[00:01:33] Download an ubuntu iso, burn it to disk, reboot, follow install instructions, come back and ask again. [00:01:59] Public SSH Key [00:02:06] Public SSH key info [00:02:10] Public SSH key [00:02:13] triplicate much? [00:05:00] Reedy: heh [00:05:11] OrenBochman: I don't have windows... [00:05:31] OrenBochman: but, putty has an SSH agent (called pagant) [00:05:35] err [00:05:38] pagent [00:05:47] pageant [00:05:54] that too [00:06:07] Add a shortcut in your startup folder to pageant [00:06:17] append to it the path to your private key [00:06:26] then on startup, you'll get a box asking for your key password [00:06:29] enter that, key loaded [00:06:41] it think it just holdes the private keys [00:06:47] I can get into bastion [00:06:49] it's an ssh agent [00:06:51] but no further [00:06:53] SSH forwarding [00:06:55] so [00:06:59] you need to tell putty to forward the agent [00:07:13] Easiest way is to make a saved session [00:07:18] then you only need to do it once [00:07:22] Btw I like how you told puppet to remove irssi off bastion xD [00:07:46] saved session to bastion ? [00:08:00] configure putty, then add it as a saved session [00:08:15] then you can double click to connect with forwarding [00:08:21] setting a proper charset is useful too [00:08:32] ok [00:08:34] meaning, configure it to connect to bastion, and to do forwarding, and likely best to make it do a socks-proxy [00:09:21] bastion works its the second part I don't know how to do ? [00:09:47] Reedy: what charset to set [00:10:10] UTF-8 is better [00:10:21] ok [00:11:02] then SSH, Auth, Allow agent forwarding [00:13:04] Ryan_Lane, you're using openstackmanager-novakeyname but the message doesn't exist [00:13:22] isn't that commented out? [00:13:24] wow... when did quota support get added to glusterfs volumes [00:13:36] not unless you've done it really recently [00:13:59] recently, I think [00:14:58] http://svn.wikimedia.org/viewvc/mediawiki/trunk/extensions/OpenStackManager/special/SpecialNovaKey.php?view=markup#l62 [00:15:15] ah [00:15:17] I see [00:15:31] that's because I've never used nova to actually store the key :) [00:15:59] What would the text be? "Key name" ? [00:16:06] or Nova key name [00:16:22] Reedy: I've st allow agent forwarding [00:16:34] Did you save the session then? [00:16:39] yes [00:16:42] good [00:16:56] Reedy: Key name, yeah [00:20:31] I'm in bastion but when I ssh to labs-mw1.pmtpa.wmflabs I get prompted for a password [00:20:40] am I missing something [00:20:59] Then your clients isn't doing agent forwarding [00:21:40] looks that way [00:24:39] Check the event log [00:24:51] Should have something like "Requesting OpenSSH-style agent forwarding" [00:37:58] I'm guessing you're doing the ors manually cause MWs query wrappers won't? [00:38:37] ors? [00:38:45] Reedy: what's that? [00:38:48] oh [00:38:55] I actually killed that off anyway [00:39:04] I'm making a shit-ton of changes to the puppet group stuff right now [00:39:07] That query? Or the whole thing? [00:39:09] incase you are in there :) [00:39:18] I was just looking at what was indexed [00:39:20] the or thing [00:39:28] my indexes are likely fucked [00:39:29] :) [00:39:35] * Ryan_Lane sucks at database work [00:39:45] you've got the obvious ones at least [00:39:50] plus like the group_name extra one [00:40:06] There's a couple of text fields you're WHERE'ing on with no indexes [00:40:15] But they're hardly gonna be big tables [00:40:15] I'm doing project and global groups right now [00:40:18] Will fix it later [00:40:30] the global groups have group_project as NULL [00:40:46] I can't think of a better way of going about that [00:41:13] but I have a feeling that's kind of dumb, since I then have to do a "where group_project is NULL" [00:42:03] I guess I could restrict the "global" name, so that it can't be used as a project, then I can search based on that string... [00:43:36] What's going on in here? [00:45:19] I'm just tidying up after Ryan ;) [00:45:20] talking about openstackmanager's shitty database code :) [00:45:39] hehe [00:46:03] so null/not null is a bad idea for global/not global, right? :) [00:46:41] Wikitext for SALs suck, I'm going to write a plugin for that https://labsconsole.wikimedia.org/wiki/Server_Admin_Log [00:46:53] sounds good to me [00:47:24] as long as we can have per-project SALs I'll be happy :) [00:48:06] being able to send a log message directly to the api would be sweet [00:48:21] AND IT'LL HAVE BLACKHACJ [00:48:23] AND HOOKERS [00:48:25] Fail [00:48:26] :D [00:48:43] :< [00:48:43] You could use a table for global groups, and one not for it [00:48:47] you could use a boolean field [00:49:27] I really don't want to make another table just for that [00:49:31] boolean field should be easy [00:49:45] yeah [00:49:58] You could put everything in one table [00:50:07] And just say it's for preformance [00:50:29] preformance? performance? bleugh [00:50:33] well, the only difference between global groups and project groups is that one has a project associated with it, and the other does not [00:50:41] otherwise it's exactly the same [00:51:10] and global groups can only be modified by admins [00:55:11] johnduhart, have you seen the whole 7 lines of CSS the extension has? :p [00:55:21] \o/ [00:55:25] and I didn't even write them!!! [00:55:29] heh [00:55:30] haha [00:55:33] I wonder how long before people want API modules for this stuff [00:55:41] I want api modules right now [00:55:49] so that I can do actions via ajax [00:55:50] I think at that point, if it's growing in popularity, we get someone to JS the frontend properly [00:56:08] and make things spin and magically appear [00:56:15] yes. that's what I'd like :) [00:56:41] I also need to paginate everything [00:56:59] it's fucking killing me right now. I'm in every project [00:57:55] We need some more css to make that tavle in the bastion group not at hte top [00:57:57] * Reedy grins [00:58:15] I'd like to collapse the user lists, by default [00:58:52] hell, I wouldn't mind collapsing the projects either [00:59:35] Do we have some jquery libs for that sort of thing? [00:59:42] yeah [00:59:56] I know there's a way to easily do this with jquery [01:00:03] johnduhart, ^ [01:00:11] Uh [01:00:18] I've been adding features rather than doing interface work, lately [01:00:42] the puppet group stuff can definitely use some love. it's functional, but ugly [01:00:46] Collapsible something was added to trunk/1.18 [01:01:10] Ryan_Lane, install CentralNotice and put Nyan cat along the top [01:01:18] https://www.mediawiki.org/wiki/RL/DM#jQuery.makeCollapsible [01:01:20] that [01:01:26] I'd like to collapse everything, except the current project being worked on [01:01:52] well, I guess the current projects being worked on anyway [01:01:53] Hmm. It's for wikitables... [01:02:24] ugh. not even for divs? [01:02:39] I'm just looking at transferring it back from wikitext to html tables [01:02:42] essentially it's the same thing [01:03:10] it's possibly just adding hte divs, and including the right jquery stuff [01:03:10] some places I use tables, and others I use headings with tables underneath it [01:03:21] Yeah the projects section would make WAY more sense if it was headers instead of one big table [01:03:32] (We're talking about this right? https://labsconsole.wikimedia.org/wiki/Special:NovaProject ) [01:03:39] yeah [01:03:42] yeah [01:03:51] and then you couuld collapse the headers [01:04:23] yep, that's what I'd like [01:05:13] setup wise, it should just be $output->addModules( 'jquery.makeCollapsible' ); [01:05:28] https://labsconsole.wikimedia.org/wiki/Special:NovaInstance [01:05:39] the instance list makes a little more sense [01:07:21] It isn't on by default? [01:07:41] oh. it is [01:07:52] the tables in the instance list are collapsible [01:08:29] just need to default to collapsed.. [01:08:41] "Adding '"mw-collapsed'" as additional class will cause the element to be initially collapsed when the page is loaded." [01:08:48] I'd like the whole heading to be [01:09:01] rather than the table [01:10:54] Things like this are more annoying when we know we can't steal someone for an hour or 2 and tell them to prettify it [01:11:23] heh [01:12:19] well, I'm not going to add collapsing without adding some javascript to keep current projects open [01:12:43] It would be incredibly annoying to go to another page, come back and have to reexpand the thing every time [01:13:05] how would MW know what the current project was? [01:13:17] I'd set a cookie when something was expanded [01:13:27] remove it when unexpanded [01:14:08] $res = $dbr->select('revision', array( 'rev_timestamp' ), array('rev_page = '.