[00:52:51] !log tools.quentinv57-tools Added some ifset($_COOKIE['']) calls to deal with the undefined notices appearing on sulinfo.php [00:52:53] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.quentinv57-tools/SAL [00:53:06] !log tools.quentinv57-tools (that was isset, not ifset) [00:53:08] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.quentinv57-tools/SAL [01:25:54] 10Tool-Labs-tools-Quentinv57's-tools: Local Stewards Log's hyperlink that links to an account (who's userrights are changed) is broken - https://phabricator.wikimedia.org/T67226#2808604 (10Krenair) 05Open>03Resolved a:03Krenair fixed [01:26:08] !log tools.quentinv57-tools also T67226 [01:26:11] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.quentinv57-tools/SAL [01:26:11] T67226: Local Stewards Log's hyperlink that links to an account (who's userrights are changed) is broken - https://phabricator.wikimedia.org/T67226 [01:30:25] 10Tool-Labs-tools-Quentinv57's-tools: sulinfo.php references several non-existent external resources - https://phabricator.wikimedia.org/T151137#2808609 (10Krenair) [01:35:34] 10Tool-Labs-tools-Quentinv57's-tools: SUL Info results incorrect for specific user - https://phabricator.wikimedia.org/T85642#951483 (10Krenair) This data comes from user.user_editcount in the wiki's database: ```MariaDB [simplewiki_p]> select user_id, user_editcount from user where user_name = 'Jan.Kamenicek';... [01:48:59] 10Tool-Labs-tools-Quentinv57's-tools, 07I18n: SUL info tool fails with unicode usernames - https://phabricator.wikimedia.org/T67144#696002 (10Krenair) According to the header at the top of sulinfo.php it's GPLv3 so probably. @Cyberpower678? [03:01:20] 10Tool-Labs-tools-Quentinv57's-tools, 07I18n: SUL info tool fails with unicode usernames - https://phabricator.wikimedia.org/T67144#2808646 (10Cyberpower678) I would think the centralauth tool does a good job now these days. Is there anything with this tool you still prefer? [05:19:15] 10Tool-Labs-tools-Xtools, 07I18n: xTools not translating some strings - https://phabricator.wikimedia.org/T150931#2808707 (10Matthewrbowker) >>! In T150931#2808303, @MusikAnimal wrote: > Most likely unrelated, there are issues with the Top Pages tool now: http://tools.wmflabs.org/xtools/pages/ Appears to work... [06:34:03] PROBLEM - ToolLabs Home Page on toollabs is CRITICAL: HTTP CRITICAL: HTTP/1.1 503 Service Not Available - string 'Magnus' not found on 'http://tools.wmflabs.org:80/' - 531 bytes in 0.031 second response time [06:41:29] looking [06:41:53] The last Puppet run was at Wed Nov 16 16:53:40 UTC 2016 (5148 minutes ago). Puppet is disabled. trail run for changeset 321786 handling /var/lib/gridengine [06:41:54] * Krenair grumbles [06:46:11] Uh... is someone messing with the webservice? [06:46:37] Or am I just skilled enough to break it in ways that weren't expected. [06:46:44] https://www.irccloud.com/pastebin/w6AfLja7/ [06:48:26] I just 'messed' with it to get it back up [06:48:31] will log in a sec but here's what I did [06:48:40] ssh'd to tools.wmflabs.org (tools-proxy-01) [06:48:45] redis-cli [06:48:49] hgetall prefix:admin [06:48:56] shows it's running at http://10.68.20.245:52990 [06:49:02] turns out this is tools-webgrid-lighttpd-1413.tools.eqiad.wmflabs. [06:49:55] then I SSH'd to login.tools.wmflabs.org (tools-bastion-03), became tools.admin [06:49:57] ran qmod -rj lighttpd-admin [06:50:31] hmm [06:50:36] it appears down again? [06:50:38] Solid 503s now. [06:50:49] interesting explanation though :) [06:51:02] yeah now the job is just gone from qstat [06:51:49] !log tools ran `qmod -rj lighttpd-admin` as tools.admin to try to get the main page back up, it worked briefly but then broke again [06:51:53] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools/SAL [06:58:48] how are you supposed to start