Skip to main content

Browse the archive

https://github.com/wikimedia/operations-puppet
12 April 2024, 10:39:33 UTC
sort by:
Revision Author Date Message Commit Date
9d2dc25 juniorsys 26 January 2017, 09:58:08 UTC ores/otrs/package_builder: Linting changes Use full names for class names, as relative names are not allowed in future Puppet versions Add trailing commas to abide by the Coding Style guidelines Bug: T93645 Change-Id: I7720660c061588c30a4daed22e67966da1493801 02 February 2017, 23:49:45 UTC
d7c6e25 James D. Forrester 02 February 2017, 22:04:18 UTC admin: Re-apply jforrester ssh key This reverts commit 2b87787472fde6f41a4561cbf45423b49bb600ce. Change-Id: I1e8adffb295998c3392fca156a60c60c28d42fb9 02 February 2017, 22:33:37 UTC
3f6acd8 andrewbogott 02 February 2017, 02:10:31 UTC bootstrap_vz and vmbuilder: Do apt-get update before first puppet run I'm not sure if this is necessary, but our first puppet run is complaining about not being able to find packages, even though they look like they should be available from the outset. Change-Id: I5055ea55e243d5f8c75e7d67740fca14438f712f 02 February 2017, 21:13:03 UTC
48cdf92 andrewbogott 02 February 2017, 02:39:49 UTC vmbuilder and boostrap_vz: Include many more packages in base image These are things that get installed on the first puppet run already... adding these to the base image should speed up the initial puppet run a fair bit. Change-Id: I2d3e2124726caa80acdd12e3d666561a3f6d9e4f 02 February 2017, 21:10:37 UTC
02cd949 andrewbogott 26 January 2017, 06:39:14 UTC New labs instances: Fail more obviously if we have DNS issues Change-Id: I06efaf3a75b108ab29de8783f3abbd9272943c6f 02 February 2017, 21:04:29 UTC
984aed7 andrewbogott 02 February 2017, 20:54:52 UTC vmbuilder: Hack around a bug with sudoers This is a temporary fix for https://bugs.launchpad.net/ubuntu/+source/vm-builder/+bug/1618899 which probably only works on Trusty. This can be reverted if the python-vm-builder upstream ever springs back to life. Change-Id: Ia060daf8a8252bd3480a3623bcedb40fa6974f22 02 February 2017, 20:57:12 UTC
50f837f Andrew Otto 02 February 2017, 19:29:07 UTC Update cdh module with zookeeper package conditional typo fix Change-Id: If9b41dbc7599e306f89204a84ff7d82c735fd410 02 February 2017, 19:29:07 UTC
378ed06 Amir Sarabadani 29 January 2017, 00:34:00 UTC dumps: Modernize design of the index page Bug: T155697 Change-Id: I77abcc300a0453cde1be79c674e9934c610b0d51 02 February 2017, 18:05:40 UTC
125f069 Tyler Cipriani 01 February 2017, 22:09:50 UTC logstash_checker: provide an absolute threshold Starting to see some false positives when using the logstash_checker for canary checks in MediaWiki deploys. This is mainly due to having a low starting error rate. When the average 10s error rate before deployment is 0.001 then seeing an average error rate that is 10 times (or even 100 times) that average error rate is not a large absolute number of errors. This sets a threshold for the absolute average 10s error rate we care about. If the error rate after deployment is higher than the larger of the absolute threshold or the pre-deploy 10s average error-rate multiplied by the fail-threshold -- then we exit non-zero. Change-Id: I1bfc88b829b63363ca0e2f373fc99ef6bff81cee 02 February 2017, 17:15:53 UTC