$article->getID()), $fname = 'Database::select', $options = array( 'ORDER BY' => 'rev_id DESC' , 'LIMIT' => '2' )); [01:14:27] This ain't python buddy [01:14:35] ...? [01:15:00] Reedy: Code from TweetANew [01:15:05] lol [01:15:17] and unescaped stuff [01:15:19] OOH [01:15:22] That's an idea [01:15:32] johnduhart, see if we can get him to add a hook for other actions [01:15:42] ooh [01:15:45] You could have the wiki tweet when new projects were added and stuff [01:15:56] That would be cool [01:16:08] and if coded right wouldn't be much work [01:16:40] If you're doing CR on that line above, 'rev_page' => $article->getID() [01:17:36] Already did [01:17:50] Not really CR just a quick lookover and formatting [01:21:42] Also, article doesn't have a getid method [01:22:39] It doens't fallthrough to another class via __get? [01:24:08] Maybe [01:24:13] We've had big hoohars about magic get methods [01:24:25] the tmhOAuth class can be autoloaded, yeah [01:24:46] Yeah, the function is getId and it's in WikiPage [01:27:57] Fixed [01:30:48] Shouldn't we mark i18n files @ignore for doxygen? [01:33:28] Could do [01:36:53] i should probably attempt to go to bed before it's 2am [01:41:40] Or you could attempt to set a world record for not sleeping! [02:00:44] Ryan_Lane: If you're not busy at some point could you add my account into nova and give it access to bots-cb (mainly just so I can play with your extension but you neve know when it might been a reboot). [02:00:52] * Damianz goes back to finding things to do while rsync bores him [06:27:07] Damianz: into nova? [06:27:15] Damianz: everyone has a nova account :) [06:27:54] and what do you mean by access to bots-cb? [06:28:12] if you'd like to play around with the extension, let's give you access to the openstack project [06:28:19] then you can use nova-production1 [06:28:28] or we can make nova-production2 [16:44:31] hey [16:44:58] Ryan_Lane: D: you rebooted some of my bots? :O [16:45:04] :D [16:45:17] that's why most of them crashed yesterday [16:45:28] or I noticed it yesterday [16:46:40] petan|wk: not purposely [16:46:45] btw you should register labs-home-wm [16:46:56] that's ok :) [16:47:00] when I was enabling live migration one of the host had a problem with upstart [16:47:05] and I had to reboot the host to fix it [16:47:07] I just didn't know why they were down, if it was a bug or what [16:47:24] btw does gluster have lvm support? [16:47:35] but couldn't live migrate the instances, because libvirtd was dead, which means nova-compute was dead [16:47:45] what do you mean? [16:47:50] ah [16:48:13] lvm on top of gluster, or gluster on top of lvm? [16:48:34] I mean if it's possible to extend fs we would have later, using gluster, I don't know much about it I just know you wanted to replace nfs later with it [16:48:52] I don't know how actually gluster works [16:49:06] if it's just a fs or entire disk management system [16:49:23] !enwiki gluster [16:49:24] ah [16:49:29] !enwiki alias wiki [16:49:29] Successfully created [16:49:38] !enwiki gluster [16:49:38] http://en.wikipedia.org/wiki/gluster [16:49:42] it's similar to nfs [16:49:46] but is distributed [16:49:56] so, if a node goes down, things can still be up [16:50:03] aha [16:50:11] so it's exactly something else than I thought [16:50:26] back in a little bit. [16:51:28] I think it would be cool to have nfs for management of the volumes, especially on sql servers, so we can extend them in case we were out of