ||
| [2010/07/21 01:05:30] @ Quit: mclarke: Quit: mclarke | ||
| [2010/07/21 01:21:27] @ Quit: yure: Remote host closed the connection | ||
| [2010/07/21 01:22:22] @ pctony joined channel #puppet | ||
| [2010/07/21 01:24:01] @ bronto joined channel #puppet | ||
| [2010/07/21 01:25:12] @ progre55 joined channel #puppet | ||
| [2010/07/21 01:35:24] @ Quit: emiddd_: *.net *.split | ||
| [2010/07/21 01:35:24] @ Quit: davea1: *.net *.split | ||
| [2010/07/21 01:35:24] @ Quit: tmz: *.net *.split | ||
| [2010/07/21 01:35:25] @ Quit: hMz: *.net *.split | ||
| [2010/07/21 01:35:25] @ Quit: FoXMaN: *.net *.split | ||
| [2010/07/21 01:35:25] @ Quit: Maliuta: *.net *.split | ||
| [2010/07/21 01:35:25] @ Quit: cparedes: *.net *.split | ||
| [2010/07/21 01:35:25] @ Quit: crb: *.net *.split | ||
| [2010/07/21 01:35:25] @ Quit: Mandus: *.net *.split | ||
| [2010/07/21 01:35:25] @ Quit: Demosthenes: *.net *.split | ||
| [2010/07/21 01:36:49] @ emiddd_ joined channel #puppet | ||
| [2010/07/21 01:36:49] @ davea1 joined channel #puppet | ||
| [2010/07/21 01:36:49] @ tmz joined channel #puppet | ||
| [2010/07/21 01:36:49] @ hMz joined channel #puppet | ||
| [2010/07/21 01:36:49] @ FoXMaN joined channel #puppet | ||
| [2010/07/21 01:36:49] @ Maliuta joined channel #puppet | ||
| [2010/07/21 01:36:49] @ cparedes joined channel #puppet | ||
| [2010/07/21 01:36:49] @ crb joined channel #puppet | ||
| [2010/07/21 01:36:49] @ Mandus joined channel #puppet | ||
| [2010/07/21 01:36:49] @ Demosthenes joined channel #puppet | ||
| [2010/07/21 01:37:36] @ itguru joined channel #puppet | ||
| [2010/07/21 01:37:58] @ Quit: itguru: Changing host | ||
| [2010/07/21 01:37:58] @ itguru joined channel #puppet | ||
| [2010/07/21 01:39:34] @ Quit: ckauhaus: Ping timeout: 252 seconds | ||
| [2010/07/21 01:42:56] <progre55> hi people! isnt it possible to use a 0.25.5 client with a 0.24.8 server? | ||
| [2010/07/21 01:43:26] <QuackNL> no | ||
| [2010/07/21 01:43:42] <Hilli> progre55: The regular advice is that the server is of the latest version. | ||
| [2010/07/21 01:44:19] <progre55> thanks | ||
| [2010/07/21 01:49:22] @ kolla joined channel #puppet | ||
| [2010/07/21 01:51:03] @ andrewh joined channel #puppet | ||
| [2010/07/21 01:51:56] @ ckauhaus joined channel #puppet | ||
| [2010/07/21 01:57:07] <progre55> the Ubuntu repos only have the latest 0.24.8, so I want to install it from a tarball, the stable version (.25.5) but does the tar package also contain puppetmaster? or do I have to seperately download it? cause I couldnt find it | ||
| [2010/07/21 01:57:23] @ Quit: itguru: Remote host closed the connection | ||
| [2010/07/21 01:57:44] <progre55> oh, apparently it does =) | ||
| [2010/07/21 01:59:21] <Volcane> debian backports have newer ones | ||
| [2010/07/21 01:59:28] <Volcane> use those, install from source sux | ||
| [2010/07/21 02:01:09] @ littleidea joined channel #puppet | ||
| [2010/07/21 02:03:13] <progre55> oh | ||
| [2010/07/21 02:03:27] <progre55> let me check it out | ||
| [2010/07/21 02:05:27] @ tazz joined channel #puppet | ||
| [2010/07/21 02:05:49] @ jab_doa joined channel #puppet | ||
| [2010/07/21 02:07:09] @ itguru joined channel #puppet | ||
| [2010/07/21 02:07:10] @ Quit: itguru: Changing host | ||
| [2010/07/21 02:07:10] @ itguru joined channel #puppet | ||
| [2010/07/21 02:07:17] <progre55> Volcane: well, honestly I've got no idea where to enable the debian backports =) anything in the ubuntu backports? | ||
| [2010/07/21 02:08:24] <Volcane> i have my own debian repos where i cherry pick what i want from debian backports, dont know how normal debian setups work | ||
| [2010/07/21 02:08:36] <barn> http://backports.org/dokuwiki/doku.php?id=instructions | ||
| [2010/07/21 02:08:38] <Volcane> or if enabling debian backports is even a good idea on an ubuntu machine | ||
| [2010/07/21 02:09:09] <progre55> barn: thanks, I'll read on that | ||
| [2010/07/21 02:09:14] <barn> enable multiverse and universe, they're similar things | ||
| [2010/07/21 02:10:06] <progre55> oh and how to find out the version of a package not installing it? | ||
| [2010/07/21 02:10:12] <progre55> I mean, on a terminal | ||
| [2010/07/21 02:13:07] @ Quit: devdas: Remote host closed the connection | ||
| [2010/07/21 02:21:46] @ gebi joined channel #puppet | ||
| [2010/07/21 02:23:29] @ Quit: littleidea: Quit: littleidea | ||
| [2010/07/21 02:32:04] <spawnyd> progre55: use dpkg IIRC | ||
| [2010/07/21 02:32:09] <joschi> progre55: ubuntu lucid ships puppet 0.25.4 | ||
| [2010/07/21 02:32:26] <joschi> progre55: you can easily backport these packages, if you're still using hardy or karmic | ||
| [2010/07/21 02:32:51] <barn> http://packages.ubuntu.com/ "search" | ||
| [2010/07/21 02:33:47] <progre55> joschi: upgrading to lucid actually =) | ||
| [2010/07/21 02:34:10] <progre55> oh and, is 0.25.4 good enough? | ||
| [2010/07/21 02:35:42] <joschi> progre55: works for me[tm] | ||
| [2010/07/21 02:36:20] @ cynicismic joined channel #puppet | ||
| [2010/07/21 02:36:37] <progre55> =) thanks | ||
| [2010/07/21 02:38:42] @ Quit: TREllis: Ping timeout: 265 seconds | ||
| [2010/07/21 02:43:48] @ TREllis joined channel #puppet | ||
| [2010/07/21 02:46:08] @ ffrank joined channel #puppet | ||
| [2010/07/21 02:47:07] <Legooolas> Might be in the docs but I can't find it : Can I add to the requires list of an already-existing item (a service in this case)? | ||
| [2010/07/21 02:47:52] <Legooolas> Won't doing something like "Myclass::Myservice <| filtering... |> { require => File["foo"] } | ||
| [2010/07/21 02:47:59] <Legooolas> override completely rather than adding to it? | ||
| [2010/07/21 02:50:26] <Legooolas> I'm being an idiot anyway, I can just use before => ... | ||
| [2010/07/21 02:50:30] * Legooolas slaps forehead | ||
| [2010/07/21 02:51:27] <ffrank> uhm. isn't before the opposite of require? | ||
| [2010/07/21 02:51:40] <Legooolas> before on the file I want the service to require | ||
| [2010/07/21 02:51:52] <Legooolas> rather than modifying the require on the service to require the file | ||
| [2010/07/21 02:51:56] <ffrank> to add to the list, use { require +> Resource[...] } | ||
| [2010/07/21 02:52:05] <Legooolas> So being the other way around works for that | ||
| [2010/07/21 02:52:15] <Legooolas> Is that a 2.6.0 addition? :) | ||
| [2010/07/21 02:52:27] <ffrank> no. all i know is it worked in 2.5 | ||
| [2010/07/21 02:52:47] <Legooolas> Ah cool, cheers :) | ||
| [2010/07/21 02:53:11] <ffrank> lowers readability somewhat, esp. if your manifest is shared among admins of varying puppet proficiency. i tend to steer clear | ||
| [2010/07/21 02:53:49] <Legooolas> Yeah. Might be useful at some point though, and I'm the only one editing our manifests at the moment. | ||
| [2010/07/21 02:55:19] <cynicismic> how easy is it to migrate the puppetmaster to a new machine? | ||
| [2010/07/21 02:55:25] <ffrank> anyway, here is my problem w/ puppetmaster 0.25.5 (client 0.24.5, don't get me started): i change the manifest on the puppetmaster, but puppetd --test won't produce changes on the client. the manifest was written some 15 minutes ago, still the client thinks it was up to date | ||
| [2010/07/21 02:55:49] <cynicismic> just a case of configuring new host the same then copying over ssl store? | ||
| [2010/07/21 02:56:25] <Legooolas> cynicismic: With the same name too? Don't see why that shouldn't work.. | ||
| [2010/07/21 02:57:01] <cynicismic> name is an alias (puppet.xxx.domain.com) - that's what should be in the cn field of the certs also | ||
| [2010/07/21 02:57:28] <ffrank> cynicismic: you can probably test by just setting it up, then manipulating /etc/hosts on a test client to see if it talks to the new master all right | ||
| [2010/07/21 02:59:53] <cynicismic> ffrank: that's a good idea, will probably go that route | ||
| [2010/07/21 03:00:25] <kuh> with 2.6.0 I can no longer use unless => "( cmd1 ) || ( cmd2 )" - puppetd tries to `which (` | ||
| [2010/07/21 03:00:51] <kuh> is this a known issue? | ||
| [2010/07/21 03:02:32] <ffrank> kuh: as a workaround, can you enclose in "sh -c '(cmd1)||(cmd2)'" ? | ||
| [2010/07/21 03:03:37] <kuh> ffrank: will try, also thought I could do unless => ["cmd1","cmd2"] | ||
| [2010/07/21 03:03:50] <Whoop> Anyone having any problems with 2.6? | ||
| [2010/07/21 03:05:04] @ Quit: davea1: Quit: Leaving | ||
| [2010/07/21 03:12:40] <Volcane> Whoop: plenty of tickets | ||
| [2010/07/21 03:12:56] <Whoop> Aye, thats what I thought :P | ||
| [2010/07/21 03:13:08] <Whoop> All these features look tasty, but dunno whether to risk it so early | ||
| [2010/07/21 03:16:00] @ docelic joined channel #puppet | ||
| [2010/07/21 03:16:16] @ Quit: docelic: Remote host closed the connection | ||
| [2010/07/21 03:16:58] @ docelic joined channel #puppet | ||
| [2010/07/21 03:21:04] @ Quit: spurge: Read error: Connection reset by peer | ||
| [2010/07/21 03:22:34] @ Quit: cristim: Read error: Operation timed out | ||
| [2010/07/21 03:30:18] @ cristim joined channel #puppet | ||
| [2010/07/21 03:31:10] <alcy> does --noop also reflect the exact order in which actions take place ? if yes, is there a definite order it(puppet) observes or decides priority of a particular type (not among different types, but same ones like which file first)? | ||
| [2010/07/21 03:31:31] <Volcane> alcy: there's no order | ||
| [2010/07/21 03:33:15] <Volcane> the only order is between specific relationships that you define with require/notify/before | ||
| [2010/07/21 03:33:28] <Volcane> but even then, thats broadly speaking order, it will happen before, not *right* before etc | ||
| [2010/07/21 03:36:36] <alcy> Volcane: ah alright. completely forgot about adding require statements ! I have something like this http://pastie.org/1053461 and was getting apprehensive if the package will overwrite the configuration template I am providing. I could simply add a require statement to take care of that, I think. | ||
| [2010/07/21 03:37:41] <alcy> a bunch of require statements I mean | ||
| [2010/07/21 03:41:29] <Volcane> your OS really shouldnt be overwriting config files :) | ||
| [2010/07/21 03:42:35] <alcy> hmm, right | ||
| [2010/07/21 03:46:05] <ffrank> also, the next run will fix it back. requires can hit readability, i think twice whether i really want them | ||
| [2010/07/21 03:46:35] <Volcane> dont think they affect readability much | ||
| [2010/07/21 03:46:50] <Volcane> i use resource requires just always more or less inside the same class | ||
| [2010/07/21 03:47:00] <Volcane> if i need to depend on something in another class i always require the class | ||
| [2010/07/21 03:47:49] <Volcane> that avoid huge arrays of many resources in a require line | ||
| [2010/07/21 03:48:22] <alcy> readability as in literally, readability of code ? | ||
| [2010/07/21 03:49:39] <Volcane> yup | ||
| [2010/07/21 03:49:44] <Volcane> very nb | ||
| [2010/07/21 03:53:46] <alcy> oh. as of now I don't care about that. plus why should I wait for things to get right in the next run ? | ||
| [2010/07/21 03:56:13] @ alban21 joined channel #puppet | ||
| [2010/07/21 03:59:58] @ Quit: alban2: Ping timeout: 260 seconds | ||
| [2010/07/21 04:01:42] @ Quit: ahasenack: Ping timeout: 265 seconds | ||
| [2010/07/21 04:04:23] <alcy> I might be wrong, obviously. :) | ||
| [2010/07/21 04:04:41] <Volcane> its best to add appropriate requires etc | ||
| [2010/07/21 04:06:49] <ffrank> the point used to be in coding best practices, but i'm not sure about now | ||
| [2010/07/21 04:07:56] <alcy> ffrank: really, I am learning, so I don't know why you said it will get fixed in the next runs ? that just leaves more room for inconsistencies, right ? | ||
| [2010/07/21 04:08:36] <Volcane> too much eventual consistancy just makes it hard to truely know when say bootstrapping machines in an automated manner when you're actually done for example | ||
| [2010/07/21 04:08:39] <Volcane> aim should be a single run | ||
| [2010/07/21 04:12:27] @ MPSimmons joined channel #puppet | ||
| [2010/07/21 04:14:32] @ herdingcat joined channel #puppet | ||
| [2010/07/21 04:16:41] @ Quit: tazz: Ping timeout: 265 seconds | ||
| [2010/07/21 04:31:16] @ Quit: spawnyd: Ping timeout: 276 seconds | ||
| [2010/07/21 04:32:08] @ tazz joined channel #puppet | ||
| [2010/07/21 04:39:23] @ Quit: alexine_dsouza: Read error: Connection reset by peer | ||
| [2010/07/21 04:56:21] @ Quit: andrewh: Read error: Connection reset by peer | ||
| [2010/07/21 05:01:18] @ Quit: erlingre: Ping timeout: 240 seconds | ||
| [2010/07/21 05:03:34] @ Quit: tazz: Ping timeout: 265 seconds | ||
| [2010/07/21 05:05:14] @ alcy left channel #puppet () | ||
| [2010/07/21 05:06:06] @ Quit: \ask: Ping timeout: 276 seconds | ||
| [2010/07/21 05:07:15] @ ahasenack joined channel #puppet | ||
| [2010/07/21 05:10:33] @ ghaskins_mobile joined channel #puppet | ||
| [2010/07/21 05:10:33] @ Quit: ghaskins_mobile: Quit: Leaving | ||
| [2010/07/21 05:10:51] @ ghaskins_mobile joined channel #puppet | ||
| [2010/07/21 05:11:54] <ghaskins_mobile> hey all, im still struggling with getting providers/plugins loaded on the client. ive read all the docs and googled like crazy, but I am still missing something | ||
| [2010/07/21 05:12:06] @ Quit: suit: Quit: leaving | ||
| [2010/07/21 05:12:25] <ghaskins_mobile> anyone familar with utilizing a provider generally? | ||
| [2010/07/21 05:13:38] <Volcane> you'll have better results if you just ask a question providing as much information as possible via something like pastie.org | ||
| [2010/07/21 05:13:55] <ghaskins_mobile> Volcane: ok, thx, will do | ||
| [2010/07/21 05:14:17] <ghaskins_mobile> i had posted more context last night so that was supposed to be a continuation of that dialog, but point taken | ||
| [2010/07/21 05:14:40] <Volcane> different people, different times of day :) | ||
| [2010/07/21 05:15:02] <ghaskins_mobile> heres what I am trying to specifically do: i want to use the zypper package provider against some managed suse nodes | ||
| [2010/07/21 05:15:20] <ghaskins_mobile> and I am struggling gettting it to work | ||
| [2010/07/21 05:15:55] <ghaskins_mobile> i had followed the doc on modules and plugins, and I can see zypper.rb is included in my distros puppet.rpm | ||
| [2010/07/21 05:16:27] <ghaskins_mobile> i have the clients setup for pluginsync=true, which resulted in some complaints that it couldnt find puppet://puppet/plugins | ||
| [2010/07/21 05:16:45] <ghaskins_mobile> i fixed that by setting up my fileserver.conf with a [plugins] share | ||
| [2010/07/21 05:16:49] <ghaskins_mobile> (currently empty) | ||
| [2010/07/21 05:16:54] <Volcane> but you say the client already has zypper.rb on them right? | ||
| [2010/07/21 05:17:11] <ghaskins_mobile> Volcane, well, no, i think that is my problem | ||
| [2010/07/21 05:17:27] <ghaskins_mobile> struggling with figuring out the "right" way to get it down there | ||
| [2010/07/21 05:17:34] <Volcane> ok | ||
| [2010/07/21 05:17:40] <Volcane> what version? | ||
| [2010/07/21 05:17:48] <ghaskins_mobile> I tried setting up $MODULEDIR/zypper/lib/puppet/...blah blah | ||
| [2010/07/21 05:17:57] @ cliff-hm joined channel #puppet | ||
| [2010/07/21 05:18:23] <ghaskins_mobile> my distro says: puppet-0.24.8-2.1.x86_64 | ||
| [2010/07/21 05:18:31] <ghaskins_mobile> i assume that is perhaps v2.4? | ||
| [2010/07/21 05:19:11] <Volcane> ok, show your fileserver.conf and exactly where you put the zypper.rb on the client, also your puppet.conf from both master and clients all on pastie.org | ||
| [2010/07/21 05:19:22] @ [Fate] joined channel #puppet | ||
| [2010/07/21 05:19:24] <ghaskins_mobile> Volcane, ok, will do | ||
| [2010/07/21 05:21:18] @ Quit: vinci: Ping timeout: 240 seconds | ||
| [2010/07/21 05:22:09] <mark> hmm... I was hoping to be able to build an array decentrally using virtual defines, which are appending to a single variable that is initialized right before the collection of those defines | ||
| [2010/07/21 05:22:19] <mark> but the variable doesn't appear to be in the same scope | ||
| [2010/07/21 05:22:50] <ghaskins_mobile> Volcane, btw: be gentle, i am a puppet newbie ;) | ||
| [2010/07/21 05:23:02] <ghaskins_mobile> probably have this completely bass-ackwards | ||
| [2010/07/21 05:23:04] <ghaskins_mobile> but here we go | ||
| [2010/07/21 05:23:06] <ghaskins_mobile> http://pastie.org/1053554 | ||