0f0cb77 elukey 02 February 2017, 14:43:56 UTC Replace Redis/Memcached shards mc200[4567] with mc202[2345] Bug: T155755 Change-Id: I8d6e1ff2025b1b5ddf8f4e9200cc75b0a611c467 02 February 2017, 14:43:56 UTC
71607cf Elukey 02 February 2017, 13:46:04 UTC Revert "Add JVM Heap usage alarms for basic Hadoop daemons" This reverts commit 4735b0f002a7ac5269186fd8399231cdcc5c2253. Change-Id: I0843e88286cdcb3d7ff41fc844c93920ac1f2fb0 02 February 2017, 13:46:04 UTC
4735b0f elukey 02 January 2017, 15:45:44 UTC Add JVM Heap usage alarms for basic Hadoop daemons Bug: T88640 Change-Id: I6d47880bd69106268eb0ff03bd9ee2293d59049c 02 February 2017, 12:55:42 UTC
04dff30 Guillaume Lederrey 23 November 2016, 14:20:54 UTC elasticsearch - reimage elasticsearch relforge servers to jessie Bug: T151326 Change-Id: I263e55bcc3e93e1758af5ef22c11bac5fd5ece40 02 February 2017, 12:30:06 UTC
ed8f0b8 elukey 02 February 2017, 11:27:24 UTC Replace mc200[23] Redis/Memcached codfw shards with mc202[01] Bug: T155755 Change-Id: Ie37006746f88dbd37ba4509e458ddbe13e55852c 02 February 2017, 11:27:24 UTC
a00ad83 Filippo Giunchedi 02 February 2017, 11:11:05 UTC conftool-data: fix typo defaul_values vs default_values Change-Id: I5ca4b06c5602ec41d149ca0c051776e82932a583 02 February 2017, 11:11:05 UTC
404458e elukey 01 February 2017, 14:47:08 UTC Replace mc2001 with mc2019 in Mediawiki Redis/Memcached shards Bug: T155755 Change-Id: I39537aa3a35d257bf17d08b87bf8a415231d1cc0 02 February 2017, 10:19:55 UTC
3d489f6 Moritz Muehlenhoff 01 February 2017, 13:46:01 UTC Remove ntfs-3g on precise/trusty Installed by default on Ubuntu, but not used (and it's setuid root, so a potential security risk). Limited to Ubuntu, since Debian doesn't pull it in by default and at least on labnodepool1001 we have a real world use case since libguestfs0 pulls in ntfs-3g. Change-Id: I42169c2e3c419173c482525fdd3546080274d08f 02 February 2017, 08:34:49 UTC
046a7b8 Eric Evans 17 January 2017, 18:18:07 UTC Enable Prometheus JMX exporter on Cassandra nodes This uses a boolean value (`jmx_exporter_enabled`) to selectively enable the exporter on a host-by-host basis. Long-term this does not make sense, and so when we're comfortable with the stability of the agent, this should be removed and the exporter deployed to all Cassandra nodes. Bug: T155120 Change-Id: I8b70ada1ede13338f36b90ce704e8bbdfb468889 02 February 2017, 08:08:45 UTC
5c628e7 Moritz Muehlenhoff 30 January 2017, 12:12:39 UTC Remove otto from aqs-admins I've written a script to run consistency checks on group memberships in the WMF cluster. Remove otto from aqs-admins since he has cluster- wide root access anyway. Bug: T142836 Change-Id: I3d14ac4f7f247b5cfa9e7a9a0885247cd68a082d 02 February 2017, 07:36:43 UTC
d421bd7 Moritz Muehlenhoff 30 January 2017, 12:04:06 UTC Remove otto from piwik-roots I've written a script to run consistency checks on group memberships in the WMF cluster. Remove otto from piwik-roots since he has cluster- wide root access anyway. Bug: T142836 Change-Id: I2700c18b43d4e81aa32ecd85694730f0ac9ca8c8 02 February 2017, 07:33:17 UTC
28b166e YuviPanda 31 January 2017, 21:14:00 UTC docker: Gently wade into new coding guidelines The reason I'd like these to be parameterized is that $volumes_to_remove should be set to a particular value in role::toollabs::k8s::worker, and splitting that out into horizon seems unnecessary. Change-Id: I0c3152e6479b2e8b4ed75a1cb0f6ab949f632c93 02 February 2017, 07:09:15 UTC