space, dunno if it's possible atm [16:51:35] * lvm not nfs [16:52:35] !monitor bots-1 [16:52:35] http://nagios.wmflabs.org/cgi-bin/nagios3/status.cgi?host=bots-1 [17:29:26] Reedy: btw you pinged me? [17:29:43] I am actually out of town now, but I've seen it in logs [17:29:56] * Reedy shrugs [17:29:58] Maybe [17:29:59] :p [17:30:13] something with extensions [17:30:38] :o [17:30:52] Errm [17:31:10] My short term memory is rather crap atm [17:31:25] It was something you'd done, or not done in both of them [17:31:27] let's see [17:31:58] I think it would be cool to make search interface to logs I wanted to start :D [17:32:23] it's called google ;) [17:32:34] To be honest, I really can't remember [17:32:37] hey that's a good idea [17:35:24] I think that google can't search live content [17:35:41] so it actually won't work I will need to create some php search [18:23:14] petan|wk: I'd prefer the DBs not live in VMs at all [18:46:48] does anyone here have time in the next few days to help Oren Bochman get his labs account set up to his liking? he writes, "Yes I have access to the instance but at this point I can only get into Bastion and not into the lab itself. Anyhow the lab is just a basic wiki and evrything needs to be set up including importing content." [18:46:55] he's working on Lucene and search [18:47:28] des he belong to a project? [18:47:30] *does [18:48:02] I do not know. [18:48:17] I am trying to keep my "on vacation" status this week. I'm happy to forward him to you or vice versa. [18:48:55] yes. he's in a project [18:49:05] he's having an issue with agent forwarding [18:49:59] I'd help him, but I don't have windows, and can't easily walk him through the process [18:50:24] you could poke me [18:50:30] ok, thanks MaxSem [18:50:48] what' [18:51:05] what he needs to set up? [18:51:38] * sumanah forwards to Max [18:52:04] Max, good luck. Back to vacation mode I go. [18:52:07] thanks. [18:54:46] Ryan_Lane: how would we have db then [18:55:14] created via a service [18:55:29] where would it run [18:55:40] on dedicated hardware [18:55:47] ah [18:55:48] dbs in vms are slow [18:55:58] right [18:56:13] replication would be set up automatically as well, which would be nice [18:56:40] OH [18:56:49] petan|wk, now I know [18:57:01] version_compare( $wgVersion, '1.17', '>=' ) ? 'userpage tools' : 'other' [18:57:04] That won't do anything [18:57:07] wait. wtf. [18:57:17] there isn't an instance in the search project [18:57:25] no wonder he can't ssh anywhere else [18:57:28] Even on >= 1.17, there's no special page group for userpage tools etiher [18:57:29] Haha [18:57:30] Ryan_Lane: I think in some case we should not use replication at all [18:57:32] that's the only other project he's in [18:57:47] on bots some dbs are private and should not be made public [18:57:53] Sounds like a big red waving flag [18:58:11] we need a community person :) [18:58:31] I think we planned on eventually hiring one once tool labs is ready [18:58:58] I guess I need to write more documentation [18:59:04] * Ryan_Lane hates writing documentation [18:59:26] well, usage documentation anyway [19:08:21] is there a list or what [19:12:26] a list of what? [19:12:31] documentation that needs to be written? [19:24:23] Vivek: https://labsconsole.wikimedia.org/wiki/Access#Initial_log_in [19:24:30] https://labsconsole.wikimedia.org/wiki/Access#Accessing_public_and_private_instances [19:24:30] ok. [19:25:18] I've added you to the puppet project [19:25:26] you should make your own instance in there, though [19:25:31] please be frugal on resources [19:25:43] but in general tiny instances are mostly useless for real work [19:25:56] t1.small is usually good [19:30:10] ok. [19:31:20] Ryan_Lane: I talked to Reedy my network doesn't work very well [19:31:32] it sent only my last sentence, not all [19:31:46] talk to Reedy about what? [19:31:52] Reedy: what is wrong with that? what can I use [19:32:06] The key doesn't exist in mw core [19:32:07] Ryan_Lane: problem in source [19:32:21] So it'll show as unknown or something [19:32:34] Reedy: ok, is there a list? [19:32:35] of keys [19:33:01] *talked [19:33:49] petan|wk, look for specialpages-group- in MessageEn.php [19:34:01] 12 of them [19:39:25] * Ryan_Lane is confused [19:42:05] 12/30/2011 - 19:42:05 - Creating a home directory for vivek at /export/home/puppet/vivek [19:42:09] 12/30/2011 - 19:42:09 - Creating a home directory for vivek at /export/home/bastion/vivek [19:43:05] 12/30/2011 - 19:43:05 - Updating keys for vivek [19:43:09] 12/30/2011 - 19:43:08 - Updating keys for vivek [19:44:51] Reedy: https://i-0000007b.pmtpa.wmflabs/wiki/Special:NovaPuppetGroup [19:45:39] you'll have to have a socks proxy, of course :) [19:45:48] oh. I guess you'd need an account on that too [19:45:50] * Ryan_Lane groans [19:45:52] nevermind :) [19:45:59] haha [19:46:05] you'll just see the changes when they go live, then [19:46:41] I made the sections collapsible, and have been making all of the list special pages more consistent [19:59:15] Vivek: there isn't a *ton* of docs, but there are some... [19:59:19] !instances [19:59:23] !security [19:59:23] https://labsconsole.wikimedia.org/wiki/Security_Groups [19:59:28] !instance [19:59:28] https://labsconsole.wikimedia.org/wiki/Instances [19:59:32] ok [19:59:37] !instances alias instance [19:59:38] Successfully created [20:00:24] I guess yuvipanda isn't online this time [20:01:02] Ryan_Lane: we talked about some mediawiki stuff totally off-topic here [20:01:09] :) [20:01:10] ah. heh [20:01:24] I do that often too :) [20:09:09] Vivek: when you make any change to a system as root, you should log it to your project's server admin log [20:09:12] !logging [20:09:15] !log [20:09:25] ok [20:09:31] !logging is !log [20:09:31] Key was added! [20:09:37] !logging [20:09:37] !log [20:09:38] is not a valid project. [20:09:43] :D [20:09:43] lol [20:09:45] damn it [20:10:08] !logging is To log a message, use the following format: !log [20:10:08] Key exist! [20:10:13] !logging del [20:10:13] Successfully removed logging [20:10:16] !logging is To log a message, use the following format: !log [20:10:16] Key was added! [20:10:18] !logging [20:10:18] To log a message, use the following format: !log [20:10:22] better :) [20:10:26] @search log [20:10:26] Results: morebots, labs-morebots, credentials, logging, [20:10:37] !labs-morebots [20:10:37] adminbot: http://svn.wikimedia.org/viewvc/mediawiki/trunk/tools/adminlogbot/ [20:12:15] ok [22:13:15] Ryan_Lane: Are you around ? [22:13:32] yes. just ask your question though. no need to ask if I'm around [22:13:36] if I am I'll answer [22:14:29] !ask is Hi, how can we help you? Just ask your question. [22:14:29] Key was added! [22:14:37] I can help out with OpenStack,Nagios,tor,puppet (already added by you), juju [22:14:54] I am now working with Orchestra at my workplace. [22:15:00] juju will be up next. [22:15:09] so, work on one thing at a time [22:15:19] We have written the Open Stack Begineers guide. [22:15:20] ok. [22:15:53] Can I also join Nagios now as I am working on it as part of orchestra ? [22:15:59] Anyways upto you :) [22:16:08] it isn't orchestra, though [22:16:14] I know [22:16:25] Orchestra is still a suite of applications. [22:16:25] unless you are going to work on something specific, there isn't a reason to be added to a project [22:16:32] ok. [22:22:42] Ryan_Lane: ssh @bastion.wmflabs.org -L 8080:labs-mw1:80 [22:22:57] also an example [22:23:02] of port forwarding [22:23:17] I recommend a socks-proxy, rather than port forwarding, though [22:23:19] !socks-proxy [22:23:19] see https://labsconsole.wikimedia.org/wiki/Access#Accessing_public_and_private_instances ssh @bastion.wmflabs.org -D [22:23:29] I completed this step but http://localhost:8080 does not grant me access. [22:23:31] ok [22:23:40] because there's nothing running on labs-mw1.... [22:23:46] ok [22:30:28] Ryan_Lane: I am not in the sudoers file on bastion1 [22:30:33] I know [22:30:38] no one is [22:30:41] it's a bastion host [22:30:42] ok [22:30:44] that would be insecure [22:31:01] So in my case how do I access