| [2010/07/21 05:23:27] <ghaskins_mobile> so that is my modules setup, where I tried to create "zypper" to include the provider | ||
| [2010/07/21 05:23:36] <Volcane> that looks fine, not sure if it being a symlink will break things | ||
| [2010/07/21 05:23:43] <Volcane> now the rest of the info :) | ||
| [2010/07/21 05:24:01] <ghaskins_mobile> http://pastie.org/1053555 | ||
| [2010/07/21 05:24:07] <ghaskins_mobile> thats the server puppet.conf | ||
| [2010/07/21 05:24:39] <Volcane> ok, add to the [puppetmasterd] section | ||
| [2010/07/21 05:24:43] <ghaskins_mobile> http://pastie.org/1053556 | ||
| [2010/07/21 05:24:50] <Volcane> or to [main] | ||
| [2010/07/21 05:24:53] <Volcane> pluginsync = true | ||
| [2010/07/21 05:25:38] <Volcane> ghaskins_mobile: and remove the directory from that fileserver.conf - take away the path line | ||
| [2010/07/21 05:25:41] <ghaskins_mobile> and this is the client: | ||
| [2010/07/21 05:25:42] <ghaskins_mobile> http://pastie.org/1053557 | ||
| [2010/07/21 05:26:06] <ghaskins_mobile> Volcane, so I need pluginsync on the server side as well? | ||
| [2010/07/21 05:26:10] <Volcane> yup | ||
| [2010/07/21 05:26:13] <ghaskins_mobile> ah! | ||
| [2010/07/21 05:26:17] * ghaskins_mobile goes to try | ||
| [2010/07/21 05:26:18] <Volcane> it needs to know to look for plugins | ||
| [2010/07/21 05:27:28] <ghaskins_mobile> Volcane, should I just remove the entire "path" line from fileserver.conf or just set it to empty | ||
| [2010/07/21 05:27:35] <mark> defines don't introduce a new scope, do they? | ||
| [2010/07/21 05:28:00] <Volcane> ghaskins_mobile: remove | ||
| [2010/07/21 05:28:06] <Volcane> mark: they do | ||
| [2010/07/21 05:29:15] <ghaskins_mobile> Now I seem to have re-broken the plugin sync | ||
| [2010/07/21 05:29:16] <ghaskins_mobile> http://pastie.org/1053559 | ||
| [2010/07/21 05:29:44] <Volcane> restarted the master? not sure if its needed | ||
| [2010/07/21 05:29:53] <ghaskins_mobile> Volcane, yep, restarted it | ||
| [2010/07/21 05:30:24] <Volcane> hmm ok, sorry, again current puppet.conf for the master | ||
| [2010/07/21 05:31:07] <ghaskins_mobile> http://pastie.org/1053560 | ||
| [2010/07/21 05:31:44] <Volcane> oh | ||
| [2010/07/21 05:31:58] <Volcane> change 'lib' to plugins i think, let me try to confirm what it used to be hold | ||
| [2010/07/21 05:32:19] <ghaskins_mobile> ok, the doc implied newer versions needed lib | ||
| [2010/07/21 05:32:24] <Volcane> yes, lib should be plugins | ||
| [2010/07/21 05:32:33] <ghaskins_mobile> but now that I think about it, they might have been referring to v2.5+ ;) | ||
| [2010/07/21 05:32:34] <Volcane> ghaskins_mobile: you're on an ancient version :P | ||
| [2010/07/21 05:32:38] * ghaskins_mobile goes to try | ||
| [2010/07/21 05:32:49] <ghaskins_mobile> nod, yeah, I was just assuming "latest" like a dummy | ||
| [2010/07/21 05:33:53] <ghaskins_mobile> same result , and I restarted the server | ||
| [2010/07/21 05:33:56] <ghaskins_mobile> brb | ||
| [2010/07/21 05:34:42] @ Quit: olly__: Remote host closed the connection | ||
| [2010/07/21 05:35:03] <Volcane> ghaskins_mobile: thats pretty odd, config seems fine, try not symlinking the rb file | ||
| [2010/07/21 05:39:00] @ vinci joined channel #puppet | ||
| [2010/07/21 05:39:56] @ andrewh joined channel #puppet | ||
| [2010/07/21 05:42:11] <hal-j> didn't read all the scrollback… ghaskins_mobile / Volcane - did you check perms on both sides, to make sure the puppet processes have access to the proper files/dirs and/or they're running as root? | ||
| [2010/07/21 05:44:42] @ \ask joined channel #puppet | ||
| [2010/07/21 05:48:14] <ghaskins_mobile> back | ||
| [2010/07/21 05:49:08] @ Quit: LeRrA: Ping timeout: 248 seconds | ||
| [2010/07/21 05:49:22] <ghaskins_mobile> Volcane, hal-j, will check both of those | ||
| [2010/07/21 05:49:39] @ LeRrA joined channel #puppet | ||
| [2010/07/21 05:50:27] @ spawnyd joined channel #puppet | ||
| [2010/07/21 05:50:43] <ghaskins_mobile> its running as "puppet" but I just checked the perms and the look ok | ||
| [2010/07/21 05:50:54] <ghaskins_mobile> im going to try getting rid of the symlink now | ||
| [2010/07/21 05:51:08] <ghaskins_mobile> btw: do I need to explicitly include the "zypper" module somehow? | ||
| [2010/07/21 05:51:27] <ghaskins_mobile> i was trying to reference that before but ran into snags | ||
| [2010/07/21 05:51:44] <Volcane> no | ||
| [2010/07/21 05:52:13] @ Quit: achester: Ping timeout: 240 seconds | ||
| [2010/07/21 05:52:15] <ghaskins_mobile> Volcane, ah, ok...so I assume it just syncs all plugins discovered in modulepath then? | ||
| [2010/07/21 05:52:32] <Volcane> yup, it finds all plugins directories and merge them onto that mount | ||
| [2010/07/21 05:52:39] <ghaskins_mobile> nice | ||
| [2010/07/21 05:53:09] <hal-j> puppetmaster is running as puppet right? not the puppetd on the client side? | ||
| [2010/07/21 05:53:29] <ghaskins_mobile> hal-j, correct | ||
| [2010/07/21 05:53:36] <ghaskins_mobile> i am not sure what the client runs as, come to think of it | ||
| [2010/07/21 05:53:37] <Legooolas> hal-j: Yes. puppetd has to run as root so it can do everything | ||
| [2010/07/21 05:53:41] @ Bass10 joined channel #puppet | ||
| [2010/07/21 05:53:52] <hal-j> Legooolas: yup, that's why I was asking :) | ||
| [2010/07/21 05:54:33] <ghaskins_mobile> Legooolas, ah, of course..that makese sense given the nature of what the client needs to do | ||
| [2010/07/21 05:54:43] <ghaskins_mobile> Volcane, hal-j, unfortunately, same result | ||
| [2010/07/21 05:54:58] <ghaskins_mobile> "Cannot access mount[plugins]" | ||
| [2010/07/21 05:55:01] <Volcane> ghaskins_mobile: your master is 0.24.8 right? | ||
| [2010/07/21 05:55:08] <Volcane> ghaskins_mobile: and what does the master logs show | ||
| [2010/07/21 05:55:11] <ghaskins_mobile> yep | ||
| [2010/07/21 05:55:16] <ghaskins_mobile> standby | ||
| [2010/07/21 05:55:17] @ Quit: [Fate]: Ping timeout: 265 seconds | ||
| [2010/07/21 05:55:21] <Volcane> (syslogs probably) | ||
| [2010/07/21 05:56:02] <ghaskins_mobile> /var/log/messages looks uninteresting...anywhere else I should look? | ||
| [2010/07/21 05:56:08] <ghaskins_mobile> (this is an opensuse server) | ||
| [2010/07/21 05:56:20] <Volcane> thats probably it, dont know suse but sounds redhat like anyway | ||
| [2010/07/21 05:56:30] <ghaskins_mobile> ah, here we go | ||
| [2010/07/21 05:56:33] <ghaskins_mobile> /var/log/puppet | ||
| [2010/07/21 05:56:54] <ghaskins_mobile> the plot thickens | ||
| [2010/07/21 05:57:03] <ghaskins_mobile> puppetmasterd.log has a perms denied issue reported | ||
| [2010/07/21 05:57:29] <Volcane> whats the exact line? | ||
| [2010/07/21 05:58:02] <ghaskins_mobile> http://pastie.org/1053595 | ||
| [2010/07/21 05:58:13] @ achester joined channel #puppet | ||
| [2010/07/21 05:58:22] <Volcane> right, show your fileserver.conf again | ||
| [2010/07/21 05:58:25] <ghaskins_mobile> looks like its defaulting to "/plugins" | ||
| [2010/07/21 05:58:52] <ghaskins_mobile> http://pastie.org/1053597 | ||
| [2010/07/21 05:59:33] <ghaskins_mobile> I had "path /var/lib/puppet/plugins" in there originally | ||
| [2010/07/21 05:59:35] <Volcane> *should* be ok | ||
| [2010/07/21 05:59:42] <Volcane> try | ||
| [2010/07/21 05:59:46] <ghaskins_mobile> (a dir I created) | ||
| [2010/07/21 05:59:47] <Volcane> allow baracus-client1.laurelwood.net | ||
| [2010/07/21 05:59:52] <Volcane> and delete that dir | ||
| [2010/07/21 06:00:57] <ghaskins_mobile> much better | ||
| [2010/07/21 06:01:12] <ghaskins_mobile> i saw the zypper.rb come down | ||
| [2010/07/21 06:01:32] <Volcane> right, something weird with allow * then | ||
| [2010/07/21 06:02:33] @ Quit: achester: Ping timeout: 240 seconds | ||
| [2010/07/21 06:02:59] <ghaskins_mobile> Volcane, yeah, i just changed it to "*.laurelwood.net" and that worked too | ||
| [2010/07/21 06:03:21] <ghaskins_mobile> doesnt like the pure wild form, for some reason | ||
| [2010/07/21 06:03:28] <Legooolas> Does allow * only work for one level of name? | ||
| [2010/07/21 06:03:46] <ghaskins_mobile> Legooolas, I dont grok the question | ||
| [2010/07/21 06:04:20] @ Quit: choffee: Read error: Operation timed out | ||
| [2010/07/21 06:04:57] <ghaskins_mobile> Legooolas, what do you mean by "one level of name" | ||
| [2010/07/21 06:05:00] @ achester joined channel #puppet | ||
| [2010/07/21 06:05:20] <ghaskins_mobile> *.net ? | ||
| [2010/07/21 06:06:20] @ choffee joined channel #puppet | ||
| [2010/07/21 06:06:44] <ghaskins_mobile> you guys rock, btw, thanks for the help | ||
| [2010/07/21 06:06:50] <ghaskins_mobile> now I am on to the next problem ;) | ||
| [2010/07/21 06:07:01] <ghaskins_mobile> so now I can see zypper.rb being synced to the client | ||
| [2010/07/21 06:07:27] <ghaskins_mobile> and I have Package { provider => zypper } in my config going to the node | ||
| [2010/07/21 06:07:54] <ghaskins_mobile> but I still get "Provider zypper is not functional on this platform" for any "package" declarations | ||
| [2010/07/21 06:08:04] <ghaskins_mobile> so is there some additional magic to enable the provider? | ||
| [2010/07/21 06:08:31] <ghaskins_mobile> (sorry if newbie question) | ||
| [2010/07/21 06:08:42] <ghaskins_mobile> the doc wasnt clear in this regard, at least, not to me | ||
| [2010/07/21 06:09:45] <Volcane> ghaskins_mobile: look in the zypper.rb somewhere it does a sanity check to determine if it can be used, like say can it find all the binaries it needs | ||
| [2010/07/21 06:10:09] @ Quit: achester: Ping timeout: 240 seconds | ||
| [2010/07/21 06:11:52] <ghaskins_mobile> Volcane, not sure I am qualified yet to know what to look for, but what I see is a facter constraint (forgive if that is wrong terminology) for "confine :operatingsystem => :suse" | ||
| [2010/07/21 06:12:03] <Volcane> yup, thats part of it | ||
| [2010/07/21 06:12:06] <ghaskins_mobile> otherwise, nothing jumps out at me | ||
| [2010/07/21 06:12:15] <Volcane> pastie the zypper.rb | ||
| [2010/07/21 06:12:19] <ghaskins_mobile> ok | ||
| [2010/07/21 06:12:56] @ Quit: p3rror: Read error: Operation timed out | ||
| [2010/07/21 06:13:04] <ghaskins_mobile> http://pastie.org/1053620 | ||
| [2010/07/21 06:13:35] <ghaskins_mobile> so i assume one thing I need to do is confirm that facter is returning operatingsystem==suse | ||
| [2010/07/21 06:14:13] <ghaskins_mobile> but i am not sure how to do that (yet) | ||
| [2010/07/21 06:15:21] @ Quit: unixdaemon: Ping timeout: 240 seconds | ||
| [2010/07/21 06:15:26] @ unixdaemon joined channel #puppet | ||
| [2010/07/21 06:16:12] @ achester joined channel #puppet | ||
| [2010/07/21 06:17:36] @ Quit: tonyskapunk: Ping timeout: 276 seconds | ||
| [2010/07/21 06:17:38] @ Quit: zooz: Ping timeout: 260 seconds | ||
| [2010/07/21 06:18:04] @ zooz joined channel #puppet | ||
| [2010/07/21 06:18:52] <Volcane> does /usr/bin/zypper exist? | ||
| [2010/07/21 06:18:57] <Volcane> and rpm | ||
| [2010/07/21 06:19:13] <ghaskins_mobile> yep | ||
| [2010/07/21 06:20:05] @ Quit: QtPlatypus: Ping timeout: 252 seconds | ||
| [2010/07/21 06:20:11] @ Quit: ajf_: Ping timeout: 240 seconds | ||
| [2010/07/21 06:20:56] <ghaskins_mobile> is there a way I can dump :operatingsystem ? | ||
| [2010/07/21 06:20:56] <Volcane> should be fine then if your facter does the right thing | ||
| [2010/07/21 06:21:06] <Volcane> just type facter | ||
| [2010/07/21 06:21:27] <ghaskins_mobile> operatingsystem => OpenSuSE | ||
| [2010/07/21 06:21:49] <ghaskins_mobile> but since that is equated to :suse, perhaps that is a define that covers that spelling | ||
| [2010/07/21 06:21:55] @ QtPlatypus joined channel #puppet | ||
| [2010/07/21 06:22:02] @ radkin joined channel #puppet | ||
| [2010/07/21 06:22:08] <Volcane> nope, i suspect thats the problem | ||
| [2010/07/21 06:22:27] <ghaskins_mobile> how can I see what :suse is ? | ||
| [2010/07/21 06:22:35] <Volcane> try changing it to :opensuse | ||
| [2010/07/21 06:22:48] <ghaskins_mobile> ok | ||
| [2010/07/21 06:22:49] @ Quit: Maliuta: Ping timeout: 246 seconds | ||
| [2010/07/21 06:22:59] <ghaskins_mobile> if I update it on the server, will it get resynced? | ||
| [2010/07/21 06:23:09] <Volcane> yup | ||
| [2010/07/21 06:23:13] <ghaskins_mobile> nice | ||
| [2010/07/21 06:23:22] * ghaskins_mobile goes to update the server side | ||
| [2010/07/21 06:23:27] @ Quit: joschi: Ping timeout: 276 seconds | ||
| [2010/07/21 06:23:35] @ joschi joined channel #puppet | ||
| [2010/07/21 06:23:40] @ Quit: Omahn: Ping timeout: 240 seconds | ||
| [2010/07/21 06:25:11] <ghaskins_mobile> new file synced, now its thinking | ||
| [2010/07/21 06:25:13] <ghaskins_mobile> something is different, I suspect its going to work | ||
| [2010/07/21 06:25:28] <ghaskins_mobile> it worked! | ||
| [2010/07/21 06:25:46] <ghaskins_mobile> thanks so much, all | ||
| [2010/07/21 06:25:46] * ghaskins_mobile snapshots the server vm so that is not lost ;) | ||
| [2010/07/21 06:26:10] <Volcane> np :) | ||
| [2010/07/21 06:26:50] @ Quit: meltemi: Ping timeout: 265 seconds | ||
| [2010/07/21 06:26:52] @ ajf_ joined channel #puppet | ||
| [2010/07/21 06:26:53] @ Quit: Dominic: Ping timeout: 260 seconds | ||
| [2010/07/21 06:26:57] <ghaskins_mobile> related question: is a "provider" restricted to being a package-provider? | ||
| [2010/07/21 06:26:58] @ Quit: lanky: Ping timeout: 260 seconds | ||
| [2010/07/21 06:27:02] @ achester_ joined channel #puppet | ||
| [2010/07/21 06:27:10] <ghaskins_mobile> or can I write *.rb files to do other things, too? | ||
| [2010/07/21 06:27:10] @ Quit: Hunner: Read error: Connection reset by peer | ||
| [2010/07/21 06:27:21] <ghaskins_mobile> for instance, add zypper repos with "zypper ar" | ||
| [2010/07/21 06:27:26] @ Quit: ed209: Ping timeout: 260 seconds | ||
| [2010/07/21 06:27:26] @ Quit: yeled: Ping timeout: 260 seconds | ||
| [2010/07/21 06:27:45] @ Maliuta joined channel #puppet | ||
| [2010/07/21 06:27:46] <ghaskins_mobile> i saw someone on the lists had done this with a perl script (or something like that) | ||
| [2010/07/21 06:27:53] @ Quit: choffee: Ping timeout: 276 seconds | ||
| [2010/07/21 06:28:32] <Volcane> there are two parts | ||
| [2010/07/21 06:28:45] <ghaskins_mobile> i guess I dont care if its a "provider" more than I care if I can extend the function of puppet to allow describing new zypper repos | ||
| [2010/07/21 06:28:47] @ RussS joined channel #puppet | ||
| [2010/07/21 06:28:47] <Volcane> a type - that is what you deal with in the .pp files | ||
| [2010/07/21 06:28:53] @ ed209 joined channel #puppet | ||
| [2010/07/21 06:28:53] <Volcane> and then a provider for every OS | ||
| [2010/07/21 06:28:56] @ Omahn joined channel #puppet | ||
| [2010/07/21 06:29:01] <ghaskins_mobile> ah, got it | ||
| [2010/07/21 06:29:06] <Volcane> so if you wanted to do something entirely different, you need both | ||
| [2010/07/21 06:29:08] <ghaskins_mobile> so provider is like an os plugin | ||
| [2010/07/21 06:29:18] <Volcane> if u just want an existing type to work on a new OS you can just make a provider | ||
| [2010/07/21 06:29:38] <ghaskins_mobile> Volcane, got it, i will look at the doc with this new understanding | ||
| [2010/07/21 06:29:46] <ghaskins_mobile> type + provider | ||
| [2010/07/21 06:29:47] <ghaskins_mobile> thx! | ||
| [2010/07/21 06:29:53] <Volcane> c'ya | ||
| [2010/07/21 06:30:01] <ghaskins_mobile> later | ||
| [2010/07/21 06:30:06] <ghaskins_mobile> (im sure I will be back ;) | ||
| [2010/07/21 06:30:08] <ghaskins_mobile> thanks again | ||
| [2010/07/21 06:30:35] @ cyrus_mc joined channel #puppet | ||
| [2010/07/21 06:30:51] @ Hunner joined channel #puppet | ||
| [2010/07/21 06:31:16] @ Mick27 joined channel #puppet | ||
| [2010/07/21 06:32:33] @ Quit: dballing: Ping timeout: 276 seconds | ||
| [2010/07/21 06:32:53] @ dballing joined channel #puppet | ||
| [2010/07/21 06:35:38] @ Hollow joined channel #puppet | ||
| [2010/07/21 06:35:38] @ Quit: RussS: Quit: Leaving | ||
| [2010/07/21 06:36:16] <Hollow> hi! after upgrading the master to 2.6.0, i get the following error message: /File[/var/lib/puppet/lib]: Could not evaluate: Could not retrieve information from source(s) puppet://puppet.example.com/plugins | ||
| [2010/07/21 06:36:20] <Hollow> any ideas? | ||
| [2010/07/21 06:36:50] @ rgsteele joined channel #puppet | ||
| [2010/07/21 06:36:51] @ tyler79 joined channel #puppet | ||
| [2010/07/21 06:37:05] <Volcane> it shouldnt break anything | ||