51f1d15 Daniel Zahn 02 February 2017, 01:41:11 UTC installserver: add firewall hole for rsync also for IPv6 Just adding the ferm hole for IPv4 is not enough, we also need to open it for IPv6 which is always preferred when you have dual-stack. Bug: T84380 Bug: T132757 Change-Id: I119e5aa83e54b8ed8613505fef3c81fb2d67253a 02 February 2017, 01:41:11 UTC
5b1f27b Daniel Zahn 27 January 2017, 01:08:24 UTC aptrepo/rsync: flip the "if"-logic around instead of a negation Avoid using a negation, make the logic more obvious. Also, fix the rsync command, missing trailing / after dir. Change-Id: I95bf8b121870a4a8b2be9268e881c57e6da86f5b 02 February 2017, 01:29:10 UTC
c8bdc19 Daniel Zahn 26 January 2017, 04:42:10 UTC aptrepo: add cron to rsync APT data automatically After testing the rsync setup and manually running the command, syncing data from carbon to install1002 and install2002, now add a cron job to make that automatic. For the moment we will still sync from carbon as source, and over to install2002. Bug: T84380 Bug: T132757 Change-Id: I0f3570890d5c949db98763b3f1ad5bcb0f35d23c 02 February 2017, 00:33:25 UTC
65688ea Madhumitha Viswanathan 01 February 2017, 23:12:53 UTC toollabs: Better exception handling for precise-reminder Change-Id: I41851637e1f5962124e610392d283faf1563d4a4 01 February 2017, 23:16:26 UTC
541d23a Madhumitha Viswanathan 01 February 2017, 22:37:40 UTC toollabs: Fix precise reminder block indent Change-Id: I37ba480a0090ec741c039002ead37a54b056edfa 01 February 2017, 22:38:10 UTC
518a3b8 Jaime Crespo 01 February 2017, 22:14:21 UTC Add replication client grants to phuser Change-Id: I1ceffab470e9405e77ed96bfd1d1509c5495badb 01 February 2017, 22:25:40 UTC
fd3d0c1 Madhumitha Viswanathan 01 February 2017, 22:10:38 UTC toollabs: Fix precise_reminder script file name Change-Id: I411ba9b9909da6aede01b678aaa5611865351034 01 February 2017, 22:10:38 UTC
57ce178 Madhumitha Viswanathan 01 February 2017, 19:55:27 UTC toollabs: Add temp role and cron to send weekly tools precise deprecation reminders Bug: T149214 Change-Id: Ia6a0e527cd10704446711e567ca363002f6adca4 01 February 2017, 21:59:23 UTC
d8619bb Andrew Otto 01 February 2017, 21:30:12 UTC Clean up analytics icinga contactgroup Change-Id: Ia0fca275708a33ae0caf91214baae3ffb40b4c99 01 February 2017, 21:30:12 UTC
7f3539e Mukunda Modell 01 February 2017, 19:22:28 UTC Phabricator: enable elasticsearch backend Search queries keep crashing mariadb, so switch to elastic for now. Change-Id: Ib77cb6e59ffa25ca4a1faec111de2bac9aeb01a2 01 February 2017, 19:22:28 UTC
50bae28 RobH 01 February 2017, 19:09:07 UTC setup params for gerrit2001 cloned existing recipe but removed swap, since system should have enough memory to do its job without it. Bug:T152525 Change-Id: I229816624721739673d621255fd56370b3955bc7 01 February 2017, 19:11:59 UTC
c88cff5 Andrew Otto 01 February 2017, 19:01:33 UTC Install python-ua-parser as an eventlogging dependency Bug: T156821 Change-Id: Ia141dfbccd72ec5a29eba42f6f1b6bcb1efd69ec 01 February 2017, 19:01:33 UTC