the damn thing [07:01:55] Hi all [07:06:55] hi Oshwah [07:07:17] Matthew_, did you notice this yourself or from the messages in IRC? [07:08:13] Oshwah, are you here about the main page too? [07:08:33] Nope just saw activity here, thought I'd say hi. [07:08:37] What's happening with the main page? [07:09:07] not sure yet, https://tools.wmflabs.org/ is down (but not the actual tools under that) [07:13:24] pfft what? [07:13:27] it just randomly started? [07:14:00] RECOVERY - ToolLabs Home Page on toollabs is OK: HTTP OK: HTTP/1.1 200 OK - 3670 bytes in 0.030 second response time [07:15:32] * Oshwah shrugs [07:15:33] Computers [07:16:12] Krenair: Actually, my monitoring software caught it. [07:16:19] Rekt [07:16:35] I get a text when xTools goes down, I use UptimeRobot. [07:16:41] xTools was down? [07:16:52] Yes. Alternating 503 and 500 [07:16:56] It has recovered. [07:17:05] Paste is down right now. [07:17:10] you really should've told me [07:17:18] Sorry... [07:17:22] I was assuming only the main page was broken [07:17:29] I thought everything was down, and I didn't want to get in your way. [07:17:51] if everything was down I would've contacted ops [07:18:14] My apologies. I guess we had a miss-communication. [07:18:38] My original link was to an error message when I tried to restart the webservice in xtools. I neglected to mention that facat. [07:18:40] *fact [07:19:53] what was the error message? [07:20:19] https://www.irccloud.com/pastebin/w6AfLja7/ was what I received when I ran "webservice restart" [07:20:35] This is for the base xtools [07:21:18] alright, I'll include that in my labs-admin mail [07:21:31] Okay. Is there anything else you need from me? [07:21:39] Paste just went back up btw [07:21:54] I need to drive home and it'll take a half hour, I can be online after that. [07:23:09] Nope [07:29:15] is xtools-ec down [07:30:43] shows nothing for my count [09:29:24] bd808: can you clarify what's the scope of "tool" on https://meta.wikimedia.org/wiki/Requests_for_comment/Abandoned_Labs_tools#Call_for_vote_on_straw_dog_proposals ? [09:30:03] i.e. what about eg. quarry? [09:32:14] how would quarry be unclear? [09:32:51] it's not part of tools is it? [09:33:11] https://quarry.wmflabs.org/ [09:33:14] it's not in tool labs. I don't see any mention of whether it's only applied to tool labs or to labs globally [09:33:21] exactly [09:34:18] at least, it's not explicit [09:34:32] I don't see anything that suggests it would apply outside of tools [09:34:43] hmm [09:35:45] and what about the more complex half-tool-labs half-outside tools such as video2commons of mine? [09:36:38] if adoption is needed, the new maintainer needs access to the video project as well [09:37:06] (I'm fine with that, but it's not stated how such should be handled) [09:37:57] I would assume only the tools half is covered and access to the video project would have to dealt with by the labs admins [09:38:13] but might be worth asking for clarification [09:42:57] 10Quarry: Add an option to export result in Wikilist - https://phabricator.wikimedia.org/T137268#2808805 (10Dvorapa) >>! In T137268#2808050, @XXN wrote: > Should there be 2 options, for each type of list: bulleted and numbered? One would be enough I think >>! In T137268#2808050, @XXN wrote: > As an alternative... [09:54:34] 10Quarry: Quarry exports integers as floats to wikitable - https://phabricator.wikimedia.org/T151106#2807913 (10zhuyifei1999) This isn't specific to wikitable. I just checked CSV and it's float as well. [10:05:33] Labs replag is quite high right now [11:11:02] PROBLEM - SSH on tools-webgrid-lighttpd-1415 is CRITICAL: CRITICAL - Socket timeout after 10 seconds [11:15:53] RECOVERY - SSH on tools-webgrid-lighttpd-1415 is OK: SSH OK - OpenSSH_6.6.1p1 Ubuntu-2ubuntu2.8 (protocol 2.0) [12:24:24] PROBLEM - Host tools-secgroup-test-102 is DOWN: CRITICAL - Host Unreachable (10.68.21.170) [13:12:45] PROBLEM - Host tools-secgroup-test-103 is DOWN: CRITICAL - Host Unreachable (10.68.21.22) [14:28:47] PROBLEM - Host secgroup-lag-102 is DOWN: CRITICAL - Host Unreachable (10.68.17.218) [15:17:22] PROBLEM - Puppet run on tools-bastion-03 is CRITICAL: CRITICAL: 30.00% of data above the critical threshold [0.0] [15:18:14] PROBLEM - Puppet run on tools-webgrid-lighttpd-1206 is CRITICAL: CRITICAL: 55.56% of data above the critical threshold [0.0] [15:21:23] Why a nodejs program returns "ERROR ... V8 is longer usable" when using jsub? [15:32:04] Seija: make sure to allow the task to use enough memory [15:32:06] -mem parameter [15:32:21] I think it's 256M by default; try e.g. 512MB [15:32:47] But just a helloworld... [15:33:15] complain to the nodejs developers ;-) [15:34:49] I changed the mem to 768m and it worked. thanks. [15:48:13] RECOVERY - Puppet run on tools-webgrid-lighttpd-1206 is OK: OK: Less than 1.00% above the threshold [0.0] [15:57:22] RECOVERY - Puppet run on tools-bastion-03 is OK: OK: Less than 1.00% above the threshold [0.0] [16:25:25] Hi all, I'm still getting errors like this at http://pastebin.com/B9rqrT8g when logging in using OAuth from the grid. When I run the same command under the same account but without the grid (at tools-bastion-02), all works as expected (http://pastebin.com/wXHn4acY). [16:25:33] Can anybody help me and stop me getting this errors? [16:25:38] Thanks in advance! [16:32:02] Urbanecm: "ERROR: OAuth authentication not supported: No module named requests_oauthlib" [16:34:30] valhallasw`cloud, I can see this error. But why it appear only when using the grid? [16:34:35] And how to fix it? [16:39:53] Urbanecm: install requests_oauthlib in a virtualenv [16:40:16] I have no clue why it is installed on tools-bastion, it shouldn't be [16:41:12] or should be installed on both [16:41:38] but it might be only for trusty and higher? [16:41:54] not sure why you're trying to run it on a precise host [16:42:18] seems to work on trusty hosts. [16:47:06] valhallasw`cloud, I run it on the default. I didn't want to run it at trusty. I'll look at it, thanks! [16:48:20] *at precise [16:52:41] PROBLEM - Free space - all mounts on tools-docker-registry-01 is CRITICAL: CRITICAL: tools.tools-docker-registry-01.diskspace.root.byte_percentfree (<100.00%) [17:01:16] 10Tool-Labs-tools-Quentinv57's-tools, 07I18n: SUL info tool fails with unicode usernames - https://phabricator.wikimedia.org/T67144#2809199 (10Huji) @Cyberpower678 you are right, Special:CenteralAuth now provides all of the information this tool provides. In fact, it might make sense to sundown this tool altog... [17:10:57] 10Quarry: Query Quarry's own database - https://phabricator.wikimedia.org/T151158#2809210 (10intracer) [17:34:45] Beta cluster seems to be down [17:35:14] https://de.wikipedia.beta.wmflabs.org/wiki/ returns MediaWiki internal error. [17:35:15] Exception caught inside exception handler. [17:35:25] this is -releng [17:35:33] releng maintains that i think [17:38:22] paladox: you've meant I shall state that issue at #wikimedia-releng, am I right? [17:38:31] Yep [17:38:33] please [17:38:40] PROBLEM - Puppet run on tools-services-01 is CRITICAL: CRITICAL: 100.00% of data above the critical threshold [0.0] [17:38:43] I belive they maintain that. [17:40:47] Obviously Sagan was faster then me ;) [17:41:42] :) [17:41:45] yep [17:42:19] :D [19:06:22] zhuyifei1999_: The policies up for vote are scoped to Tool Labs which I would define as the "tools" project that is hosted in Wikimedia Labs and the software and services it supplies. [19:07:20] There was a suggestion in the prior discussion period by legoktm that the right to fork policy be extended to all of Wikimedia Labs, but