| [2010/07/21 06:37:27] <Volcane> but you can just disable pluginsync on your clients | ||
| [2010/07/21 06:38:08] <Hollow> is pluginsync deprecated? | ||
| [2010/07/21 06:38:21] <mackn> bah.. damn youz telehouse | ||
| [2010/07/21 06:38:27] <Volcane> Hollow: no | ||
| [2010/07/21 06:39:25] @ p3rror joined channel #puppet | ||
| [2010/07/21 06:39:36] <Volcane> mackn: but you need to enable it on the server, you probably just dnot have it | ||
| [2010/07/21 06:39:41] <Hollow> then it does really make sense to disable it, how should my clients get the plugins? | ||
| [2010/07/21 06:39:58] <Hollow> i've tried adding a plugins section to fileserver.conf | ||
| [2010/07/21 06:40:00] <Volcane> do you have any plugins? | ||
| [2010/07/21 06:40:01] <Hollow> but that didn't help | ||
| [2010/07/21 06:40:15] <Hollow> yes, i have a bunch of facts and providers | ||
| [2010/07/21 06:40:37] <Volcane> right, sorry assumed like the 100s of other times we get the same question :) | ||
| [2010/07/21 06:41:00] <Volcane> erm, not sure not used 2.6 long just enough to know i ran into bugs | ||
| [2010/07/21 06:41:21] <Hollow> heh, ok, thanks anyway | ||
| [2010/07/21 06:42:39] @ Quit: ckauhaus: Quit: Leaving. | ||
| [2010/07/21 06:45:51] @ vachon joined channel #puppet | ||
| [2010/07/21 06:46:27] <vachon> im seeing some very odd behavior | ||
| [2010/07/21 06:46:30] <vachon> http://pastebin.com/RafYvyaD | ||
| [2010/07/21 06:46:45] <vachon> anyone ever seen that? | ||
| [2010/07/21 06:47:27] @ yeled joined channel #puppet | ||
| [2010/07/21 06:47:58] <vachon> basically its failing to find the class in the module some times, and it works most times... server is running rack via passenger | ||
| [2010/07/21 06:48:35] @ choffee joined channel #puppet | ||
| [2010/07/21 06:49:04] @ Quit: ffrank: Ping timeout: 246 seconds | ||
| [2010/07/21 06:49:08] @ alban21 is now known as alban2 | ||
| [2010/07/21 06:49:47] @ meltemi joined channel #puppet | ||
| [2010/07/21 06:50:23] @ Quit: \ask: Remote host closed the connection | ||
| [2010/07/21 06:50:47] @ \ask joined channel #puppet | ||
| [2010/07/21 06:52:06] @ lanky joined channel #puppet | ||
| [2010/07/21 06:53:42] @ RussS joined channel #puppet | ||
| [2010/07/21 06:53:51] @ Dominic joined channel #puppet | ||
| [2010/07/21 06:54:19] @ Quit: cyrus_mc: Ping timeout: 246 seconds | ||
| [2010/07/21 06:54:51] @ Quit: itguru: Remote host closed the connection | ||
| [2010/07/21 06:55:57] @ Quit: \ask: Ping timeout: 276 seconds | ||
| [2010/07/21 06:59:03] @ pheezy joined channel #puppet | ||
| [2010/07/21 07:02:04] <Hollow> Volcane: seems to be a passenger/apache issue. when running with webrick everything works fine | ||
| [2010/07/21 07:02:40] <vachon> Hollow: was that meant for me or Volcane | ||
| [2010/07/21 07:02:47] <Hollow> Volcane: | ||
| [2010/07/21 07:02:47] @ flakrat joined channel #puppet | ||
| [2010/07/21 07:02:49] <vachon> ok | ||
| [2010/07/21 07:02:51] <Hollow> :) | ||
| [2010/07/21 07:02:56] <vachon> i think mine is related | ||
| [2010/07/21 07:03:28] <Hollow> vachon: yes, i saw the same behaviour with 2.6.0 and passenger | ||
| [2010/07/21 07:03:33] <Hollow> no catalog is ever being applied | ||
| [2010/07/21 07:03:43] <vachon> it gets applied | ||
| [2010/07/21 07:03:48] <vachon> just not 100% of the time | ||
| [2010/07/21 07:04:00] <Hollow> i see, for me it was 100% of the time | ||
| [2010/07/21 07:04:04] <Hollow> but i didn;t try too hard ;-) | ||
| [2010/07/21 07:05:50] <vachon> yea, i think it has something to do with mod_rails caching | ||
| [2010/07/21 07:07:44] <Hollow> well, webrick works for the time being | ||
| [2010/07/21 07:07:49] @ bonyfish joined channel #puppet | ||
| [2010/07/21 07:07:59] <bonyfish> G'day guys and gals. | ||
| [2010/07/21 07:08:02] <vachon> not for me | ||
| [2010/07/21 07:08:13] <bonyfish> Any chance of asking for some help on puppet-dashboard in here? | ||
| [2010/07/21 07:08:21] <bonyfish> Ahh, spotted it in the title.. | ||
| [2010/07/21 07:08:49] @ jdcasey joined channel #puppet | ||
| [2010/07/21 07:08:57] @ vachon left channel #puppet () | ||
| [2010/07/21 07:12:01] <bonyfish> Right. Got lastest puppet, and puppet dashboard up. Added report=true in the relevant conf, and out of forty machines, only thirty appear in puppet-dashboard. Which is strange, as all changes were pushed by puppet, and I confirmed them afterwards, I.e. by making sure report=true was actually there. I stuck my head in/var/lib/puppet/reports/ and all the reports are there, all forty of them, but when i try to browse directly to them, | ||
| [2010/07/21 07:12:06] <bonyfish> Any idea what could be up? | ||
| [2010/07/21 07:12:22] <bonyfish> Any idea why reports are being generated, but not picked up by the web frontend? | ||
| [2010/07/21 07:15:03] <bonyfish> Just renamed the directory under: /var/lib/puppet/reports/f.q.d.n, new one got created with same problem | ||
| [2010/07/21 07:15:38] @ ecapriolo joined channel #puppet | ||
| [2010/07/21 07:17:14] @ reyjrar joined channel #puppet | ||
| [2010/07/21 07:18:40] @ \ask joined channel #puppet | ||
| [2010/07/21 07:18:46] @ joe-mac joined channel #puppet | ||
| [2010/07/21 07:19:05] <joe-mac> in 2.6, do you still need to restart the master to see changes to your custom functions? | ||
| [2010/07/21 07:20:41] @ rcrowley joined channel #puppet | ||
| [2010/07/21 07:24:49] <joschi> Anyone brave enough to test new Debian/Ubuntu packages for Foreman 0.1-5? -> http://theforeman.org/issues/357 | ||
| [2010/07/21 07:26:02] @ Quit: RussS: Quit: Leaving | ||
| [2010/07/21 07:27:25] @ notbrien joined channel #puppet | ||
| [2010/07/21 07:28:34] @ squirrel_labs joined channel #puppet | ||
| [2010/07/21 07:30:10] <bonyfish> Very weird, I get the same error from the web frontend when there is or isnt anything in the reports directory. | ||
| [2010/07/21 07:30:15] @ kaptk2 joined channel #puppet | ||
| [2010/07/21 07:32:07] @ Quit: Schmidt: Read error: Operation timed out | ||
| [2010/07/21 07:32:26] @ Quit: wilmoore: Remote host closed the connection | ||
| [2010/07/21 07:34:37] @ Schmidt joined channel #puppet | ||
| [2010/07/21 07:37:45] <progre55> anyone? I'm on jaunty, and want to install puppet. but jaunty repos have an old version of the package. So is it possible to add lucid as some backport into my apt sources and install it from the lucid repos? | ||
| [2010/07/21 07:38:47] @ jumex joined channel #puppet | ||
| [2010/07/21 07:39:46] <jens_> progre55, jaunty is very old, I wouldn´t except newer package versions | ||
| [2010/07/21 07:40:08] @ tonyskapunk joined channel #puppet | ||
| [2010/07/21 07:42:47] @ zooz left channel #puppet () | ||
| [2010/07/21 07:43:22] <progre55> jens_: too bad | ||
| [2010/07/21 07:47:07] <joe-mac> jaunty isn't very old... | ||
| [2010/07/21 07:47:29] <joe-mac> that's like 9.04 isn't it? | ||
| [2010/07/21 07:47:43] <joe-mac> i wouldn't say a year old is 'very old' lol | ||
| [2010/07/21 07:48:09] <joe-mac> your best bet progre55 is to grab a newer version from ftp.ubuntu, or grab the dsc and build it yourself | ||
| [2010/07/21 07:48:11] <jens_> joe-mac, there are of course different viewpoints, imho and for my needs, jaunty is old | ||
| [2010/07/21 07:48:44] @ Quit: tonyskapunk: Quit: ERC Version 5.3 (IRC client for Emacs) | ||
| [2010/07/21 07:48:53] <jumex> <n00b-question>Can someone point me in the right direction for having puppet install and configure MySQL on a Mac OS X server (I don't want to use the builtin MySQL). I was thinking of using Homebrew (http://bit.ly/cERS9L). Am I way off base?</n00b-question> | ||
| [2010/07/21 07:50:47] <barn> yeah, that's not really what puppet is for | ||
| [2010/07/21 07:50:53] <barn> either use macports or homebrew? | ||
| [2010/07/21 07:51:25] <progre55> joe-mac, jens_: thanks guys | ||
| [2010/07/21 07:51:35] <bonyfish> Weirder still.... is there any reason a node is not being added to the mysql db? | ||
| [2010/07/21 07:53:02] @ static^ joined channel #puppet | ||
| [2010/07/21 07:54:40] @ mauve joined channel #puppet | ||
| [2010/07/21 07:55:14] <jumex> barn: The reason I was thinking of it was because of the package resource (http://bit.ly/c2vWk3). My end goal is to be able to automatically provision a box with one command. I can use a bash script to execute multiple commands, but I think there is a more elegent solution out tthere (I just haven't figured it out yet). | ||
| [2010/07/21 07:55:24] <mauve> How can I refer to a resource specified in another class? | ||
| [2010/07/21 07:55:52] <mauve> Effectively want to do require => java::jre::Package['sun-java6-jre'] | ||
| [2010/07/21 07:56:39] <jbooth> mauve: Just Package['foo'] | ||
| [2010/07/21 07:58:04] <mauve> Oh, I thought I'd tried that and found it didn't work | ||
| [2010/07/21 07:58:11] <mauve> Thanks | ||
| [2010/07/21 07:59:32] <mauve> err: Could not run Puppet configuration client: Could not find dependency Package[sun-java6-jre] for Package[tomcat] at /etc/puppet/modules/tomcat/manifests/server.pp:30 | ||
| [2010/07/21 08:00:31] @ Quit: ceren: Quit: ceren | ||
| [2010/07/21 08:02:30] @ linuxfreak joined channel #puppet | ||
| [2010/07/21 08:03:00] <linuxfreak> Require some help with puppetmaster redundancy | ||
| [2010/07/21 08:03:13] <hal-j> mauve : are you sure that the node's config is somehow finding the package statement? | ||
| [2010/07/21 08:03:33] <hal-j> and it's not excluded due to a conditional somewhere, and that whatever _does_ set up that package is being included? | ||
| [2010/07/21 08:03:43] <mauve> It is included, yes | ||
| [2010/07/21 08:05:17] @ zz_shenson is now known as shenson | ||
| [2010/07/21 08:07:04] <linuxfreak> puppet redundancy | ||
| [2010/07/21 08:10:04] <dballing> Does anyone know when EPEL will have the 2.6 package(s)? | ||
| [2010/07/21 08:12:11] @ Quit: linuxfreak: Quit: Page closed | ||
| [2010/07/21 08:13:02] @ Quit: pmorillo: Quit: pmorillo | ||
| [2010/07/21 08:14:47] <progre55> jens_: oh what I actually meant was.. the puppet version for jaunty is old, but there is a new version in the lucid repos. Is it possible to install the lucid puppet on a jaunty server? =) | ||
| [2010/07/21 08:16:42] @ Quit: bug: Quit: bug | ||
| [2010/07/21 08:18:37] <tmz> dballing: Not until it's been thoroughly tested and we're confident it doesn't break existing manifests unduly. | ||
| [2010/07/21 08:18:59] <tmz> http://tmz.fedorapeople.org/repo/puppet/ | ||
| [2010/07/21 08:19:08] <joe-mac> uhhh | ||
| [2010/07/21 08:19:08] <joe-mac> Jul 21 11:11:20 thordur puppetd[27146]: Could not retrieve catalog from remote server: Could not intern from pson: regexp buffer overflow | ||
| [2010/07/21 08:19:08] <joe-mac> anyone seen that? | ||
| [2010/07/21 08:19:13] <tmz> dballing: Use that if you want to test now. | ||
| [2010/07/21 08:19:52] <mauve> Gah, I was including java::jre in the outer class, and trying to access Package[sun-java6-jre] in an inner class | ||
| [2010/07/21 08:20:47] <jbooth> joe-mac: Nope. You get unique errors don't you? ;-) | ||
| [2010/07/21 08:21:36] <joe-mac> lol, yea, openbsd | ||
| [2010/07/21 08:21:48] <joe-mac> but sadly, due to some problems we've been having i am being forced to learn cisco mostl ikely | ||
| [2010/07/21 08:21:58] @ WALoeIII joined channel #puppet | ||
| [2010/07/21 08:24:36] <dballing> touching on something I asked about yesterday: if 2.6.0 clients try to send a report to a 0.25.5 server that will not work. :-) | ||
| [2010/07/21 08:24:42] @ stewartl421 joined channel #puppet | ||
| [2010/07/21 08:25:35] <jens_> progre55, that might be working, but even if yes I wouldn´t do that on a "server", I name it repo mixery and have bad experience with such salats | ||
| [2010/07/21 08:26:10] <progre55> jens_: so should I better install a tarball? | ||
| [2010/07/21 08:27:07] <jens_> jens_, hmm maybe better yes, but also no excelent way as you guess :) | ||
| [2010/07/21 08:27:17] <jens_> s/jens_/progre55 | ||
| [2010/07/21 08:27:32] <joe-mac> dballing: yea, generally clients with higher versions won't work with newer masters | ||
| [2010/07/21 08:27:33] @ Quit: lkoranda: Quit: Ex-Chat | ||
| [2010/07/21 08:28:02] @ littleidea joined channel #puppet | ||
| [2010/07/21 08:28:57] <progre55> jens_: well then I guess I'll download it from the official site | ||
| [2010/07/21 08:30:05] <radkin> does anyone know what this is about "Could not retrieve configuration: Wrong size" ? | ||
| [2010/07/21 08:31:17] <dballing> I'm still wondering (since I don't have a good environment to test this yet)... if I've got 0.25.5 working with apache front-ending it (according to the docs on the wiki) how much of a drop-in replacement (in terms of configuration, etc.) it will be for 2.6? will it just be a matter of installing the new package and restarting apache? | ||
| [2010/07/21 08:32:41] @ Quit: benoit__: Ping timeout: 276 seconds | ||
| [2010/07/21 08:32:57] <tmz> dballing: I don't even think you'll have to restart apache. The puppet-server package should do a condrestart on the puppetmasterd processes. | ||
| [2010/07/21 08:33:26] @ ceren joined channel #puppet | ||
| [2010/07/21 08:33:44] <Volcane> tmz: but not if you're using passenger. | ||
| [2010/07/21 08:33:50] <tmz> I've not tested an install with mongrel or passenger yet. But I'm expecting it will work without change. (Famous last words, of course.) | ||
| [2010/07/21 08:34:08] <tmz> Volcane: Ahh, I didn't know passenger was that much different than mogrel. | ||
| [2010/07/21 08:34:28] @ alcy joined channel #puppet | ||
| [2010/07/21 08:34:34] <Volcane> yeah passenger takes care of spawning the master then, not the service script | ||
| [2010/07/21 08:35:30] <tmz> Good to know. I'll try to remember that. | ||
| [2010/07/21 08:37:50] <mauve> Why might puppet be incredibly slow when I run it on a VirtualBox VM, connecting to a puppetmaster on the host? | ||
| [2010/07/21 08:39:01] <joe-mac> because virtual machines are slow | ||
| [2010/07/21 08:39:24] @ Quit: kolla: Remote host closed the connection | ||
| [2010/07/21 08:39:52] <Volcane> dns :P | ||
| [2010/07/21 08:40:15] @ Quit: achester: Quit: Lost terminal | ||
| [2010/07/21 08:41:07] @ Hypnoz joined channel #puppet | ||
| [2010/07/21 08:42:03] @ benoit__ joined channel #puppet | ||
| [2010/07/21 08:43:20] <alcy> I need help in this situation: need to populate a LDAP database with an initial set of records. these records are generated at client using an erb template (things like doman name taken from facter, brokend down intobase object, DNs etc). Problem is, to actually populate the db, I need to execute a command (slapadd or ldapadd) on the template (which will be saved on the client as an ldif... | ||
| [2010/07/21 08:43:22] <alcy> ...file) which I don't how to do without using exec. any clue ? | ||
| [2010/07/21 08:43:40] @ Quit: stewartl421: Ping timeout: 245 seconds | ||
| [2010/07/21 08:43:57] @ Quit: herdingcat: Remote host closed the connection | ||
| [2010/07/21 08:44:02] <alcy> or can I use exec in this case ? :D | ||
| [2010/07/21 08:44:50] <Volcane> yeah if you can be reliably sure that it wont run multiple times and if it does that it wont cause corruption | ||
| [2010/07/21 08:46:25] <alcy> well, it shouldn't run multiple times, as it is just meant to create the initial records. so, is there a Puppet way around this problem ? | ||
| [2010/07/21 08:46:55] <Volcane> not something I'd consider 100% trustworthy :) | ||
| [2010/07/21 08:47:57] @ Quit: WALoeIII: Quit: WALoeIII | ||
| [2010/07/21 08:48:27] <Volcane> and thats fine, you wouldnt automate restoring backups directly into mysql with puppet | ||
| [2010/07/21 08:48:35] <Volcane> some stuff is just not in puppets realm | ||
| [2010/07/21 08:49:05] @ thekad-afk is now known as thekad | ||
| [2010/07/21 08:49:19] <alcy> hmmm, alright. need to make sure it runs only once then. I wish there was scuh an option for exec. or something like self destruct after first run. :D | ||
| [2010/07/21 08:49:45] <Volcane> there are ways, if say it creates a file as a result of running then puppet can use that to avoid future execs | ||
| [2010/07/21 08:49:56] <Volcane> till that file ends up in lost+found due to minor data corruption | ||
| [2010/07/21 08:50:01] <Volcane> which then has the exec run again | ||
| [2010/07/21 08:50:07] <Volcane> which then causes major data corruption | ||