the puppet host ? [22:31:04] if you create an instance in the puppet project, you'll have i tthere [22:31:31] create an instance, wait for it to finish building (when the puppet run finished - you can see this in the instance's console output), then you can ssh into it [22:31:40] your sudo password is your wiki password [22:31:44] @search sudo [22:31:44] No results found! :| [22:32:29] !sudo You have sudo in any project that you are a member of, excluding global projects (like bastion). Your sudo password is your labsconsole wiki password. [22:32:37] !sudo is You have sudo in any project that you are a member of, excluding global projects (like bastion). Your sudo password is your labsconsole wiki password. [22:32:37] Key was added! [22:32:41] !sudo [22:32:41] You have sudo in any project that you are a member of, excluding global projects (like bastion). Your sudo password is your labsconsole wiki password. [22:34:33] You want me to use the m1.tiny right ? [22:35:52] m1.small is likely better [22:36:06] ok [22:36:06] tiny instances can be useful for *some* things, but are often worthless [22:36:14] ok. [22:36:31] Do you suggest that I create my own instance or use an existing one ? [22:37:11] create your own [22:37:18] someone is likely doing something on the others [22:37:22] Doing that now [22:41:01] Is there any instance naming convention ? [22:41:15] Going for a m1.small with Oneric [22:44:57] it's likely best to use lucid [22:45:05] oneric is there for testing, mostly [22:45:26] we use LTS for basically everything, though [22:45:35] Ryan_Lane: I use Oneric at work. [22:45:38] no specific naming convention, just try to use a name that will be unique [22:45:42] we use lucid here ;) [22:46:02] I am naming it vivek-puppet [22:50:34] Ryan_Lane: When I am about to create a puppet instance do I need the only the pupper option ? [22:50:49] you don't need any options [22:50:56] ok. [22:50:59] and as the warning implies, you shouldn't set them on creation [22:51:05] Got it. [22:52:18] Created instance i-000000c7 with image ami-00000023 and hostname i-000000c7.pmtpa.wmflabs. [22:53:20] Used a m1.small oneric, oneric should not be an issue :) [22:53:57] puppet builds the system, and it doesn't run properly in oneric [22:54:16] some of the manifests break because we don't have an apt repo set up for oneric [22:54:45] and our apt repo contains packages that the manifests are looking for [22:54:50] ok Then let me go back to deleting the instance and creating it in Lucid. [22:55:13] anything other than LTS releases we only ever use for testing [22:55:22] ok. [22:57:46] Created instance i-000000c8 with image ami-0000001d and hostname i-000000c8.pmtpa.wmflabs. [22:58:22] * Damianz debats if Vivek is a bot [22:58:35] lol [22:59:06] Damianz: Welcome come to the Open Source way of System Administration :) [23:01:21] Ryan_Lane: Now how do I access the instance ? [23:01:31] !access [23:01:34] vivek@bastion1:~$ ssh vivek@10.4.0.11 [23:01:34] Permission denied (publickey). [23:01:35] hmm [23:01:53] first, use the instance name, it's easier :) [23:02:01] second, did you forward your key? [23:02:11] third, did the instance fully build? [23:02:45] ugh [23:02:48] well, you hit a bug [23:02:55] you'll need to delete and recreate the instance [23:03:02] if you look in the console output you'll see it [23:03:07] Failed to fetch http://ubuntu.wikimedia.org/ubuntu/pool/main/r/ruby1.8/ruby1.8_1.8.7.249-2_amd64.deb Hash Sum mismatch [23:03:08] E: Unable to fetch some archives, maybe run apt-get update or try with --fix-missing? [23:03:16] we have some issue with that package in our repo [23:03:23] occasionally it causes this bug [23:03:26] ok. [23:03:35] usually deleting and recreating will help [23:03:49] Deleting and recreating it. [23:03:53] PROBLEM Current Load is now: CRITICAL on vivek-puppet vivek-puppet output: Connection refused by host [23:04:38] PROBLEM Current Users is now: CRITICAL on vivek-puppet vivek-puppet output: Connection refused or timed out [23:07:07] Ryan_Lane: The recreation