e4fb49c Marius Hoch 01 February 2017, 15:52:30 UTC Use 5 instead of 4 shards when dumping Wikidata A full run takes ~16 hours by now, so let's add another shard to get it down to a (more reasonable) 13 hours. Change-Id: I494b78a9aa2fbe122e32593fd6807788532802d3 01 February 2017, 15:52:30 UTC
924aafb elukey 31 January 2017, 12:13:19 UTC Require cassandra-wmf-tools and jvm-utils for Cassandra Change-Id: Ie1401abe7fb1580017309ada8ef71ceb12df6442 01 February 2017, 15:32:11 UTC
904f2fc elukey 31 January 2017, 11:04:19 UTC Extend role memcached to the new codfw mc hosts The current mc1001-->mc1018 eqiad Redis hosts are replicated to 16 codfw hosts, mc2001->mc2016, since the first and the last are running two Redis instances each to match the difference with eqiad. This patch separates also the regexes to match the hosts to ease the work of the final puppet cleanup. Bug: T155755 Change-Id: I1a1ba9effd9e6adf9bb8c6c642faa6a92ce6eb15 01 February 2017, 14:13:43 UTC
19ab81f Jcrespo 01 February 2017, 12:38:54 UTC Revert "phabricator: Increase phabricator dbs buffer pool" This reverts commit 4044898240185c4c38fab229f1b03ae0ac5fc130. Change-Id: I113af60b0f32aa1df57196e047b3e12ee88434e7 01 February 2017, 12:39:18 UTC
dbc93da Moritz Muehlenhoff 01 February 2017, 09:34:38 UTC Remove access credentials for jhobs Change-Id: I6b0f693a8b2f73a93b30c68f03bb7910da9c92f6 01 February 2017, 09:34:38 UTC
ddea7ef Filippo Giunchedi 01 February 2017, 09:12:52 UTC lvs: make thumbor.svc alert non-critical Thumbor is receiving shadow traffic and not serving responses back to users. Don't page in case something goes wrong until launch. Change-Id: I979d78a760acb2ca95b90a6391d7e1e600e49ae7 01 February 2017, 09:13:56 UTC
4e9f212 elukey 30 January 2017, 11:49:16 UTC Add aqs1008-a to the AQS Cassandra cluster Bug: T155654 Change-Id: I05b89d12cc164c566d09296977c9b732275376ff 01 February 2017, 07:27:33 UTC
9658f35 Moritz Muehlenhoff 30 January 2017, 12:10:19 UTC Remove aaron from logstash-roots I've written a script to run consistency checks on group memberships in the WMF cluster. Remove Aaron from logstash-roots since he has cluster- wide root access anyway. Bug: T142836 Change-Id: I6c7bcce816510e024bc0ac1367c2f2278c505317 01 February 2017, 04:21:38 UTC
2093f50 Moritz Muehlenhoff 30 January 2017, 12:19:32 UTC Remove aaron from contint-users I've written a script to run consistency checks on group memberships in the WMF cluster. Remove aaron from contint-users since he has cluster- wide server access anyway. Bug: T142836 Change-Id: I7cf7dda97b28d883d7d65c002f69f5573952c37a 01 February 2017, 04:14:55 UTC
f1dc558 Daniel Zahn 01 February 2017, 03:11:32 UTC installserver::nginx: use "do_acme" in Hiera instead of custom var Instead of using the custom "apt::wikimedia::active-host" variable in Hiera, use the standard "do_acme: false" to disable Letsencrypt cert creation. This should avoid issues with nginx not starting when the LE config snippet is not generated. Change-Id: Ica12256f6ecf1f62c43c1246f357af0ef7ce6131 01 February 2017, 03:18:03 UTC
adbda44 Daniel Zahn 01 February 2017, 03:00:11 UTC installserver: add install1002/2002 to hiera Add freshly installed install1002/2002 to Hiera to have debdeploy grains and skip the Letsencrypt cert creation. Bug: T156440 Bug: T132757 Change-Id: I2fd03a4866416d7f5fe96991389e2168aa505d09 01 February 2017, 03:00:11 UTC