| [2010/07/21 08:50:34] <hal-j> IMHO - best way to do it would be to do the load via a script that is smart enough to check for the existence of the records in the DB | ||
| [2010/07/21 08:50:44] <hal-j> IOW - change the process slightly so that it _can_ be run multiple times | ||
| [2010/07/21 08:50:51] <Volcane> yeah | ||
| [2010/07/21 08:50:54] <hal-j> and then, let it | ||
| [2010/07/21 08:51:04] <Volcane> the lock file thing just is no good | ||
| [2010/07/21 08:51:19] <hal-j> no, a lockfile that once removed makes the world explode is not a good idea ;) | ||
| [2010/07/21 08:51:31] @ Quit: squirrel_labs: Quit: See ya! | ||
| [2010/07/21 08:52:53] <alcy> i was thinking something like exec {"slapadd" : onlyif=>"test `magic code to see if there are 0 records in the db`) | ||
| [2010/07/21 08:54:46] <alcy> looks ok ? | ||
| [2010/07/21 08:56:38] <alcy> gotta write the magic code there. tomorrow. | ||
| [2010/07/21 08:57:07] @ Quit: Hypnoz: Ping timeout: 276 seconds | ||
| [2010/07/21 09:00:20] <joe-mac> does slapadd create any files? | ||
| [2010/07/21 09:00:30] <joe-mac> creates => is often easier to use than onlyif imo | ||
| [2010/07/21 09:00:47] <Volcane> joe-mac: see comment about lost+found | ||
| [2010/07/21 09:03:36] @ gniks joined channel #puppet | ||
| [2010/07/21 09:03:53] @ Quit: gniks: Client Quit | ||
| [2010/07/21 09:04:46] <joe-mac> o, i've never run into that | ||
| [2010/07/21 09:05:05] <joe-mac> very rarely do i run into fs corruption | ||
| [2010/07/21 09:05:08] @ Quit: jdcasey: Remote host closed the connection | ||
| [2010/07/21 09:05:16] <joe-mac> it's more likely i will experience a raid fubar due to double drive faults | ||
| [2010/07/21 09:06:10] <Volcane> :) still i wouldnt be doing much thats potentially data destructive with puppet | ||
| [2010/07/21 09:06:26] <Volcane> how easy is it to mess up a regex in vi or a simple finger problem that might change your creates without you noticing | ||
| [2010/07/21 09:06:30] <Volcane> way too fragile | ||
| [2010/07/21 09:07:31] @ ckauhaus joined channel #puppet | ||
| [2010/07/21 09:08:35] <ohadlevy> joe-mac: i just got 200GB in lost+found | ||
| [2010/07/21 09:08:43] <alcy> joe-mac: slapadd just adds entries into a database ( .bdb ). Volcane, maybe I could keep those few shell scripts ? :) | ||
| [2010/07/21 09:09:12] <joe-mac> ohadlevy: ouch, how? what FS? | ||
| [2010/07/21 09:09:39] <ohadlevy> joe-mac: donno how, ext3, maybe it helps its an external usb drive :) | ||
| [2010/07/21 09:09:49] @ Quit: progre55: Remote host closed the connection | ||
| [2010/07/21 09:10:00] <hal-j> I think ultimately it doesn't much matter. FS corruption can happen, as can silly mistakes like removing a lockfile by mistake. I wouldn't want either of those two to cause further damage by corrupting my ldap DB as well | ||
| [2010/07/21 09:10:40] <joe-mac> yea, however it can still happen with an onlyif | ||
| [2010/07/21 09:10:41] <hal-j> especially if it's easy enough to wrap the initial insert in such a way that it assures me that the insert won't happen twice (like, check each entry's existence before inserting it) | ||
| [2010/07/21 09:10:50] <joe-mac> it's a risk you're taking for the sake of automation | ||
| [2010/07/21 09:11:05] <hal-j> that's why I suggested changing the insert process, rather than using onlyif | ||
| [2010/07/21 09:12:02] <hal-j> wouldn't be difficult to wrap something that checks for each entry before slapadding it | ||
| [2010/07/21 09:12:42] <alcy> hal-j: thanks, will try to work through that approach. yup, wouldn't want to risk the db. :) | ||
| [2010/07/21 09:14:57] @ stewartl42 joined channel #puppet | ||
| [2010/07/21 09:15:07] @ Quit: ahasenack: Quit: Leaving | ||
| [2010/07/21 09:16:00] @ WALoeIII joined channel #puppet | ||
| [2010/07/21 09:16:07] @ Quit: WALoeIII: Remote host closed the connection | ||
| [2010/07/21 09:16:13] @ WALoeIII joined channel #puppet | ||
| [2010/07/21 09:16:45] <bitfield> if there's something that you really can't afford to have happen, you need to make sure it can't happen, and also that even if it does happen, it can't happen. | ||
| [2010/07/21 09:17:02] <bitfield> bit too wordy for an aphorism, but i'll work on it | ||
| [2010/07/21 09:17:39] <alcy> reminds me of - if something can go wrong, it will go wrong. meh. | ||
| [2010/07/21 09:19:56] <hal-j> bitfield: pretty much what I'm getting at, I just didn't say it so…. elegantly :) | ||
| [2010/07/21 09:20:29] <hal-j> start with the assumption that puppet _will_ run this loader twice. with that assumption in mind, lets make sure it's not destructive | ||
| [2010/07/21 09:20:35] @ warreng joined channel #puppet | ||
| [2010/07/21 09:20:47] <hal-j> then you won't be caught off guard when it does happen :) | ||
| [2010/07/21 09:22:33] @ conconcatenate joined channel #puppet | ||
| [2010/07/21 09:23:11] @ bug joined channel #puppet | ||
| [2010/07/21 09:25:03] <alcy> didn't think about the consequences to this extent. thanks folks, will try to tke care of these. | ||
| [2010/07/21 09:25:52] <alcy> 'night everyone. | ||
| [2010/07/21 09:26:00] @ alcy left channel #puppet () | ||
| [2010/07/21 09:26:56] @ gniks joined channel #puppet | ||
| [2010/07/21 09:28:19] @ Quit: spawnyd: Ping timeout: 276 seconds | ||
| [2010/07/21 09:28:33] @ jdcasey joined channel #puppet | ||
| [2010/07/21 09:33:54] @ Quit: rmiller4pi8: Ping timeout: 276 seconds | ||
| [2010/07/21 09:34:11] @ Quit: mauve: Quit: Leaving | ||
| [2010/07/21 09:35:42] @ Quit: bonyfish: Quit: Page closed | ||
| [2010/07/21 09:38:06] @ alfism joined channel #puppet | ||
| [2010/07/21 09:38:25] @ jmccune joined channel #puppet | ||
| [2010/07/21 09:38:48] @ Quit: TREllis: Ping timeout: 260 seconds | ||
| [2010/07/21 09:41:56] @ mclarke joined channel #puppet | ||
| [2010/07/21 09:45:31] @ Quit: ceren: Ping timeout: 240 seconds | ||
| [2010/07/21 09:45:50] <cparedes> anyone here using puppet to manage SGE clusters? | ||
| [2010/07/21 09:46:02] @ TREllis joined channel #puppet | ||
| [2010/07/21 09:46:32] @ Quit: gebi: Read error: Operation timed out | ||
| [2010/07/21 09:48:20] @ adrian_broher joined channel #puppet | ||
| [2010/07/21 09:48:50] @ plathrop-away is now known as plathrop | ||
| [2010/07/21 09:50:19] @ Quit: MattyM: Remote host closed the connection | ||
| [2010/07/21 09:54:03] @ Quit: littleidea: Ping timeout: 276 seconds | ||
| [2010/07/21 09:57:25] @ littleidea joined channel #puppet | ||
| [2010/07/21 09:58:39] <pheezy> wow this dashboard is so wicked unhelpful if something goes wrong | ||
| [2010/07/21 10:01:16] @ Quit: ckauhaus: Read error: Operation timed out | ||
| [2010/07/21 10:03:16] @ rmiller4pi8 joined channel #puppet | ||
| [2010/07/21 10:06:20] @ Quit: jaredrhine: Quit: Leaving. | ||
| [2010/07/21 10:07:14] @ benkevan joined channel #puppet | ||
| [2010/07/21 10:07:40] <benkevan> can I do a conditional statement for if a file exists on a certain os type? | ||
| [2010/07/21 10:07:48] @ Quit: littleidea: Quit: littleidea | ||
| [2010/07/21 10:08:38] <AK6L> case $operatingsystem { "someOS": { file { "..." } } } | ||
| [2010/07/21 10:08:48] <AK6L> there's also a ternary operator, but i can't remember the syntax offhand. | ||
| [2010/07/21 10:10:34] @ lak joined channel #puppet | ||
| [2010/07/21 10:11:02] <benkevan> thanks.. now trying to decide if I should put this in the resolver module, or if I should create a module for SLES and put it in there.. hummmm | ||
| [2010/07/21 10:12:01] <benkevan> would: case $operatingsystem { | ||
| [2010/07/21 10:12:01] <benkevan> SLES|suse: | ||
| [2010/07/21 10:12:03] <benkevan> work? | ||
| [2010/07/21 10:12:09] <benkevan> or does it have to be SLES then suse ? | ||
| [2010/07/21 10:13:54] <AK6L> not sure to be honest. when i have questions like that i write up a quick dummy manifest and test it | ||
| [2010/07/21 10:14:05] <benkevan> I'm the big dummy :D jk | ||
| [2010/07/21 10:14:20] <benkevan> I'm still pretty new to puppet.. so I'm just all over the place haha | ||
| [2010/07/21 10:14:37] @ tonyskapunk joined channel #puppet | ||
| [2010/07/21 10:22:37] @ littleidea joined channel #puppet | ||
| [2010/07/21 10:24:14] @ Quit: lak: Quit: lak | ||
| [2010/07/21 10:24:28] @ Quit: yannL: Remote host closed the connection | ||
| [2010/07/21 10:24:53] @ Quit: jdcasey: Remote host closed the connection | ||
| [2010/07/21 10:24:58] @ Quit: TREllis: Ping timeout: 260 seconds | ||
| [2010/07/21 10:33:17] @ Quit: rmiller4pi8: Ping timeout: 258 seconds | ||
| [2010/07/21 10:35:28] @ Quit: alban2: Quit: Leaving. | ||
| [2010/07/21 10:35:47] <yakub> i have a test class i inflict on servers for.... testing | ||
| [2010/07/21 10:38:14] <benkevan> hummmm.. my puppet doesn't seem to be taking one of my modules.. --debug doesn't really show that it's even trying to get it | ||
| [2010/07/21 10:38:37] @ Quit: littleidea: Quit: littleidea | ||
| [2010/07/21 10:38:38] <benkevan> oh.. wait | ||
| [2010/07/21 10:38:41] <benkevan> i'm just an idiot.. | ||
| [2010/07/21 10:39:08] <benkevan> ok.. here's a weird question.. where should I define the variables.. that I use within my modules? | ||
| [2010/07/21 10:42:08] @ Quit: warreng: Quit: Leaving. | ||
| [2010/07/21 10:44:14] @ Hypnoz joined channel #puppet | ||
| [2010/07/21 10:45:09] <benkevan> err: Could not run Puppet configuration client: SIGQUIT | ||
| [2010/07/21 10:46:16] <benkevan> yay.. think this is working | ||
| [2010/07/21 10:49:22] <duritong> masterzen: you have nginx in front of mongrel, right? do you have any special buffer sizes defined? because I keep getting that the requests are too big on certain resources | ||
| [2010/07/21 10:50:02] @ rmiller4pi8 joined channel #puppet | ||
| [2010/07/21 10:55:12] @ warreng joined channel #puppet | ||
| [2010/07/21 10:56:11] <benkevan> man.. after setting all this up, i'm going to have to think of environments.. so I have a place to test.. | ||
| [2010/07/21 10:58:33] @ nexx joined channel #puppet | ||
| [2010/07/21 10:59:44] <masterzen> duritong: I have proxy_buffer_size 16k and proxy_buffers 8 32k | ||
| [2010/07/21 11:02:13] @ kolla joined channel #puppet | ||
| [2010/07/21 11:05:14] @ bitfield_ joined channel #puppet | ||
| [2010/07/21 11:05:52] <duritong> masterzen: ok, thanks | ||
| [2010/07/21 11:07:13] @ Quit: kolla: Remote host closed the connection | ||
| [2010/07/21 11:07:38] @ kolla joined channel #puppet | ||
| [2010/07/21 11:09:31] @ Quit: joe-mac: Read error: Connection reset by peer | ||
| [2010/07/21 11:10:04] @ Quit: bitfield_: Ping timeout: 265 seconds | ||
| [2010/07/21 11:14:49] @ bitfield_ joined channel #puppet | ||
| [2010/07/21 11:16:09] @ Quit: bitfield_: Remote host closed the connection | ||
| [2010/07/21 11:20:56] @ lak joined channel #puppet | ||
| [2010/07/21 11:22:05] @ Quit: rmiller4pi8: Quit: Leaving. | ||
| [2010/07/21 11:22:39] @ jaredrhine joined channel #puppet | ||
| [2010/07/21 11:22:58] @ rmiller4pi8 joined channel #puppet | ||
| [2010/07/21 11:23:50] @ mclarke_ joined channel #puppet | ||
| [2010/07/21 11:24:01] @ Quit: mclarke: Read error: Connection reset by peer | ||
| [2010/07/21 11:24:01] @ mclarke_ is now known as mclarke | ||
| [2010/07/21 11:24:15] @ Quit: kolla: Ping timeout: 252 seconds | ||
| [2010/07/21 11:24:34] @ Quit: adrian_broher: Quit: Verlassend | ||
| [2010/07/21 11:27:56] @ Quit: bgupta: Quit: bgupta | ||
| [2010/07/21 11:30:29] @ bgupta joined channel #puppet | ||
| [2010/07/21 11:30:45] <geekatcmu> You know what I hate? Software that ships with three config files in /etc (for no discernable reason), yet still has config.pm files scattered around, and hard-coded config stuff in other random files. | ||
| [2010/07/21 11:31:05] <geekatcmu> Especially when one of the shipped libraries includes a parser for pulling all the information out of the /etc files. | ||
| [2010/07/21 11:31:37] <ReinH> pheezy: can you possibly define "wicked unhelpful" or suggest ways we can make it more helful? | ||
| [2010/07/21 11:31:42] <geekatcmu> Admittedly, it only took me 20 minutes to kill off one of the major config.pm files so that it just extracts all the relevant data from the /etc file. Still. | ||
| [2010/07/21 11:31:54] <geekatcmu> Hrm. Wrong window. | ||
| [2010/07/21 11:32:01] <Tonnerre> What's even better is software which ships with config files it never reads or uses | ||
| [2010/07/21 11:32:58] @ _lucid joined channel #puppet | ||
| [2010/07/21 11:35:27] <_lucid> i need to be able to support both monitrc and monit.conf ... has anyone invented that wheel yet? | ||
| [2010/07/21 11:45:05] @ Quit: benoit__: Ping timeout: 276 seconds | ||
| [2010/07/21 11:46:15] @ adrian_broher joined channel #puppet | ||
| [2010/07/21 11:46:56] @ jdcasey joined channel #puppet | ||
| [2010/07/21 11:51:20] @ littleidea joined channel #puppet | ||
| [2010/07/21 11:54:04] @ HouseAway is now known as AimanA | ||
| [2010/07/21 12:02:16] <pheezy> I'll need a minute ReinH, someone changing the libdir per the README.markdown has hosed all my certificates :-/ | ||
| [2010/07/21 12:02:35] <ReinH> pheezy: o_O | ||
| [2010/07/21 12:02:48] <pheezy> I don't think it liked the colon in there | ||
| [2010/07/21 12:03:04] <ReinH> pheezy: yeah, that's unfortunately a Puppet bug | ||
| [2010/07/21 12:03:25] <ReinH> pheezy: I'd recommend using the current reductivelabs git master, we're about to release it as a new point version that addresses a lot of those issues | ||
| [2010/07/21 12:03:30] <ReinH> and also provides far better documentation | ||
| [2010/07/21 12:03:40] <pheezy> hmm ok | ||
| [2010/07/21 12:03:46] <ReinH> for installing on both Puppet 0.25.x and 2.6.0 | ||
| [2010/07/21 12:04:22] <ReinH> pheezy: installing on 0.25.x is still a bit tedious but thankfully less so | ||
| [2010/07/21 12:04:31] <ReinH> and also has the advantage of not hosing your certificates :/ | ||
| [2010/07/21 12:06:48] @ Pyretic joined channel #puppet | ||
| [2010/07/21 12:06:52] <pheezy> Yeah wow, so somehow there was now a directory in /etc/puppet/ssl with all sorts of puppet ssl stuff | ||
| [2010/07/21 12:06:58] <pheezy> on the clients o_O | ||
| [2010/07/21 12:07:58] <Pyretic> that's usually in /var | ||
| [2010/07/21 12:08:55] <benkevan> I have the following 2 errors.. http://www.fpaste.org/xprC/ .. | ||
| [2010/07/21 12:09:04] <benkevan> for the second one.. do I need to define zypper an rpm source? | ||
| [2010/07/21 12:09:33] <ReinH> pheezy: that, er, doesn't sound like something dashboard would do... | ||
| [2010/07/21 12:10:16] <ReinH> dashboard doesn't write anything to your system (outside of its project root and its database) or make any changes to clients at all (aside from the changes puppet makes via using it as an external nodes tool) | ||
| [2010/07/21 12:10:18] <jbooth> benkevan: Your first error -- what are your client and server versions? Your second, why are you using the rpm package provider? | ||
| [2010/07/21 12:11:12] @ Quit: andrewh: Ping timeout: 276 seconds | ||
| [2010/07/21 12:11:55] <benkevan> jbooth: 25.5 and 25.5 server / client, and I dunno.. just what I've setup I guess (not even sure how I setup rpm as the provider) | ||
| [2010/07/21 12:12:52] <jbooth> benkevan: What base OS? | ||
| [2010/07/21 12:12:57] <benkevan> jbooth: SLES 11 | ||
| [2010/07/21 12:13:03] <benkevan> jbooth: SLES 11 SP1 | ||
| [2010/07/21 12:13:25] @ Quit: \ask: Remote host closed the connection | ||
| [2010/07/21 12:13:42] <jbooth> benkevan: In that case I have no idea. Smells like a sles problem. I'd check rpm is the 'best' provider for it to be using, cause I'm kinda doubting it is. Maybe you're missing a package or something. | ||
| [2010/07/21 12:13:49] @ \ask joined channel #puppet | ||
| [2010/07/21 12:14:05] <benkevan> jbooth: how can I change the provider? (I guess I don't know how to do that).. I imagine zypper would be the best provider | ||
| [2010/07/21 12:14:55] <pheezy> ReinH: yeah I'm aware of that, not blaming dashboard, think i typoed on $libdir and now POW o_O | ||
| [2010/07/21 12:18:22] @ Quit: \ask: Ping timeout: 260 seconds | ||
| [2010/07/21 12:19:36] @ kolla joined channel #puppet | ||
| [2010/07/21 12:20:26] <pheezy> thank $deity for multiple environments o_O | ||
| [2010/07/21 12:21:25] <benkevan> i don't have any "provider" in /var/lib/puppet | ||
| [2010/07/21 12:21:27] <benkevan> should I ? | ||
| [2010/07/21 12:21:47] <nico> directly ? no | ||