seems to be done ok, logs show no errors. [23:08:18] The key forwarding has not been done... [23:08:53] PROBLEM host: vivek-puppet is DOWN address: vivek-puppet CRITICAL - Host Unreachable (vivek-puppet) [23:10:19] Ryan_Lane: Need help to proceed with the key forwarding part. [23:11:13] Ryan_Lane: Do I need to import a new key from the bastion to the instance ? [23:13:52] PROBLEM Current Load is now: CRITICAL on vivek-pupper vivek-pupper output: CHECK_NRPE: Error - Could not complete SSL handshake. [23:13:53] Ryan_Lane: I am in vivek-puppet [23:13:54] :) [23:14:32] PROBLEM Current Users is now: CRITICAL on vivek-pupper vivek-pupper output: CHECK_NRPE: Error - Could not complete SSL handshake. [23:14:57] brb [23:15:12] PROBLEM Disk Space is now: CRITICAL on vivek-pupper vivek-pupper output: CHECK_NRPE: Error - Could not complete SSL handshake. [23:16:12] PROBLEM Free ram is now: CRITICAL on vivek-pupper vivek-pupper output: CHECK_NRPE: Socket timeout after 10 seconds. [23:16:40] Vivek: I don't see any instance in the list that you created [23:16:54] I just deleted it once more now. [23:16:58] why? [23:17:01] Recreating it now. [23:17:10] you should check the console logs..... [23:17:24] "get console output" action [23:17:32] PROBLEM Total Processes is now: CRITICAL on vivek-pupper vivek-pupper output: Connection refused or timed out [23:17:35] you can refresh it to see the progress [23:18:12] PROBLEM dpkg-check is now: CRITICAL on vivek-pupper vivek-pupper output: Connection refused or timed out [23:19:31] I did that. [23:21:15] this one is building properly [23:21:17] don't delete it [23:21:29] since you deleted it, the DNS is cached for the wrong IP [23:21:35] gimme a sec [23:22:28] ok, now whenever it finished building, it should work [23:22:51] "notice: Finished catalog run in 140.19 seconds" [23:22:56] ok [23:23:03] vivek@i-000000ca.pmtpa.wmflabs's password [23:23:30] It permitted me without the password earlier. [23:23:49] again, you should use vivek-puppet, rather than i-00000ca [23:23:52] PROBLEM Current Load is now: CRITICAL on vivek-puppet vivek-puppet output: CHECK_NRPE: Error - Could not complete SSL handshake. [23:23:57] either way, it should work [23:24:05] ok [23:24:32] PROBLEM Current Users is now: CRITICAL on vivek-puppet vivek-puppet output: CHECK_NRPE: Error - Could not complete SSL handshake. [23:24:36] It works now. [23:24:44] it accepted your password? [23:24:46] I need to fix that [23:24:54] you should use your key, not a password [23:25:12] PROBLEM Disk Space is now: CRITICAL on vivek-puppet vivek-puppet output: CHECK_NRPE: Error - Could not complete SSL handshake. [23:25:16] your agent is likely not forwarded properly [23:25:27] Yep, it asked for password. [23:25:47] From the bastion to vivek-puppet [23:26:02] PROBLEM Free ram is now: CRITICAL on vivek-puppet vivek-puppet output: CHECK_NRPE: Error - Could not complete SSL handshake. [23:27:02] And it accepted the password. [23:27:22] PROBLEM Total Processes is now: CRITICAL on vivek-puppet vivek-puppet output: CHECK_NRPE: Error - Could not complete SSL handshake. [23:28:12] PROBLEM dpkg-check is now: CRITICAL on vivek-puppet vivek-puppet output: CHECK_NRPE: Error - Could not complete SSL handshake. [23:52:22] RECOVERY Total Processes is now: OK on vivek-puppet vivek-puppet output: PROCS OK: 78 processes [23:53:12] RECOVERY dpkg-check is now: OK on vivek-puppet vivek-puppet output: All packages OK [23:53:52] RECOVERY Current Load is now: OK on vivek-puppet vivek-puppet output: OK - load average: 0.02, 0.03, 0.07 [23:54:32] RECOVERY Current Users is now: OK on vivek-puppet vivek-puppet output: USERS OK - 0 users currently logged in [23:55:12] RECOVERY Disk Space is now: OK on vivek-puppet vivek-puppet output: DISK OK [23:56:02] RECOVERY Free ram is now: OK on vivek-puppet vivek-puppet output: OK: 91% free memory [23:57:41] Ryan_Lane: Did you fix the key exchange issue ? [23:57:57] that has nothing to do with me [23:58:10] I meant you being able to log in with a password is something I'm going to fix