I asked that we focus on Tool Labs first before attempting to expand scope. [19:07:39] makes sense [19:08:26] Do you think this warrants some clarification on either the polices themselves or the RfC page? [19:10:17] it would be better if it's said that "This policy applies to Tool Labs only" somewhere explicitly. many instances are using "tool", and that word is quite broad [19:11:03] and how do you think those cross-project tools should be handled? [19:12:34] zhuyifei1999_: I think that cross project hybrids are a topic for the enforcement committee to discuss. :) [19:12:48] If we aim for perfect we will stay in the world of no policy [19:12:56] um sure [19:14:00] world of no policy... utopia? [19:14:53] as far as I know there are only a very small number of hybrid tools. paws and video2commons are actually the only ones that come immediately to mind [19:15:34] paws may have actually all moved into tools too. I'm not entirely shure [19:16:02] * zhuyifei1999_ even failed to remember paws [19:17:52] the reasonable initial policy would be that the right to fork and abandoned tool policies are strictly scoped to the software and services hosted in the tools project. [19:18:43] there are other "hybrid" tools that orchestrate things outside of Tool Labs & Labs and ti would be difficult to draw the line otherwise [19:21:01] an example of that second type of hybrid would be the tool that sends images to google's image api for text conversion [19:21:26] hmm [19:22:18] I would think google's image api is some external service. the only thing secret would be the api key [19:22:38] (but that's a critical part as well) [19:22:56] and the software in use on Google's servers. that is not FLOSS [19:24:26] If we required all software used via the tools to be FLOSS, forkable and usurpable then we would be implicitly disallowing tools that interact with non-FLOSS services hosted outside of Labs [19:24:58] well, I don't think our ISPs use 100% FLOSS, but we use them anyways, don't we? [19:25:43] I would certainly *encourage* the video2commons project and similar hybrids to publish their code under an OSI-approved license. The terms of service for all of Labs currently require that. [19:25:57] it is on github [19:26:18] with gpl [19:27:00] *nod* so that particular case is handled [19:27:30] I'm the licensing thing isn't really new, but the adoption is [19:27:45] *~~I'm~~ [19:27:47] I'm fairly confident that the number of purposefully non-FLOSS tools is very small [19:28:13] but we have not done much work on verification and enforcement in the past [19:28:52] I can control what my tools are licensed under, and fix it (almost) forever, but adoption, I have no say [19:28:52] In part because there has never been a clear list of expectations nor an assigned enforcement body [19:30:02] judging by the number of tickets for tool takeover that have been filed in phabricator it will not be a common activity [19:30:20] hmm sure [19:30:40] to be worth the effort of takeover a tool would need to be fairly popular, badly broken, and without an active maintainers [19:31:13] oh rotatebot on commons [19:31:51] in that case it was forked [19:32:03] *nod* which is a much easier path [19:32:17] as long as the source can be found and shown to be licensed [19:32:46] making a wiki account is trivial and even getting a bot flag is not too hard [19:34:11] I'll look at the policies and see if I can reasonably add a "scope" section to both [19:34:43] k [19:43:41] zhuyifei1999_: https://wikitech.wikimedia.org/w/index.php?title=Help:Tool_Labs/Right_to_fork_policy&diff=998307&oldid=941464 [19:44:59] cool :) much clearer now [19:45:25] I made the same edit to the abandoned tool policy as well [19:45:35] thanks for asking the question [19:45:59] np [19:46:36] (why do I find np a bit weird here)