| [2010/07/21 12:21:59] <benkevan> then how can I see what providers are available? | ||
| [2010/07/21 12:22:30] <nico> package providers ? | ||
| [2010/07/21 12:22:37] <benkevan> nico: agreed | ||
| [2010/07/21 12:22:51] <nico> in /usr/lib/ruby/1.8/puppet/provider/package | ||
| [2010/07/21 12:23:00] <nico> (on my debian, ymmv) | ||
| [2010/07/21 12:23:06] <Volcane> benkevan: the docs also have a handy list complete with capabilities and features | ||
| [2010/07/21 12:23:15] <pheezy> puppetdoc -r providers should help | ||
| [2010/07/21 12:23:20] @ Quit: warreng: Quit: Leaving. | ||
| [2010/07/21 12:23:34] <Naresh> congrats on the new release! | ||
| [2010/07/21 12:23:37] <nico> pheezy: would it show plugin distributed ones ? | ||
| [2010/07/21 12:23:56] <nico> there is a (pending iirc) patch provider for zypper | ||
| [2010/07/21 12:24:03] <benkevan> weird.. I don't have puppet doc.. oh well.. | ||
| [2010/07/21 12:24:04] <pheezy> hmm that I don't know, you might have to do puppetdoc --pluginsync -r providers | ||
| [2010/07/21 12:24:08] <benkevan> nico: oh.. it hasn't been in yet? | ||
| [2010/07/21 12:24:17] <pheezy> but i'm not sure if puppetdoc supports that | ||
| [2010/07/21 12:24:24] <benkevan> whereis puppetdoc | ||
| [2010/07/21 12:24:24] <pheezy> i'm on 0.25.5 too btw | ||
| [2010/07/21 12:24:24] <benkevan> puppetdoc: | ||
| [2010/07/21 12:24:42] <benkevan> pheezy: on what underline os? | ||
| [2010/07/21 12:24:45] @ Quit: toi: Remote host closed the connection | ||
| [2010/07/21 12:24:46] <benkevan> what packages did you use to install? | ||
| [2010/07/21 12:25:19] <pheezy> centos/redhat, just instaleld puppet and puppet-server where necessary | ||
| [2010/07/21 12:25:32] <benkevan> pheezy: from epel? | ||
| [2010/07/21 12:25:50] <pheezy> i don't think puppetdoc agrees with the rubygems-rdoc package provided by epel though, so you can't generate the rdoc html files :( | ||
| [2010/07/21 12:25:51] <pheezy> yeah | ||
| [2010/07/21 12:26:10] <benkevan> hummm.. i created my own packages.. It didn't create puppetdoc maybe I'll have to take a peak at that | ||
| [2010/07/21 12:26:11] <pheezy> puppetdoc: /usr/bin/puppetdoc /usr/share/man/man8/puppetdoc.8.gz | ||
| [2010/07/21 12:27:22] <benkevan> hey.. I just saw 2.6 | ||
| [2010/07/21 12:27:29] <benkevan> as latest.. when did that get released? | ||
| [2010/07/21 12:28:02] <fenris02> yesterday? | ||
| [2010/07/21 12:28:11] <benkevan> nice | ||
| [2010/07/21 12:28:17] @ kolla_ joined channel #puppet | ||
| [2010/07/21 12:28:30] <fenris02> yeah, the link in the topic has the announcement | ||
| [2010/07/21 12:29:03] <benkevan> anyone have it on their machine, that can do a puppetdoc to let me know if zypper is included as a package provider? | ||
| [2010/07/21 12:29:06] @ Quit: kolla: Ping timeout: 260 seconds | ||
| [2010/07/21 12:30:15] @ Quit: MPSimmons: Quit: Leaving. | ||
| [2010/07/21 12:31:46] <fenris02> loads of new features. looks nifty. | ||
| [2010/07/21 12:31:49] <tmz> benkevan: If you want to poke 2.6.0 packages for fedora/epel, take a look at http://tmz.fedorapeople.org/repo/puppet/ (grab the puppet.repo file for the appropriate dist and yum install) | ||
| [2010/07/21 12:32:04] <tmz> Those should match what ends up in epel pretty closely. | ||
| [2010/07/21 12:32:36] * tmz apologizes for whoring for testing and feedback. | ||
| [2010/07/21 12:32:41] * fenris02 ^5s tmz | ||
| [2010/07/21 12:33:18] * benkevan is on SLES 11 :( | ||
| [2010/07/21 12:36:01] <nico> benkevan: zypper support is in 2.6.0 | ||
| [2010/07/21 12:36:42] <benkevan> nico: YAY.. going to build 2.6 .. and look to see why I didn't get the puppetdoc | ||
| [2010/07/21 12:39:56] <benkevan> any one know the facter requirement for 2.6.0 ? | ||
| [2010/07/21 12:40:07] <nico> use the latest | ||
| [2010/07/21 12:43:24] @ toi joined channel #puppet | ||
| [2010/07/21 12:47:30] <benkevan> god.. the suse package was HORRIBLE | ||
| [2010/07/21 12:47:35] <benkevan> ralsh / pi etc in /usr/sbin | ||
| [2010/07/21 12:47:37] <benkevan> and not in /usr/bin | ||
| [2010/07/21 12:47:46] * benkevan stabs novell | ||
| [2010/07/21 12:52:17] @ Deesl joined channel #puppet | ||
| [2010/07/21 12:53:34] @ Guest1718 joined channel #puppet | ||
| [2010/07/21 12:57:23] @ rellis joined channel #puppet | ||
| [2010/07/21 12:59:18] <ghaskins_mobile> ok, i am close now | ||
| [2010/07/21 12:59:40] <fenris02> benkevan, i'm sure rhel folks would warmly welcome you should you decide to switch 8-) | ||
| [2010/07/21 12:59:46] <ghaskins_mobile> the last problem i seem to be having is that my puppetd never comes back after it initially connects and doesnt have a signed cert | ||
| [2010/07/21 13:00:04] <ghaskins_mobile> if I sign the cert and wait, the client never seems to ever come back | ||
| [2010/07/21 13:00:12] @ Quit: adrian_broher: *.net *.split | ||
| [2010/07/21 13:00:13] @ Quit: pctony: *.net *.split | ||
| [2010/07/21 13:00:13] @ Quit: carlasouza: *.net *.split | ||
| [2010/07/21 13:00:13] @ Quit: huggie: *.net *.split | ||
| [2010/07/21 13:00:14] <ghaskins_mobile> but if I manually run puppetd --test, it works | ||
| [2010/07/21 13:00:38] <ghaskins_mobile> I believe I am just using the standard daemonized puppetd, not cron, etc | ||
| [2010/07/21 13:00:50] <ghaskins_mobile> any ideas on ways to debug this? | ||
| [2010/07/21 13:01:19] <ghaskins_mobile> my expectation is that it would come back within the default waitforcert=120 interval | ||
| [2010/07/21 13:01:25] <ghaskins_mobile> could it be waiting for 30m? | ||
| [2010/07/21 13:01:47] @ TREllis joined channel #puppet | ||
| [2010/07/21 13:01:59] @ Quit: bug: Quit: bug | ||
| [2010/07/21 13:02:43] @ ckauhaus joined channel #puppet | ||
| [2010/07/21 13:03:15] @ huggie joined channel #puppet | ||
| [2010/07/21 13:05:09] @ Quit: Guest1718: Quit: Leaving... | ||
| [2010/07/21 13:07:15] @ radkin1 joined channel #puppet | ||
| [2010/07/21 13:07:56] <benkevan> jamesturnbull: I found some more hoopla's with SUSE's packaging .spec.. I didn't realize they didn't include puppetdoc (they explicitly exclude it for some reason).. | ||
| [2010/07/21 13:08:55] @ Quit: radkin: Ping timeout: 258 seconds | ||
| [2010/07/21 13:09:47] @ OpenMedia joined channel #puppet | ||
| [2010/07/21 13:10:59] @ pctony joined channel #puppet | ||
| [2010/07/21 13:11:45] @ carlasouza joined channel #puppet | ||
| [2010/07/21 13:12:07] @ adrian_broher joined channel #puppet | ||
| [2010/07/21 13:12:46] @ Mikey joined channel #puppet | ||
| [2010/07/21 13:13:14] @ Mikey is now known as Guest94198 | ||
| [2010/07/21 13:13:56] @ Quit: Guest94198: Excess Flood | ||
| [2010/07/21 13:14:41] @ bug joined channel #puppet | ||
| [2010/07/21 13:15:27] @ Mikey_ joined channel #puppet | ||
| [2010/07/21 13:15:57] @ Quit: Mikey_: Excess Flood | ||
| [2010/07/21 13:17:13] @ Mikey_ joined channel #puppet | ||
| [2010/07/21 13:17:24] @ Quit: AK6L: Remote host closed the connection | ||
| [2010/07/21 13:18:40] @ Quit: pheezy: Remote host closed the connection | ||
| [2010/07/21 13:19:09] @ kc7zzv joined channel #puppet | ||
| [2010/07/21 13:19:28] @ AK6L joined channel #puppet | ||
| [2010/07/21 13:20:08] @ pheezy joined channel #puppet | ||
| [2010/07/21 13:20:38] @ Quit: Mikey_: Remote host closed the connection | ||
| [2010/07/21 13:21:19] @ Mikey_ joined channel #puppet | ||
| [2010/07/21 13:21:21] @ Quit: Mikey_: Max SendQ exceeded | ||
| [2010/07/21 13:21:27] <kc7zzv> I think I might make a ruby module to manage cups soon. Is there any way I could make it delete non-managed printers? I know how to delete printers. Is there a good way that a type can get a list of all other resources of that type? | ||
| [2010/07/21 13:21:44] @ Mikey_ joined channel #puppet | ||
| [2010/07/21 13:21:46] @ Quit: Mikey_: Excess Flood | ||
| [2010/07/21 13:22:07] @ Quit: pheezy: Remote host closed the connection | ||
| [2010/07/21 13:22:22] @ olafrv joined channel #puppet | ||
| [2010/07/21 13:22:25] <olafrv> Hi | ||
| [2010/07/21 13:22:55] <olafrv> Anybody can give an example of using "refresh" in exec type | ||
| [2010/07/21 13:23:03] @ Mikey_ joined channel #puppet | ||
| [2010/07/21 13:23:04] @ Quit: Mikey_: Excess Flood | ||
| [2010/07/21 13:23:08] @ Quit: littleidea: Quit: littleidea | ||
| [2010/07/21 13:23:28] @ pheezy joined channel #puppet | ||
| [2010/07/21 13:23:33] @ Mikey_ joined channel #puppet | ||
| [2010/07/21 13:23:34] @ Quit: Mikey_: Excess Flood | ||
| [2010/07/21 13:24:34] <kc7zzv> You mean refresh_only? (I think that's spelled right) | ||
| [2010/07/21 13:24:57] @ Mikey_ joined channel #puppet | ||
| [2010/07/21 13:25:00] @ Quit: Mikey_: Excess Flood | ||
| [2010/07/21 13:25:50] @ Mikey_ joined channel #puppet | ||
| [2010/07/21 13:25:53] @ Quit: Mikey_: Excess Flood | ||
| [2010/07/21 13:26:31] @ Quit: bug: Quit: bug | ||
| [2010/07/21 13:26:41] <olafrv> No just refresh as you can see on Exec type reference | ||
| [2010/07/21 13:27:05] @ Quit: Deesl: Remote host closed the connection | ||
| [2010/07/21 13:27:14] <olafrv> http://docs.puppetlabs.com/references/stable/type.html#refresh | ||
| [2010/07/21 13:27:24] <kc7zzv> olafrv: I think you just put another command on that parameter. | ||
| [2010/07/21 13:27:44] <olafrv> And what is that command for? | ||
| [2010/07/21 13:28:02] <mackn> if it has a special form explicitly for refreshing | ||
| [2010/07/21 13:28:11] <mackn> like a restart | ||
| [2010/07/21 13:28:17] <mackn> or a stop/start | ||
| [2010/07/21 13:28:19] <mackn> *shrug* | ||
| [2010/07/21 13:28:28] @ Quit: toi: Quit: Ex-Chat | ||
| [2010/07/21 13:28:43] <kc7zzv> Or if you want to use "reload" instead of start/restart. | ||
| [2010/07/21 13:28:56] <mackn> something other than the 'default' action for a refresh | ||
| [2010/07/21 13:29:31] <mackn> depending on which provider it uses there is a default (like /etc/init.d/blah restart) | ||
| [2010/07/21 13:29:36] @ Quit: juniper: Quit: She who quits and runs away... | ||
| [2010/07/21 13:30:00] @ Quit: _lucid: Ping timeout: 252 seconds | ||
| [2010/07/21 13:30:01] <mackn> but if it was something wonky like /etc/init.d/blah redoit then you would put that in for the 'refresh' parameter | ||
| [2010/07/21 13:31:58] <kc7zzv> Pure speculation: Might have been more useful before the service resource could do so much. | ||
| [2010/07/21 13:32:49] <olafrv> ok | ||
| [2010/07/21 13:33:32] @ littleidea joined channel #puppet | ||
| [2010/07/21 13:35:02] @ be200fx joined channel #puppet | ||
| [2010/07/21 13:36:51] @ Quit: jumex: Quit: Page closed | ||
| [2010/07/21 13:37:41] @ bleything joined channel #puppet | ||
| [2010/07/21 13:38:35] <bleything> so I'm following the config instructions, and I got to this part: "puppetd --server myserver.domain.com --waitforcert 60 --test" | ||
| [2010/07/21 13:38:45] <bleything> and when I run that, I get "err: Could not request certificate: undefined method `closed?' for nil:NilClass" | ||
| [2010/07/21 13:39:00] <bleything> which I googled, and turned up a ticket that suggested setting the fqdn in the server config... which I did, and didn't seem to help | ||
| [2010/07/21 13:40:51] @ Quit: littleidea: Quit: littleidea | ||
| [2010/07/21 13:42:06] <bleything> oh jeez. apparently I had to install the puppetmaster package too. sorry, I'm not good at ubuntu. | ||
| [2010/07/21 13:42:59] @ Quit: nexx: Quit: quit | ||
| [2010/07/21 13:44:08] @ littleidea joined channel #puppet | ||
| [2010/07/21 13:47:01] <bleything> no, actually, new problem. https://gist.github.com/a6a016f81a9e4ca8fb29 | ||
| [2010/07/21 13:47:35] @ Mikey_ joined channel #puppet | ||
| [2010/07/21 13:47:36] @ Quit: Mikey_: Excess Flood | ||
| [2010/07/21 13:49:17] @ yannL joined channel #puppet | ||
| [2010/07/21 13:56:06] @ Quit: littleidea: Quit: littleidea | ||
| [2010/07/21 14:00:21] @ Quit: kc7zzv: Quit: kc7zzv | ||
| [2010/07/21 14:01:26] @ Quit: rgsteele: Remote host closed the connection | ||
| [2010/07/21 14:03:50] @ Quit: static^: Remote host closed the connection | ||
| [2010/07/21 14:06:28] @ Quit: kaptk2: Read error: Operation timed out | ||
| [2010/07/21 14:08:49] @ kaptk2 joined channel #puppet | ||
| [2010/07/21 14:09:40] @ littleidea joined channel #puppet | ||
| [2010/07/21 14:13:54] @ Quit: ckauhaus: Quit: Leaving. | ||
| [2010/07/21 14:14:23] @ radkin joined channel #puppet | ||
| [2010/07/21 14:16:23] @ Quit: radkin1: Ping timeout: 258 seconds | ||
| [2010/07/21 14:18:04] @ Quit: Mick27: Quit: This computer has gone to sleep | ||
| [2010/07/21 14:20:15] @ kc7zzv joined channel #puppet | ||
| [2010/07/21 14:22:47] @ Mick27 joined channel #puppet | ||
| [2010/07/21 14:24:18] @ Quit: kc7zzv: Ping timeout: 240 seconds | ||
| [2010/07/21 14:24:49] @ Quit: Mick27: Client Quit | ||
| [2010/07/21 14:26:25] @ Quit: adrian_broher: Quit: Verlassend | ||
| [2010/07/21 14:27:01] @ bug joined channel #puppet | ||
| [2010/07/21 14:29:15] @ plathrop is now known as plathrop-away | ||
| [2010/07/21 14:29:50] @ Quit: reyjrar: Quit: Leaving. | ||
| [2010/07/21 14:31:02] @ alban2 joined channel #puppet | ||
| [2010/07/21 14:32:21] @ Quit: kolla_: Ping timeout: 252 seconds | ||
| [2010/07/21 14:34:31] <pheezy> ReinH: you still around? | ||
| [2010/07/21 14:37:04] @ fredden joined channel #puppet | ||
| [2010/07/21 14:37:05] @ stewartl42 left channel #puppet () | ||
| [2010/07/21 14:37:54] @ Quit: olafrv: Quit: Page closed | ||
| [2010/07/21 14:40:55] @ Quit: mattock: Ping timeout: 258 seconds | ||
| [2010/07/21 14:45:11] <ReinH> pheezy: hiya | ||
| [2010/07/21 14:45:33] <pheezy> heya so I took your advice and got the latest and greatest dashboard, still no live aggregation though | ||
| [2010/07/21 14:46:47] <pheezy> http://pastie.org/1054395 | ||
| [2010/07/21 14:47:03] @ jatencio joined channel #puppet | ||
| [2010/07/21 14:47:09] <ReinH> thanks for the pastie... | ||
| [2010/07/21 14:47:19] <ReinH> pheezy: which puppet version? | ||
| [2010/07/21 14:47:51] @ Mikey joined channel #puppet | ||
| [2010/07/21 14:47:52] @ Quit: Mikey: Excess Flood | ||
| [2010/07/21 14:48:01] <pheezy> ReinH: 0.25.5 | ||
| [2010/07/21 14:48:15] <ReinH> pheezy: can you look at the reporting instructions here: http://github.com/reinh/puppet-dashboard/blob/master/README.markdown | ||
| [2010/07/21 14:48:23] <ReinH> and see if there are discrepancies with what you've done? | ||
| [2010/07/21 14:48:30] <ReinH> we're in the process of updating the official docs | ||
| [2010/07/21 14:49:00] <ReinH> which we should be releasing today, you just caught us while we're working on them :) | ||
| [2010/07/21 14:49:59] <pheezy> Yeah those are the ones I followed | ||
| [2010/07/21 14:50:24] <ReinH> ok | ||
| [2010/07/21 14:50:32] <ReinH> one sec | ||
| [2010/07/21 14:51:08] <ReinH> pheezy: and reports are not showing up in dashboard? | ||
| [2010/07/21 14:51:15] <pheezy> no sir | ||
| [2010/07/21 14:51:32] <pheezy> your docs show your libdir as /var/lib/puppet but mine comes up /var/lib/puppet/lib, only difference i see | ||
| [2010/07/21 14:51:33] <ReinH> so you're running dashboard on port 3000 on localhost via passenger? | ||
| [2010/07/21 14:51:37] <ReinH> hmm | ||
| [2010/07/21 14:51:41] <pheezy> yep | ||
| [2010/07/21 14:51:59] <pheezy> tcp 0 0 :::3000 :::* LISTEN 28225/httpd | ||
| [2010/07/21 14:52:04] <ReinH> so (sanity check) http://localhost:3000 loads dashboard | ||
| [2010/07/21 14:52:09] <pheezy> yep | ||
| [2010/07/21 14:52:09] <ReinH> nod | ||
| [2010/07/21 14:52:37] <ReinH> what does puppetmasterd --configprint reports show? | ||
| [2010/07/21 14:53:05] <pheezy> store, puppet_dashboard | ||
| [2010/07/21 14:53:15] <ReinH> well, that's good | ||
| [2010/07/21 14:54:05] <ReinH> ok, give me one sec to boot up a vm... | ||
| [2010/07/21 14:54:14] <pheezy> I tried putting the report processor in /usr/lib/ruby/site_ruby/1.8/puppet/reports as well, but nada | ||
| [2010/07/21 14:54:19] <pheezy> ok no rush | ||
| [2010/07/21 14:54:28] <pheezy> afk 5 min | ||
| [2010/07/21 14:54:33] @ Quit: lak: Quit: lak | ||
| [2010/07/21 14:55:47] @ Quit: ecapriolo: Quit: KVIrc Insomnia 4.0.0, revision: 4030, sources date: 20100125, built on: 2010-02-25 23:12:54 UTC http://www.kvirc.net/ | ||
| [2010/07/21 14:56:22] <ReinH> k | ||
| [2010/07/21 14:56:47] @ Mikey joined channel #puppet | ||
| [2010/07/21 14:57:04] @ Quit: Mikey: Excess Flood | ||
| [2010/07/21 14:57:19] <ReinH> when you get back, can you please (from dashboard's root): grep ReportsController#create log/{production,development}.log |wc -l | ||
| [2010/07/21 14:57:39] <ReinH> or just production.log if that's your environment | ||
| [2010/07/21 14:57:55] @ Quit: yannL: Remote host closed the connection | ||
| [2010/07/21 15:00:50] @ Quit: radkin: Remote host closed the connection | ||
| [2010/07/21 15:01:16] @ Quit: littleidea: Quit: littleidea | ||
| [2010/07/21 15:02:55] <pheezy> I get 0 | ||
| [2010/07/21 15:02:57] <hMz> is there any reason that storeconfigs makes all the scheme but doesnt make the index that i'm suppose to copy & paste from the doc? | ||
| [2010/07/21 15:03:09] <hMz> is it a ymmv? | ||
| [2010/07/21 15:05:02] @ Quit: bgupta: Quit: bgupta | ||
| [2010/07/21 15:05:18] <ReinH> pheezy: ok, so it isn't actually making requests... | ||
| [2010/07/21 15:06:35] <ReinH> pheezy: the README is incorrect btw, default libdir should be /var/lib/puppet/lib | ||
| [2010/07/21 15:06:44] <pheezy> oh ok | ||
| [2010/07/21 15:06:46] <ReinH> pheezy: and your report is in /var/lib/puppet/lib/puppet/reports/puppet_dashboard.rb, so that's correct | ||
| [2010/07/21 15:07:15] @ Quit: taihen: Read error: Operation timed out | ||
| [2010/07/21 15:07:42] @ spawnyd joined channel #puppet | ||
| [2010/07/21 15:08:06] <ReinH> hmm... | ||
| [2010/07/21 15:09:26] <ReinH> one sec | ||
| [2010/07/21 15:10:42] <ReinH> pheezy: can you grep puppetmasterd's logs for "puppet_dashboard"? | ||
| [2010/07/21 15:11:46] <pheezy> no hits in masterhttp.log :( | ||
| [2010/07/21 15:11:48] @ Quit: londo: Remote host closed the connection | ||
| [2010/07/21 15:12:33] <ReinH> ugh. | ||
| [2010/07/21 15:12:41] <pheezy> it's like the report isn't even being loaded | ||
| [2010/07/21 15:13:36] @ Quit: TREllis: Quit: Lost terminal | ||
| [2010/07/21 15:13:57] <ReinH> indeed | ||
| [2010/07/21 15:14:08] <ReinH> this is why I added the http processor to puppet 2.6.0 | ||
| [2010/07/21 15:14:11] <zirpu> abadr: /c | ||
| [2010/07/21 15:14:16] <zirpu> erp. ignore that. | ||
| [2010/07/21 15:14:24] <ReinH> trying to figure out where to put puppet_dashbaord.rb is... frustrating | ||
| [2010/07/21 15:14:30] <ReinH> so | ||
| [2010/07/21 15:14:44] <ReinH> apparently it doesn't lib /var/lib/puppet/lib/puppet/reports | ||
| [2010/07/21 15:15:08] <slapayoda> potentially stupid question alert! | ||
| [2010/07/21 15:15:29] <slapayoda> should I expect to see logging of client catalog compilations in my logdest here? | ||
| [2010/07/21 15:15:31] <slapayoda> puppet 14992 0.0 0.2 139172 36192 ? Ssl 15:06 0:00 /usr/bin/ruby /usr/sbin/puppetmasterd --masterport=28140 --logdest /var/log/puppet/puppetmasterd-test.log --servertype=mongrel --pidfile=/var/run/puppet/puppetmaster.28140.pid --trace --verbose --debug | ||
| [2010/07/21 15:15:53] <slapayoda> /var/log/puppet/puppetmasterd-test.log is showing plenty of debug-level information about startup | ||
| [2010/07/21 15:15:59] <ReinH> pheezy: hmm, reportserver is not related to dashboard | ||
| [2010/07/21 15:16:03] <slapayoda> but doesn't log anything when I start running clients against it | ||
| [2010/07/21 15:16:08] <ReinH> pheezy: you probably don't want that set | ||
| [2010/07/21 15:16:30] <pheezy> eh? but thats where the clients send reports right | ||
| [2010/07/21 15:16:52] <ReinH> pheezy: do you have dashboard running on a different puppetmaster from the one the clients talk to? | ||
| [2010/07/21 15:17:02] <ReinH> (from the one they get their configurations from) | ||
| [2010/07/21 15:17:07] <pheezy> yeah | ||
| [2010/07/21 15:17:10] <ReinH> oh | ||
| [2010/07/21 15:17:12] <slapayoda> I was expecting to see at least "Compiled catalog for hostname in 0.21 seconds" | ||
| [2010/07/21 15:17:12] <pheezy> and the CA is different as well | ||
| [2010/07/21 15:17:15] <ReinH> ok | ||
| [2010/07/21 15:17:28] <pheezy> i get all the store reports | ||
| [2010/07/21 15:17:44] <ReinH> and that host is at puppet-dashboard.example.com (I'm assuming the name has been changed to protect the innocent) | ||
| [2010/07/21 15:17:59] <pheezy> indeed sir, the same one running both a puppetmaster and the dashboard | ||
| [2010/07/21 15:18:05] <ReinH> ok | ||
| [2010/07/21 15:18:18] <ReinH> and *that* puppetmaster is the one that is configured for dashboard? | ||
| [2010/07/21 15:18:26] <pheezy> correct | ||
| [2010/07/21 15:18:28] <ReinH> ok | ||
| [2010/07/21 15:18:32] <ReinH> blargh | ||
| [2010/07/21 15:18:35] <ReinH> hmm | ||
| [2010/07/21 15:18:53] <ReinH> and report=true is set on all the clients (that you want to report)? | ||
| [2010/07/21 15:18:59] <ReinH> (it's insufficient to set it on the puppetmaster) | ||
| [2010/07/21 15:19:34] <pheezy> yep | ||
| [2010/07/21 15:19:52] <ReinH> or, instead, are the reports being stored as yaml in $reportdir? (usually /var/lib/puppet/reports) | ||
| [2010/07/21 15:20:23] @ verwilst joined channel #puppet | ||
| [2010/07/21 15:20:29] <pheezy> yes, they are being stored there | ||
| [2010/07/21 15:20:37] <ReinH> ok, so the reports are getting to the puppetmaster | ||
| [2010/07/21 15:20:39] <slapayoda> ah, nevermind. setting --masterlog=/var/log/puppet/puppetmasterd-test.log on top of those seems to give me what I wanted. | ||
| [2010/07/21 15:20:40] <ReinH> where is that reportdir, btw? | ||
| [2010/07/21 15:21:09] <pheezy> /var/lib/puppet/reports | ||
| [2010/07/21 15:21:13] <ReinH> ok | ||
| [2010/07/21 15:21:22] <ReinH> so it's most likely that the puppet_dashboard.rb report processor isn't getting loaded | ||
| [2010/07/21 15:21:33] <ReinH> have you restarded the puppetmasterd on the dashboard host? | ||
| [2010/07/21 15:21:51] <pheezy> Yes, multiple times. | ||
| [2010/07/21 15:22:18] <ReinH> interesting. | ||
| [2010/07/21 15:22:29] <ReinH> and your libdir is /var/lib/puppet/lib, not /var/lib/puppet | ||
| [2010/07/21 15:22:36] <pheezy> that is correct | ||
| [2010/07/21 15:22:44] <ReinH> ok, one sec | ||
| [2010/07/21 15:25:29] @ Guest34535 joined channel #puppet | ||
| [2010/07/21 15:25:42] <ReinH> if you specify a report processor that puppetmasterd can't find, it will log a warning when it the client sends it a report | ||
| [2010/07/21 15:25:46] <ReinH> like: warning: No report named 'foobar' | ||
| [2010/07/21 15:25:57] <ReinH> and you don't see any lines like that in your puppetmaster's log? | ||
| [2010/07/21 15:26:17] @ Quit: tonyskapunk: Quit: ERC Version 5.3 (IRC client for Emacs) | ||
| [2010/07/21 15:26:31] <ReinH> can you run it with --debug --verbose --no-daemonize and trigger a puppet run? | ||
| [2010/07/21 15:27:32] <pheezy> yeah, hold on | ||
| [2010/07/21 15:27:34] <ReinH> thanks | ||
| [2010/07/21 15:28:26] <pheezy> ah hah | ||
| [2010/07/21 15:28:27] <pheezy> err: Report puppet_dashboard failed: getaddrinfo: Name or service not known | ||
| [2010/07/21 15:28:32] <ReinH> ah ha | ||
| [2010/07/21 15:28:42] <ReinH> new errors are progress | ||
| [2010/07/21 15:28:44] <ReinH> of a sort | ||
| [2010/07/21 15:29:15] @ Quit: kaptk2: Quit: Leaving. | ||
| [2010/07/21 15:29:16] <ReinH> pheezy: can you pasteb in the contents of the puppet_dashboard.rb you have? | ||
| [2010/07/21 15:29:20] <ReinH> *pastebin | ||
| [2010/07/21 15:30:24] <pheezy> http://pastie.org/1054480 | ||
| [2010/07/21 15:30:39] <ReinH> ok | ||
| [2010/07/21 15:31:21] <ReinH> to clarify, the puppetmasterd that uses the puppet_dashboard processor is on the same machine as the dasboard? | ||
| [2010/07/21 15:31:25] <pheezy> yeah | ||
| [2010/07/21 15:31:27] <pheezy> i found the problem | ||
| [2010/07/21 15:31:34] <ReinH> oh? | ||
| [2010/07/21 15:31:46] <pheezy> /etc/hosts was missing a localhost entry | ||
| [2010/07/21 15:31:51] * pheezy kicks vmware | ||
| [2010/07/21 15:31:52] <jatencio> ls | ||
| [2010/07/21 15:31:52] <ReinH> ... | ||
| [2010/07/21 15:31:54] <ReinH> awesome | ||
| [2010/07/21 15:31:59] <ReinH> you know, you can use puppet to manage that | ||
| [2010/07/21 15:32:00] * ReinH ducks | ||
| [2010/07/21 15:32:10] <ReinH> ralsh :p | ||
| [2010/07/21 15:32:35] <ReinH> yeah, my next question would have been: can you curl to localhost:3000 on that box? | ||
| [2010/07/21 15:32:36] @ tanto joined channel #puppet | ||
| [2010/07/21 15:32:41] <ReinH> thanks for reading my mind ;) | ||
| [2010/07/21 15:34:19] <ReinH> ok, please tell me it works now ;) | ||
| [2010/07/21 15:34:42] @ fbe joined channel #puppet | ||
| [2010/07/21 15:35:19] <pheezy> wow thanks for helping with | ||
| [2010/07/21 15:35:35] <pheezy> i think i have to use puppet to manage that now :X | ||
| [2010/07/21 15:35:37] <ReinH> yeah, np, I guess I should start with: "is your /etc/hosts sane?" | ||
| [2010/07/21 15:35:46] <ReinH> "do you have a cp command?" | ||
| [2010/07/21 15:35:46] <ReinH> etc | ||
| [2010/07/21 15:35:47] <ReinH> :p | ||
| [2010/07/21 15:35:58] @ zorzar_ joined channel #puppet | ||
| [2010/07/21 15:36:04] <ReinH> are your reports making their way into dashboard? | ||
| [2010/07/21 15:36:09] <pheezy> why the heck people bind the hostname to 127.0.0.1 is beyond me | ||
| [2010/07/21 15:36:13] <ReinH> yeah | ||
| [2010/07/21 15:36:13] <pheezy> yeah, they are pouring in | ||
| [2010/07/21 15:36:14] <ReinH> meh | ||
| [2010/07/21 15:36:25] <ReinH> pheezy: run rake reports:import to grab the stored yaml reports | ||
| [2010/07/21 15:36:48] <pheezy> yeah saw that, just did it :D | ||
| [2010/07/21 15:36:48] <ReinH> it may take a while depending on how many there are | ||
| [2010/07/21 15:36:50] <ReinH> :D | ||
| [2010/07/21 15:36:54] <ReinH> awesome | ||
| [2010/07/21 15:36:56] @ jatencio left channel #puppet () | ||
| [2010/07/21 15:37:21] <ReinH> I was about to do a final dashboard test on 0.25.x anyway, now I don't have to :D | ||
| [2010/07/21 15:38:19] <ReinH> pheezy: we'll be releasing 1.0.3 today so you can get yourself on an official release instead of edge | ||
| [2010/07/21 15:38:24] <ReinH> and thanks for helping test it :) | ||
| [2010/07/21 15:38:47] <pheezy> np, thanks for helping me out | ||
| [2010/07/21 15:38:57] @ Quit: zorzar: Ping timeout: 240 seconds | ||
| [2010/07/21 15:39:00] <pheezy> so tonight i can update my puppet-dashboard module :) | ||
| [2010/07/21 15:39:25] <ReinH> you have a puppet module for puppet-dashboard? | ||
| [2010/07/21 15:39:47] <pheezy> yeah | ||
| [2010/07/21 15:39:52] <ReinH> can you add it to the module site? that would be totally sweet | ||
| [2010/07/21 15:39:59] <pheezy> yeah i need to polish it up a bit | ||
| [2010/07/21 15:40:17] <ReinH> great! | ||
| [2010/07/21 15:40:19] <pheezy> and it's totally dependent on it being an RPM, though I did consider using ya'lls repo provider | ||
| [2010/07/21 15:40:28] <ReinH> ah | ||
| [2010/07/21 15:40:28] <pheezy> but sure i'll put it up there | ||
| [2010/07/21 15:40:42] <ReinH> yeah, we'll review it and possibly suggest improvements | ||
| [2010/07/21 15:40:53] <ReinH> and then let people know it's available as part of the dashboard docs | ||
| [2010/07/21 15:40:59] <ReinH> that would be fantastic :) | ||
| [2010/07/21 15:41:23] <tanto> how are you guys autoprovisioning cloud servers? | ||
| [2010/07/21 15:41:35] <pheezy> np, will do it, probably today or tomorrow | ||
| [2010/07/21 15:41:38] @ Quit: notbrien: Quit: notbrien | ||
| [2010/07/21 15:42:22] @ bgupta joined channel #puppet | ||
| [2010/07/21 15:42:37] <ReinH> pheezy: sweet, can you email me rein<at>puppetlabs.com when it's up? | ||
| [2010/07/21 15:42:43] @ be200fx left channel #puppet () | ||
| [2010/07/21 15:42:45] <pheezy> yep will do! | ||
| [2010/07/21 15:42:52] <ReinH> also, forge site is http://forge.puppetlabs.com/ (ofc) | ||
| [2010/07/21 15:42:56] <ReinH> awesome | ||
| [2010/07/21 15:43:24] <ReinH> pheezy: and I idle here all the time if you have issues | ||
| [2010/07/21 15:43:31] @ lak joined channel #puppet | ||
| [2010/07/21 15:43:32] <ReinH> and tickets are always appreciated | ||
| [2010/07/21 15:43:44] <pheezy> yeah i've got a list, need to put them in redmine | ||
| [2010/07/21 15:43:48] <ReinH> please do | ||
| [2010/07/21 15:43:55] <pheezy> anyway, must run, thanks again for the help! | ||
| [2010/07/21 15:44:01] <ReinH> yeah np, thank you as well | ||
| [2010/07/21 15:44:13] @ Quit: pheezy: Remote host closed the connection | ||
| [2010/07/21 15:44:15] <hal-j> Tanto: I know it's not the answer you want to hear, but the transient ec2 instances that come and go are started from images I cut from a production instance | ||
| [2010/07/21 15:44:22] @ Quit: verwilst: Quit: Ex-Chat | ||
| [2010/07/21 15:44:45] <hal-j> and I re-cut those images when I do upgrades to the system in question | ||
| [2010/07/21 15:45:13] <hal-j> the constantly running instances are all managed by puppet | ||
| [2010/07/21 15:45:44] <tanto> i want to spin up new instances and not have to configure anything on them =) | ||
| [2010/07/21 15:49:18] <hal-j> pretty much what I get, although it takes a bit of effort to keep them updated | ||
| [2010/07/21 15:51:18] <hal-j> the "manual" process, for me anyhow, for bringing up new instances that I don't have precut images for — bring up N instances, log in, set hostname, add hostname(s) to a node definition (all instances that share a common use, share a node definition) and kick off puppet | ||
| [2010/07/21 15:51:28] <hal-j> so it's not that tough, but I haven't gotten around to making it better :) | ||
| [2010/07/21 15:53:43] <Volcane> hal-j, tanto: http://mcollective.blip.tv/file/3876576/ there's a video of a generic vm being taken to a specific use using puppet and some stuff i built | ||
| [2010/07/21 15:54:22] <ReinH> Volcane: wait, mcollective has a blip.tv station?! | ||
| [2010/07/21 15:54:27] <ReinH> holy awesome. | ||
| [2010/07/21 15:54:40] <Volcane> moving things slowly over to it yeah | ||
| [2010/07/21 15:54:47] <ReinH> needs more romantic comedy | ||
| [2010/07/21 15:54:50] <Volcane> :P | ||
| [2010/07/21 15:54:51] <hal-j> Volcane: thanks. I'll take a look | ||
| [2010/07/21 15:54:55] <ReinH> a puppet meets a message bus | ||
| [2010/07/21 15:54:57] <ReinH> hilarity ensues | ||
| [2010/07/21 15:54:58] <ReinH> etc | ||
| [2010/07/21 15:54:58] <Volcane> i f'n hate blip.tv | ||
| [2010/07/21 15:55:13] <hal-j> working on that is in my TODO list, but I haven't gotten to it | ||
| [2010/07/21 15:55:36] <tanto> thanks guys | ||
| [2010/07/21 15:55:50] <hal-j> TBH - I moved from our own datacenter to ec2 in a panic'd rush last fall, when as part of the restructuring / acquisition we went through, meant I was losing all proper datacenters within 2 weeks | ||
| [2010/07/21 15:56:06] <Volcane> hal-j: that sux | ||
| [2010/07/21 15:56:12] <hal-j> a feat I couldn't have accomplished without puppet, mind you | ||
| [2010/07/21 15:56:55] <tanto> volcane you don't want to run autosign? | ||
| [2010/07/21 15:57:07] <Volcane> tanto: no | ||
| [2010/07/21 15:57:20] <tanto> blip.tv doesn't seem to like chrome | ||
| [2010/07/21 15:57:22] * hal-j won't use autosign either | ||
| [2010/07/21 15:57:23] <Volcane> autosign doesnt solve the problem either | ||
| [2010/07/21 15:57:37] <Volcane> next time u get a node with a matching hostname, autosign doesnt autoclean | ||
| [2010/07/21 15:58:20] <Volcane> so you need to deal with cleanup anyway, if u can automate cleanup you can automate sign | ||
| [2010/07/21 15:58:25] @ bgupta_ joined channel #puppet | ||
| [2010/07/21 16:00:53] @ Quit: bgupta: Ping timeout: 265 seconds | ||
| [2010/07/21 16:01:18] @ FOCer joined channel #puppet | ||
| [2010/07/21 16:03:01] @ Quit: Guest34535: Remote host closed the connection | ||
| [2010/07/21 16:04:33] @ Mikey_ joined channel #puppet | ||
| [2010/07/21 16:04:34] @ Quit: Mikey_: Max SendQ exceeded | ||
| [2010/07/21 16:04:37] @ Quit: akoma1s: Remote host closed the connection | ||
| [2010/07/21 16:04:40] @ akoma1s joined channel #puppet | ||
| [2010/07/21 16:04:54] <hal-j> re: auto-provisioning, there's only one subsystem I'd need it for, and I've got that managed pretty well as it is. and I don't know how long I'll be _in_ ec2 | ||
| [2010/07/21 16:05:29] <Volcane> does ec2 work well for you? | ||
| [2010/07/21 16:05:40] <Volcane> compared to self hosting etc? | ||
| [2010/07/21 16:06:00] <hal-j> well | ||
| [2010/07/21 16:06:33] <hal-j> ec2 is a tool, like anything else. If you read the forums you'd think it's horribly unreliable and incapable of running a production environmenty | ||
| [2010/07/21 16:06:58] <hal-j> however, if you plan for any instance to fail at any point in time, and design the system to tolerate that, you do fine | ||
| [2010/07/21 16:07:09] <hal-j> and I've been very happy with it | ||
| [2010/07/21 16:07:35] <Volcane> yah, but given how you moved - without that planning presumably, how'd you find it? did you have to change architectures a fair bit to cope with failure? | ||
| [2010/07/21 16:08:06] @ taihen joined channel #puppet | ||
| [2010/07/21 16:08:11] <hal-j> I changed some things on the fly, but we had a pretty robust system to start with | ||
| [2010/07/21 16:08:18] <Volcane> nice | ||
| [2010/07/21 16:08:42] @ Quit: jdcasey: Remote host closed the connection | ||
| [2010/07/21 16:08:53] <Volcane> all the imaging stuff is frustrating, and ebs volumes probably not ideal for every vm | ||
| [2010/07/21 16:09:00] <Volcane> i dont use it much, labs style stuf mostly | ||
| [2010/07/21 16:09:05] <Volcane> just curious about your experience | ||
| [2010/07/21 16:09:11] <hal-j> and despite the fact that I was a dev at the company initially, and became the entire ops team during the restructuring, I knew the system pretty well, so moving it wasn't all that hard | ||
| [2010/07/21 16:09:39] <hal-j> in the end, it's VERY similar to having a datacenter that you can rack up new hosts in, in a matter of minutes | ||
| [2010/07/21 16:10:16] <hal-j> without a bit of the flexibility, and with some additional (but different flexibility) | ||
| [2010/07/21 16:10:25] <Volcane> sure but the shorter perceived life expectancy of nodes means you need to push automation into places you might not have had it like load balancers | ||
| [2010/07/21 16:10:44] <Volcane> and monitoring capabilities need to be more dynamic etc | ||
| [2010/07/21 16:11:21] @ Mikey_ joined channel #puppet | ||
| [2010/07/21 16:11:41] <Volcane> its nice for sure if u have a compatible workload etc, and if u can justify the cost by needing to scale quickly etc | ||
| [2010/07/21 16:17:21] <hal-j> well, we had a pretty strict rule well before moving to ec2, that all changes were in puppet, so replacing lost instances has been pretty easy | ||
| [2010/07/21 16:17:52] <Volcane> nice, just shows, doing things right pays off | ||
| [2010/07/21 16:18:14] <hal-j> moving from proper load balancers previously, to the ELBs offered by AWS was a bit of a hassle, but eventually I settled on what worked, and worked around what didn't | ||
| [2010/07/21 16:19:48] <hal-j> ultimately, I've had no insurmountable problems, and I didn't need to make too many sacrifices | ||
| [2010/07/21 16:19:57] <Volcane> nice | ||
| [2010/07/21 16:20:14] <hal-j> yeah, if not for the automation already in place, there's no way I could have pulled it off. | ||
| [2010/07/21 16:20:59] <hal-j> (thanks, in a large part, to the ops team that preceded me) | ||
| [2010/07/21 16:22:41] <Volcane> kewl, well thanks for telling me your experiences. time for bed | ||
| [2010/07/21 16:23:06] <hal-j> night | ||
| [2010/07/21 16:23:56] <benkevan> in the new 2.6.0 i notice that $puppetd is being called with ruby /usr/sbin/puppetd instead of /usr/bin/ruby /usr/sbin/puppetd | ||
| [2010/07/21 16:24:06] <benkevan> which causes lsb error codes of 7 (startproc) | ||
| [2010/07/21 16:24:14] <benkevan> anyone know where this call is coming from? | ||
| [2010/07/21 16:25:37] <benkevan> is it because /usr/sbin/puppetd is now called with #!/usr/bin/env ruby instead of /usr/bin/ruby specifically? | ||
| [2010/07/21 16:26:30] @ Quit: fbe: Ping timeout: 265 seconds | ||
| [2010/07/21 16:28:13] @ [GuS] joined channel #puppet | ||
| [2010/07/21 16:28:13] @ Quit: [GuS]: Changing host | ||
| [2010/07/21 16:28:13] @ [GuS] joined channel #puppet | ||
| [2010/07/21 16:30:42] @ Quit: Mikey_: Remote host closed the connection | ||
| [2010/07/21 16:31:56] <hal-j> #!/usr/bin/env ruby will cause the first ruby in your path to be used | ||
| [2010/07/21 16:32:07] @ Quit: spawnyd: Ping timeout: 276 seconds | ||
| [2010/07/21 16:32:21] <hal-j> what does "which ruby" tell you, when running as the user that puppetd runs as (almost certainly root) | ||
| [2010/07/21 16:32:57] <benkevan> in this instance.. it doesn't seem to be working .. once I hardcode /usr/bin/ruby $puppetd in init scripts.. it starts with correct startproc exit code of 0 | ||
| [2010/07/21 16:33:23] <benkevan> otherwise.. it doesn't exit correctly (give an exit of 7).. but I see ruby /usr/sbin/puppetd running | ||
| [2010/07/21 16:33:49] @ Guest87063 joined channel #puppet | ||
| [2010/07/21 16:34:13] <benkevan> the initscripts are broken.. because pidof is broken | ||
| [2010/07/21 16:34:43] <benkevan> guess i'll look at it tomorrow.. time to go home.. | ||
| [2010/07/21 16:34:53] <benkevan> but the packages I just built for SLES11 are broken because of this env change.. | ||
| [2010/07/21 16:34:54] @ Quit: Guest87063: Client Quit | ||
| [2010/07/21 16:37:30] @ Quit: benkevan: Quit: Lost terminal | ||
| [2010/07/21 16:42:54] @ londo joined channel #puppet | ||
| [2010/07/21 16:45:21] @ Quit: lak: Quit: lak | ||
| [2010/07/21 16:47:08] @ Quit: alfism: Read error: Operation timed out | ||
| [2010/07/21 16:52:15] @ kolla joined channel #puppet | ||
| [2010/07/21 16:55:59] @ alfism joined channel #puppet | ||
| [2010/07/21 16:56:20] @ Quit: alfism: Client Quit | ||
| [2010/07/21 16:59:18] @ Quit: gniks: Quit: Leaving. | ||
| [2010/07/21 17:04:31] @ Quit: conconcatenate: Ping timeout: 252 seconds | ||
| [2010/07/21 17:09:37] @ Wandrewvious joined channel #puppet | ||
| [2010/07/21 17:09:56] @ Quit: bgupta_: Quit: bgupta_ | ||
| [2010/07/21 17:10:17] @ bgupta joined channel #puppet | ||
| [2010/07/21 17:13:59] @ Quit: WALoeIII: Ping timeout: 276 seconds | ||
| [2010/07/21 17:14:21] @ Quit: Wandrewvious: Ping timeout: 265 seconds | ||
| [2010/07/21 17:14:50] @ Quit: bgupta: Ping timeout: 265 seconds | ||
| [2010/07/21 17:15:28] @ ScottD joined channel #puppet | ||
| [2010/07/21 17:15:39] <ScottD> I am having a problem with swtiching to puppet 2.6 | ||
| [2010/07/21 17:15:57] <ScottD> I used to run it locally on a machine with | ||
| [2010/07/21 17:16:10] <ScottD> sudo puppet --modulepath=`pwd`/modules --templatedir=`pwd`/templates --verbose manifests/site.pp | ||
| [2010/07/21 17:16:21] <ScottD> I tried switching to | ||
| [2010/07/21 17:16:21] <ScottD> sudo puppet apply --modulepath=`pwd`/modules --templatedir=`pwd`/templates --verbose manifests/site.pp | ||
| [2010/07/21 17:17:03] <ScottD> But now I get errors like: | ||
| [2010/07/21 17:17:03] <ScottD> No route to host - connect(2) Could not retrieve file metadata for puppet:///modules/snmp/snmpd.conf: No route to host - connect(2) at /home/scottd/puppet_manifests/modules/snmp/manifests/init.pp:23 | ||
| [2010/07/21 17:17:18] <ScottD> But I am not trying to connect to a server, I just want to run and test on a single machine. | ||
| [2010/07/21 17:25:47] @ Djelibeybi joined channel #puppet | ||
| [2010/07/21 17:25:51] <nevyn> hrm | ||
| [2010/07/21 17:30:33] <ScottD> I have had to roll back to 0.25.5 for now | ||
| [2010/07/21 17:32:39] @ littleidea joined channel #puppet | ||
| [2010/07/21 17:35:06] @ gniks joined channel #puppet | ||
| [2010/07/21 17:36:22] @ Quit: jaredrhine: Quit: Leaving. | ||
| [2010/07/21 17:39:06] @ Quit: ScottD: Quit: Page closed | ||
| [2010/07/21 17:44:55] @ Quit: littleidea: Quit: littleidea | ||
| [2010/07/21 17:54:46] @ Quit: Hypnoz: Ping timeout: 252 seconds | ||
| [2010/07/21 17:56:15] @ Quit: jab_doa: Quit: Verlassend | ||
| [2010/07/21 18:01:57] <Djelibeybi> jamesturnbull: http://www.puppetlabs.com/misc/download-options/ has the latest stable as 0.25.5 | ||
| [2010/07/21 18:05:15] <jamesturnbull> Djelibeybi: thanks fixed | ||
| [2010/07/21 18:06:40] @ Djelibeybi left channel #puppet ("Leaving") | ||
| [2010/07/21 18:08:50] @ thekad is now known as thekad-afk | ||
| [2010/07/21 18:12:20] <rcrowley> I've just created a new master on 2.6 and upgraded one client to 2.6. Everything's awesome except its inability to load a plugin that defines a type and a provider for that type. The files are created as expected but they are empty. | ||
| [2010/07/21 18:12:37] <rcrowley> Every single run shows a line like this for each file, always exactly the same line: notice: /File[/var/lib/puppet/lib/puppet/type/users.rb]/content: content changed '{md5}d41d8cd98f00b204e9800998ecf8427e' to '{md5}5b3a13c1bc9031a404c10b1ff841dfb9' | ||
| [2010/07/21 18:13:10] <rcrowley> Is this a file serving issue? A permissions issue? | ||
| [2010/07/21 18:16:47] <jamesturnbull> rcrowley: that's after multiple runs? | ||
| [2010/07/21 18:17:05] <jamesturnbull> rcrowley: that sounds like a bug but not one I've seen - platform? mongrel? webrick? passenger? | ||
| [2010/07/21 18:17:25] <jamesturnbull> rcrowley: no other debug messages? have you run with --debug --trace ? | ||
| [2010/07/21 18:20:30] <rcrowley> This has been happening for hours, despite all my sprinkling puts through the Puppet gem. I'm on Ubuntu 10.04, running under Apache/Passenger. I'm checking now whether running standalone changes things. | ||
| [2010/07/21 18:23:05] @ pheezy joined channel #puppet | ||
| [2010/07/21 18:24:10] <rcrowley> It appears the problem is Rack/Apache/Passenger related. There are other problems I'm looking into but the files have the correct content when I ran against standalone Puppet master. | ||
| [2010/07/21 18:24:35] <cparedes> this is probably the worst channel to ask this in, but i'm wondering what sort of other conf mgmt tools are viable that are not ruby based? | ||
| [2010/07/21 18:24:42] * cparedes prefers puppet over everything, but this is for work. | ||
| [2010/07/21 18:24:54] @ gniks1 joined channel #puppet | ||
| [2010/07/21 18:24:55] @ Quit: gniks: Ping timeout: 265 seconds | ||
| [2010/07/21 18:24:56] <cparedes> and i don't have much of a say in what choice of language the thing is coded up in | ||
| [2010/07/21 18:24:57] <jamesturnbull> cparedes: cfengine | ||
| [2010/07/21 18:25:08] <nevyn> cfengine BCF2 if you have XML love | ||
| [2010/07/21 18:25:09] <jamesturnbull> cparedes: oh you said viable? | ||
| [2010/07/21 18:25:15] @ Quit: gniks1: Client Quit | ||
| [2010/07/21 18:25:17] <jamesturnbull> :P | ||
| [2010/07/21 18:25:18] <cparedes> jamesturnbull: :) | ||
| [2010/07/21 18:25:22] <nevyn> there's an automation framework for data centres that IBM is backing. | ||
| [2010/07/21 18:25:24] <nevyn> umm. | ||
| [2010/07/21 18:25:30] <jamesturnbull> BMC has a tool | ||
| [2010/07/21 18:25:31] <jamesturnbull> as does CA | ||
| [2010/07/21 18:25:32] <nevyn> most of them are aweful tho. | ||
| [2010/07/21 18:25:37] <nevyn> jamesturnbull: no they don't | ||
| [2010/07/21 18:25:37] <jamesturnbull> 7 figures naturally | ||
| [2010/07/21 18:25:42] <cparedes> ah, man. | ||
| [2010/07/21 18:25:42] <nevyn> not in this space. | ||
| [2010/07/21 18:26:03] <nevyn> I suppose Spectrum automation manager is sorta like puppet kinda.. | ||
| [2010/07/21 18:26:03] <jamesturnbull> nevyn: they do - they're just crapola :) | ||
| [2010/07/21 18:26:04] <cparedes> i really want to get puppet deployed. :x the rest of the team doesn't like ruby, unfortunately | ||
| [2010/07/21 18:26:15] <nevyn> but it's really not. | ||
| [2010/07/21 18:26:20] <jamesturnbull> cparedes: what's it got to do with Ruby? | ||
| [2010/07/21 18:26:25] <jamesturnbull> cparedes: you don't need to know any Ruby | ||
| [2010/07/21 18:26:32] <jamesturnbull> cparedes: to write Puppet manifests | ||
| [2010/07/21 18:26:34] <cparedes> jamesturnbull: possibly the fact that it sucks up a lot of memory | ||
| [2010/07/21 18:26:38] <cparedes> i know. :) | ||
| [2010/07/21 18:26:46] <nevyn> jamesturnbull: people are all like we shouldn't install another scripting language on servers | ||
| [2010/07/21 18:26:49] <cparedes> i've written a ton of manifests at my last job. | ||
| [2010/07/21 18:26:55] <nevyn> increased attack surface blah blah blah | ||
| [2010/07/21 18:27:05] <jamesturnbull> cparedes: lot's of ways around that - run clients with cron, etc | ||
| [2010/07/21 18:27:12] <cparedes> aye. | ||
| [2010/07/21 18:27:14] <cparedes> hm | ||
| [2010/07/21 18:27:18] <nevyn> jamesturnbull: you going to lca next year? given it's now the other side of the world. | ||
| [2010/07/21 18:27:28] <jamesturnbull> nevyn: unlikely - someone from the company will be | ||
| [2010/07/21 18:27:28] <cparedes> i mostly have to type up a comparison sheet between puppet and a few other alternatives | ||
| [2010/07/21 18:27:45] <nevyn> puppet: Awesome | ||
| [2010/07/21 18:27:47] @ Quit: pheezy: Remote host closed the connection | ||
| [2010/07/21 18:27:50] <nevyn> others..not awesome | ||
| [2010/07/21 18:27:54] <cparedes> and i'm really trying hard not to evangelize for puppet (it'd be really disingenuous, IMO.) but man. i really hope they pick puppet. | ||
| [2010/07/21 18:28:00] <nevyn> there you are whack it in a table. | ||
| [2010/07/21 18:28:03] <nevyn> ;) | ||
| [2010/07/21 18:28:39] <rcrowley> jamesturnbull: Read through --debug --trace again in more detail. The line I pasted above with MD5 sums is the most telling. There are also a lot of these: "warning: Loaded puppet/type/users but no class was created" | ||
| [2010/07/21 18:28:56] <rcrowley> They seem to be direct results of the files being empty. | ||
| [2010/07/21 18:29:02] @ gniks joined channel #puppet | ||
| [2010/07/21 18:29:11] @ pheezy joined channel #puppet | ||
| [2010/07/21 18:29:42] <jamesturnbull> rcrowley: can you log a ticket please? | ||
| [2010/07/21 18:30:29] <nevyn> arrg. | ||
| [2010/07/21 18:30:41] <rcrowley> Certainly. This smells like file serving config to me. Is there any documentation of fileserver.conf and how to set it up for plugin syncing? | ||
| [2010/07/21 18:30:44] <jamesturnbull> rcrowley: I can't replicate that - what passenger version? and what's in your config.ru? | ||
| [2010/07/21 18:30:46] <nevyn> jamesturnbull: one of the things I really liked about reductive was there was pricing for the suppport options on the website. | ||
| [2010/07/21 18:30:56] <jamesturnbull> rcrowley: you shouldn't have to do anything | ||
| [2010/07/21 18:30:59] <nevyn> this has gone away now. which makes me sad. | ||
| [2010/07/21 18:31:16] <rcrowley> "anything" as in empty fileserver.conf? | ||
| [2010/07/21 18:31:32] <jamesturnbull> rcrowley: yep - Puppet automatically creates the plugins mount | ||
| [2010/07/21 18:31:35] @ Quit: raven: Disconnected by services | ||
| [2010/07/21 18:31:47] <rcrowley> That's excellent! | ||
| [2010/07/21 18:31:49] @ raven_ joined channel #puppet | ||
| [2010/07/21 18:31:56] <jamesturnbull> rcrowley: and you'd be seeing it on webrick if it was file serving | ||
| [2010/07/21 18:32:06] <jamesturnbull> if it's just passenger then it's somewhere in that code | ||
| [2010/07/21 18:32:18] @ mboyd_ joined channel #puppet | ||
| [2010/07/21 18:32:19] @ Quit: mboyd: Disconnected by services | ||
| [2010/07/21 18:32:28] <rcrowley> That's true. | ||
| [2010/07/21 18:37:28] <jamesturnbull> nevyn: it now varies | ||
| [2010/07/21 18:37:36] <jamesturnbull> nevyn: email scott@puppetlabs.com | ||
| [2010/07/21 18:37:58] @ mboyd_ is now known as mboyd | ||
| [2010/07/21 18:45:02] <hMz> becoming big time | ||
| [2010/07/21 18:45:11] <hMz> soon you'll have sales guys who wont talk to you without coming out with persentations ;P | ||
| [2010/07/21 18:45:29] <jamesturnbull> hMz: not us :) | ||
| [2010/07/21 18:45:34] <jamesturnbull> hMz: we're pretty low key | ||
| [2010/07/21 18:46:01] <hMz> oh because i'm good at making presentations *coughs* | ||
| [2010/07/21 18:46:05] <hMz> :D | ||
| [2010/07/21 18:54:47] @ Quit: jmccune: Ping timeout: 264 seconds | ||
| [2010/07/21 18:58:15] @ jaredrhine joined channel #puppet | ||
| [2010/07/21 19:01:55] @ davea1 joined channel #puppet | ||
| [2010/07/21 19:04:34] @ Quit: pheezy: Remote host closed the connection | ||
| [2010/07/21 19:05:51] @ crdant joined channel #puppet | ||
| [2010/07/21 19:06:21] @ jcharette joined channel #puppet | ||
| [2010/07/21 19:06:42] <jcharette> anyone have a debian package for 2.6.0? | ||
| [2010/07/21 19:06:42] <davea1> I am setting up a new webserver via puppet and I am at the point of creating the vhosts- I am doing the same thing for each one- and the only difference between them is the domain name- how would I process the array to create the same defines over and over? | ||
| [2010/07/21 19:07:13] <davea1> that is, assuming that I put all the domains in an array | ||
| [2010/07/21 19:08:09] <jamesturnbull> davea1: there are some examples of this at the Puppet Forge - http://forge.puppetlabs.com | ||
| [2010/07/21 19:08:26] * davea1 is looking now- thanks! | ||
| [2010/07/21 19:09:34] <jcharette> james, as much as I'd love to help you guys out ops wise, i can't move to portland :( | ||
| [2010/07/21 19:12:23] <jamesturnbull> jcharette: :( | ||
| [2010/07/21 19:12:42] @ Quit: mclarke: Quit: mclarke | ||
| [2010/07/21 19:14:03] <jcharette> any chance of an ubuntu package being created for 2.6.0? | ||
| [2010/07/21 19:17:13] <cparedes> i'd love to move to portland, but not yet. maybe in a few years. :P | ||
| [2010/07/21 19:17:22] @ alexine_dsouza joined channel #puppet | ||
| [2010/07/21 19:19:42] <davea1> my wife said that Texas was as far west as she ever wanted to move :/ | ||
| [2010/07/21 19:20:57] @ Quit: jaredrhine: Ping timeout: 240 seconds | ||
| [2010/07/21 19:32:40] @ Quit: Bass10: Ping timeout: 240 seconds | ||
| [2010/07/21 19:33:31] @ ohadlevy left channel #puppet () | ||
| [2010/07/21 19:40:49] @ ohadlevy joined channel #puppet | ||
| [2010/07/21 19:42:45] @ Quit: lutter: Quit: Leaving. | ||
| [2010/07/21 19:46:29] @ pheezy joined channel #puppet | ||
| [2010/07/21 19:58:18] @ Quit: jcharette: Ping timeout: 240 seconds | ||
| [2010/07/21 20:00:30] @ Quit: rcrowley: Quit: rcrowley | ||
| [2010/07/21 20:02:41] @ cyrus_mc joined channel #puppet | ||
| [2010/07/21 20:06:27] @ Quit: pheezy: Remote host closed the connection | ||
| [2010/07/21 20:09:38] @ Quit: [GuS]: Read error: Connection reset by peer | ||
| [2010/07/21 20:13:30] @ rcrowley joined channel #puppet | ||
| [2010/07/21 20:14:38] @ mclarke joined channel #puppet | ||
| [2010/07/21 20:14:56] @ squirrel_labs joined channel #puppet | ||
| [2010/07/21 20:14:56] @ Quit: squirrel_labs: Client Quit | ||
| [2010/07/21 20:15:05] @ Quit: mboyd: Read error: Connection reset by peer | ||
| [2010/07/21 20:15:42] @ mboyd joined channel #puppet | ||
| [2010/07/21 20:16:43] @ Quit: mboyd: Read error: Connection reset by peer | ||
| [2010/07/21 20:18:29] @ mboyd_ joined channel #puppet | ||
| [2010/07/21 20:18:31] @ mboyd_ is now known as mboyd | ||
| [2010/07/21 20:24:27] @ pheezy joined channel #puppet | ||
| [2010/07/21 20:26:37] @ Quit: rcrowley: *.net *.split | ||
| [2010/07/21 20:26:37] @ Quit: davea1: *.net *.split | ||
| [2010/07/21 20:26:37] @ Quit: alban2: *.net *.split | ||
| [2010/07/21 20:26:37] @ Quit: huggie: *.net *.split | ||
| [2010/07/21 20:26:37] @ Quit: Schmidt: *.net *.split | ||
| [2010/07/21 20:26:38] @ Quit: choffee: *.net *.split | ||
| [2010/07/21 20:26:38] @ Quit: p3rror: *.net *.split | ||
| [2010/07/21 20:26:38] @ Quit: tyler79: *.net *.split | ||
| [2010/07/21 20:26:38] @ Quit: emiddd_: *.net *.split | ||
| [2010/07/21 20:26:38] @ Quit: tmz: *.net *.split | ||
| [2010/07/21 20:26:38] @ Quit: hMz: *.net *.split | ||
| [2010/07/21 20:26:39] @ Quit: FoXMaN: *.net *.split | ||
| [2010/07/21 20:26:39] @ Quit: cparedes: *.net *.split | ||
| [2010/07/21 20:26:39] @ Quit: crb: *.net *.split | ||
| [2010/07/21 20:26:39] @ Quit: Mandus: *.net *.split | ||
| [2010/07/21 20:26:39] @ Quit: Demosthenes: *.net *.split | ||
| [2010/07/21 20:28:45] @ choffee joined channel #puppet | ||
| [2010/07/21 20:31:56] @ jcharette joined channel #puppet | ||
| [2010/07/21 20:31:57] @ rcrowley joined channel #puppet | ||
| [2010/07/21 20:31:57] @ davea1 joined channel #puppet | ||
| [2010/07/21 20:31:57] @ alban2 joined channel #puppet | ||
| [2010/07/21 20:31:57] @ huggie joined channel #puppet | ||
| [2010/07/21 20:31:57] @ Schmidt joined channel #puppet | ||
| [2010/07/21 20:31:57] @ p3rror joined channel #puppet | ||
| [2010/07/21 20:31:57] @ tyler79 joined channel #puppet | ||
| [2010/07/21 20:31:57] @ emiddd_ joined channel #puppet | ||
| [2010/07/21 20:31:57] @ tmz joined channel #puppet | ||
| [2010/07/21 20:31:57] @ FoXMaN joined channel #puppet | ||
| [2010/07/21 20:31:57] @ cparedes joined channel #puppet | ||
| [2010/07/21 20:31:57] @ crb joined channel #puppet | ||
| [2010/07/21 20:31:57] @ Mandus joined channel #puppet | ||
| [2010/07/21 20:31:57] @ Demosthenes joined channel #puppet | ||
| [2010/07/21 20:33:48] @ Quit: cyrus_mc: Ping timeout: 240 seconds | ||
| [2010/07/21 20:50:21] @ Quit: pheezy: Remote host closed the connection | ||
| [2010/07/21 20:51:31] @ Quit: mboyd: Read error: Connection reset by peer | ||
| [2010/07/21 20:52:12] @ mboyd joined channel #puppet | ||
| [2010/07/21 20:52:13] @ Quit: jcharette: Ping timeout: 246 seconds | ||
| [2010/07/21 20:58:34] @ jense joined channel #puppet | ||
| [2010/07/21 20:58:35] @ Quit: jens_: Read error: Operation timed out | ||
| [2010/07/21 21:00:58] @ WALoeIII joined channel #puppet | ||
| [2010/07/21 21:02:13] @ AimanA is now known as HouseAway | ||
| [2010/07/21 21:02:35] @ RussS joined channel #puppet | ||
| [2010/07/21 21:02:56] @ wilmoore joined channel #puppet | ||
| [2010/07/21 21:04:28] <davea1> jamesturnbull, thanks I finally got it working!!! 5 lines replace 25 lines!!! | ||
| [2010/07/21 21:13:37] @ Quit: OpenMedia: Quit: Leaving. | ||
| [2010/07/21 21:24:26] @ Quit: rcrowley: *.net *.split | ||
| [2010/07/21 21:24:26] @ Quit: davea1: *.net *.split | ||
| [2010/07/21 21:24:26] @ Quit: alban2: *.net *.split | ||
| [2010/07/21 21:24:26] @ Quit: huggie: *.net *.split | ||
| [2010/07/21 21:24:26] @ Quit: Schmidt: *.net *.split | ||
| [2010/07/21 21:24:26] @ Quit: p3rror: *.net *.split | ||
| [2010/07/21 21:24:26] @ Quit: tyler79: *.net *.split | ||
| [2010/07/21 21:24:27] @ Quit: emiddd_: *.net *.split | ||
| [2010/07/21 21:24:27] @ Quit: tmz: *.net *.split | ||
| [2010/07/21 21:24:27] @ Quit: FoXMaN: *.net *.split | ||
| [2010/07/21 21:24:27] @ Quit: cparedes: *.net *.split | ||
| [2010/07/21 21:24:27] @ Quit: crb: *.net *.split | ||
| [2010/07/21 21:24:27] @ Quit: Mandus: *.net *.split | ||
| [2010/07/21 21:24:27] @ Quit: Demosthenes: *.net *.split | ||
| [2010/07/21 21:24:56] @ jaredrhine joined channel #puppet | ||
| [2010/07/21 21:25:46] @ gebi joined channel #puppet | ||
| [2010/07/21 21:29:48] @ rcrowley joined channel #puppet | ||
| [2010/07/21 21:29:48] @ davea1 joined channel #puppet | ||
| [2010/07/21 21:29:48] @ alban2 joined channel #puppet | ||
| [2010/07/21 21:29:48] @ huggie joined channel #puppet | ||
| [2010/07/21 21:29:48] @ Schmidt joined channel #puppet | ||
| [2010/07/21 21:29:48] @ p3rror joined channel #puppet | ||
| [2010/07/21 21:29:48] @ tyler79 joined channel #puppet | ||
| [2010/07/21 21:29:48] @ emiddd_ joined channel #puppet | ||
| [2010/07/21 21:29:48] @ tmz joined channel #puppet | ||
| [2010/07/21 21:29:48] @ FoXMaN joined channel #puppet | ||
| [2010/07/21 21:29:48] @ cparedes joined channel #puppet | ||
| [2010/07/21 21:29:48] @ crb joined channel #puppet | ||
| [2010/07/21 21:29:49] @ Mandus joined channel #puppet | ||
| [2010/07/21 21:29:49] @ Demosthenes joined channel #puppet | ||
| [2010/07/21 21:37:19] @ xetorthio joined channel #puppet | ||
| [2010/07/21 21:37:47] <xetorthio> hi everyone! I'm new to puppet and I wanted to validate with you my use case to see how feasible it is | ||
| [2010/07/21 21:38:45] <xetorthio> we have several teams, each one of them manage a different independent part of a big app | ||
| [2010/07/21 21:39:56] <xetorthio> and every developer from that team needs to be able to reproduce the entire app for himself. I was thinking using a cloud and pupper to give that developer his own dependencies on a cloud, and manage all that with pupper. does it make any sense? | ||
| [2010/07/21 21:40:11] <jamesturnbull> xetorthio: it does | ||
| [2010/07/21 21:40:17] @ Quit: cliff-hm: Ping timeout: 258 seconds | ||
| [2010/07/21 21:40:40] <xetorthio> jamesturnbull: great! any good reading about how to achieve that? | ||
| [2010/07/21 21:40:51] <jamesturnbull> xetorthio: it means if you make sure the puppet code is identical that it doesn't matter which piece of the app each team maintains they get the same infrastructure | ||
| [2010/07/21 21:41:13] <jamesturnbull> xetorthio: you can start with the Docs - http://docs.puppetlabs.com | ||
| [2010/07/21 21:41:45] <jamesturnbull> xetorthio: break your infrastructure into components and put those into modules - you can see examples of those at http://forge.puppetlabs.com | ||
| [2010/07/21 21:43:04] <fenris02> is there a way to run puppet in a slightly more verbose mode? i'd like to run it over the example42 git repo to see what order it loads things in and what gets set, without actually doing any changes. the repo is massive and does strange/duplicate items. | ||
| [2010/07/21 21:43:38] <fenris02> something akin to 'bash -x ./somescript' would be perfect | ||
| [2010/07/21 21:44:14] <xetorthio> jamesturnbull: right.. I was thinking of commiting a puppet script with every related app piece | ||
| [2010/07/21 21:44:55] <jamesturnbull> xetorthio: people do that | ||
| [2010/07/21 21:45:06] <jamesturnbull> xetorthio: you can have a module per piece etc | ||
| [2010/07/21 21:45:20] <jamesturnbull> fenris02: --verbose? | ||
| [2010/07/21 21:45:34] <jamesturnbull> fenris02: or --verbose --debug if you're really keen | ||
| [2010/07/21 21:46:12] <fenris02> --verbose was copious and redundant. unless puppet really does the same check 1000x | ||
| [2010/07/21 21:46:26] <fenris02> i'm afraid of what --debug would entail | ||
| [2010/07/21 21:46:42] <jamesturnbull> fenris02: so you're actually asking for LESS verbose? | ||
| [2010/07/21 21:46:45] <xetorthio> jamesturnbull: sounds great! I will start digging into it | ||
| [2010/07/21 21:46:49] <xetorthio> thanks! | ||
| [2010/07/21 21:46:55] <jamesturnbull> xetorthio: de nada | ||
| [2010/07/21 21:47:08] <fenris02> jamesturnbull, somewhat, or a way to filter it down to a tree structure of what gets called when | ||
| [2010/07/21 21:47:20] <jamesturnbull> fenris02: you can print the graph | ||
| [2010/07/21 21:47:26] <fenris02> o_O? | ||
| [2010/07/21 21:47:42] <fenris02> how do i generate said graph? | ||
| [2010/07/21 21:48:15] <jamesturnbull> fenris02: http://projects.puppetlabs.com/projects/puppet/wiki/Frequently_Asked_Questions#How+do+I+use+Puppet's+graphing+support? | ||
| [2010/07/21 21:48:22] <jamesturnbull> FAQ for the win :) | ||
| [2010/07/21 21:48:30] @ cyrus_mc joined channel #puppet | ||
| [2010/07/21 21:49:20] <jamesturnbull> fenris02: remember thta unless a dependency is explicitly defined then the order will depend on how the graph came together | ||
| [2010/07/21 21:49:25] * jamesturnbull has to run | ||
| [2010/07/21 21:49:40] <fenris02> jamesturnbull, thanks again | ||
| [2010/07/21 21:53:58] <fenris02> mode => '488', | ||
| [2010/07/21 21:54:23] <fenris02> err, how is that possible? (output from "puppetd --genmanifest") | ||
| [2010/07/21 21:56:48] <fenris02> even better. mode => '1005', ... I think i'm confused | ||
| [2010/07/21 21:57:21] @ littleidea joined channel #puppet | ||
| [2010/07/21 21:59:43] @ Quit: WALoeIII: Quit: Bai. | ||
| [2010/07/21 22:03:44] @ Quit: xetorthio: Quit: Get MacIrssi - http://www.sysctl.co.uk/projects/macirssi/ | ||
| [2010/07/21 22:09:46] @ Quit: cyrus_mc: Quit: leaving | ||
| [2010/07/21 22:28:00] @ Quit: littleidea: Quit: littleidea | ||
| [2010/07/21 22:35:14] <ohadlevy> did passenger configuration changed for 2.6? | ||
| [2010/07/21 22:38:54] @ mattock joined channel #puppet | ||
| [2010/07/21 22:39:58] <charlieS> any puppeters in EU looking for a job, let me know :) http://webapps.ubuntu.com/employment/canonical_EURAPR/ | ||
| [2010/07/21 22:40:15] <charlieS> er, this probably isn't the best time of day to spam that. | ||
| [2010/07/21 22:40:30] @ littleidea joined channel #puppet | ||
| [2010/07/21 22:47:31] @ TREllis joined channel #puppet | ||
| [2010/07/21 22:48:40] @ Quit: fredden: Quit: Leaving | ||
| [2010/07/21 22:52:33] @ Quit: littleidea: Quit: littleidea | ||
| [2010/07/21 22:53:52] <sejo> morning all | ||
| [2010/07/21 22:54:01] <scooby2> evening | ||
| [2010/07/21 22:54:10] <sejo> is there a way to use a dict/hash in 0.25.5? | ||
| [2010/07/21 22:54:26] <sejo> I would like to define my nics with all their properties | ||
| [2010/07/21 22:54:34] <sejo> and than per nic run a template | ||
| [2010/07/21 22:56:09] @ ckauhaus joined channel #puppet | ||
| [2010/07/21 22:57:59] @ Quit: TREllis: Ping timeout: 265 seconds | ||
| [2010/07/21 22:59:11] <ohadlevy> sejo: facter doesnt support hashes just yet | ||
| [2010/07/21 23:00:43] <sejo> ohadlevy: hmm I would define them per node, no the only thing I'd be using from facter is the macaddress | ||
| [2010/07/21 23:01:48] <ohadlevy> sejo: 2.6 supports hashes | ||
| [2010/07/21 23:01:55] <sejo> I created a define that I can run conf_nic{$nic, $ipaddr, $gateway, $netmask:} and all our hosts have static IPS so that wouldn't be a problem | ||
| [2010/07/21 23:01:58] <sejo> coon | ||
| [2010/07/21 23:02:05] <sejo> going to remake the rpm's for 2.6 then | ||
| [2010/07/21 23:02:41] <ohadlevy> sejo: there are already rpms for it | ||
| [2010/07/21 23:03:51] <sejo> w00t | ||
| [2010/07/21 23:06:22] @ Quit: ckauhaus: Read error: Operation timed out | ||
| [2010/07/21 23:11:06] @ Djelibeybi joined channel #puppet | ||
| [2010/07/21 23:13:23] @ bgupta joined channel #puppet | ||
| [2010/07/21 23:13:58] @ TREllis joined channel #puppet | ||
| [2010/07/21 23:19:11] @ Quit: TREllis: Ping timeout: 258 seconds | ||
| [2010/07/21 23:22:02] @ Quit: mclarke: Quit: mclarke | ||
| [2010/07/21 23:23:00] @ Quit: davea1: Quit: Leaving | ||
| [2010/07/21 23:23:36] @ ckauhaus joined channel #puppet | ||
| [2010/07/21 23:25:12] <sejo> ohadlevy: in 0.26.0 my defines failed, is there a list of changes and how to resolve deprications? | ||
| [2010/07/21 23:25:23] <sejo> *deprecations | ||
| [2010/07/21 23:25:52] <ohadlevy> there is a changelog on the wiki, and look over the release announcement | ||
| [2010/07/21 23:27:02] <sejo> ok thanks | ||
| [2010/07/21 23:30:36] <sejo> oo pure ruby! | ||
| [2010/07/21 23:30:38] <sejo> w00t | ||
| [2010/07/21 23:35:19] @ Quit: alban2: Ping timeout: 246 seconds | ||
| [2010/07/21 23:39:36] <tim|mac> 0.26.0? | ||
| [2010/07/21 23:39:37] <tim|mac> you mean 2.6.0 right? ;-) | ||
| [2010/07/21 23:39:41] @ ahasenack joined channel #puppet | ||
| [2010/07/21 23:39:56] @ pmorillo joined channel #puppet | ||
| [2010/07/21 23:39:56] @ hMz joined channel #puppet | ||
| [2010/07/21 23:40:53] @ sdog joined channel #puppet | ||
| [2010/07/21 23:42:07] @ Quit: stahnma: Read error: Connection reset by peer | ||
| [2010/07/21 23:42:12] @ Quit: fuser: Read error: Operation timed out | ||
| [2010/07/21 23:42:21] @ stahnma joined channel #puppet | ||
| [2010/07/21 23:42:21] @ Quit: stahnma: Changing host | ||
| [2010/07/21 23:42:21] @ stahnma joined channel #puppet | ||
| [2010/07/21 23:42:24] @ fuser joined channel #puppet | ||
| [2010/07/21 23:42:34] @ Quit: Edgan: Quit: Edgan | ||
| [2010/07/21 23:42:39] <eric0> 10x power | ||
| [2010/07/21 23:42:56] @ Quit: ckauhaus: Ping timeout: 252 seconds | ||
| [2010/07/21 23:43:20] <eric0> i wish the production fires would at least subside to a cheery blaze so i can get back to 2.6 | ||
| [2010/07/21 23:43:43] @ Edgan joined channel #puppet | ||
| [2010/07/21 23:43:54] <ohadlevy> anyon saw this error on 2.6? :/usr/lib/ruby/site_ruby/1.8/puppet/module.rb:79: warning: multiple values for a block parameter (0 for 1) | ||
| [2010/07/21 23:45:39] <sejo> ohadlevy: not yet | ||
| [2010/07/21 23:45:51] <ohadlevy> sejo: not very helpful ;) | ||
| [2010/07/21 23:46:20] @ fbe joined channel #puppet | ||
| [2010/07/21 23:48:57] <sejo> ohadlevy: I know, sorry | ||
| [2010/07/21 23:48:57] @ yannL joined channel #puppet | ||
| [2010/07/21 23:50:29] <sejo> the only thing that I came across till now was that I used to put my defines in a class, and inherited that class in other classes. That doesn't work anymore, i just defined them in the module | ||
| [2010/07/21 23:51:28] <sejo> also puppet-dashboard reports a noop as a fail | ||
| [2010/07/21 23:52:20] @ ckauhaus joined channel #puppet | ||
| [2010/07/21 23:53:49] @ Quit: sdog: Changing host | ||
| [2010/07/21 23:53:49] @ sdog joined channel #puppet | ||
| [2010/07/21 23:53:52] @ pinoyskull joined channel #puppet |
Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!