3a52c5d Daniel Zahn 01 February 2017, 02:16:06 UTC DHCP: add install1001,install2001 Install 1002 from 1001 and 2002 from 2001, then delete 1001 and 2001. Bug: T156440 Bug: T132757 Change-Id: I6787c7f583830ac35545335529138d89265b81ae 01 February 2017, 02:31:28 UTC
cac0c1f Filippo Giunchedi 07 January 2017, 00:06:25 UTC ganglia: display deprecation banner user_header.tpl is automatically included by header.tpl on each page if present. Display a deprecation banner and remember the user's answer for 14 days. Bug: T145659 Change-Id: I30a027171a17989853972bb4e8024d02914e3dad 01 February 2017, 00:52:30 UTC
d6ededb juniorsys 26 January 2017, 07:04:38 UTC deployment: Linting fixes Use full names for class names, as relative names are not allowed in future Puppet versions Add trailing commas to abide by the Coding Style guidelines Bug: T93645 Change-Id: If8e28cd43be46ef7490c7835caf8c3d69f4c12b3 01 February 2017, 00:45:23 UTC
0b76412 juniorsys 26 January 2017, 07:06:40 UTC dnsrecursor: Linting fixes Use full names for class names, as relative names are not allowed in future Puppet versions Bug: T93645 Change-Id: I171637fca05d43c627cb562740897ee003d07747 01 February 2017, 00:40:48 UTC
3ac6719 juniorsys 26 January 2017, 07:28:42 UTC jupyterhub/keyholder: Linting fixes Add trailing commas to abide by the Coding Style guidelines Bug: T93645 Change-Id: I0926583dc77d9cd242a03bfa9b4912931808e431 01 February 2017, 00:36:25 UTC
93d2c9b Daniel Zahn 01 February 2017, 00:22:46 UTC wikimania_scholarships, planet: one line per include, full class names Lint fixes as suggested by Juniorsys. (Change-Id: I31cf5969dc8142d et al) Change-Id: I372b86efbf893a39f40868af33c3a90b4303f5ae 01 February 2017, 00:22:46 UTC
65521bd Daniel Zahn 01 February 2017, 00:13:12 UTC planet: use one line per include, full class names Lint fixes as suggested by Juniorsys. (Change-Id: I31cf5969dc8142d et al) Change-Id: I12b1e750926c1a84b0ac0716d7f87152c1ef635d 01 February 2017, 00:14:53 UTC
93a6f2e Daniel Zahn 31 January 2017, 23:10:45 UTC add install1002/2002 to replace 1001/2001 Creating install1002 and install2002 which will be installed as identical VMs with reasoanble disk space. They will be installed using the current install servers and then replace them. Bug: T156440 Bug: T132757 Change-Id: Ia88332f2d939254c11da9171b2879fac2be2a187 31 January 2017, 23:25:16 UTC
62b4b26 Andrew Otto 31 January 2017, 22:06:05 UTC Require proper file for rsync-published-datasets cron Change-Id: Ie0657ee3f07d586bb557206204d1613f6868e497 31 January 2017, 22:06:05 UTC
fe9a43d Andrew Otto 31 January 2017, 20:30:55 UTC Create /srv/published-datasets on stat* boxes, and hardsync them to analytics.wikimedia.org/datasets Bug: T125854 T132594 Change-Id: I37b11112be6265bcd781abc93d2c5b52c5f105d4 31 January 2017, 21:59:50 UTC
a92c5f1 Andrew Otto 26 January 2017, 22:05:54 UTC Add hardsync shell script This uses hardlinks to sync multiple source directory content into a single directory. Bug: T125854 Change-Id: I58fbe3242a5895d288a2170df6226e0cf7cf09a1 31 January 2017, 20:48:02 UTC
c5b959a Andrew Otto 31 January 2017, 20:17:38 UTC Revert some of the changes last week to datasets.wm.org - we will cleanup at analytics.wm.org instead Bug: T125854 Change-Id: I9aa2ddcfdcf26fb3e62bd744b4678a9f3cd4555f 31 January 2017, 20:20:19 UTC
881623d juniorsys 26 January 2017, 10:38:40 UTC planet/pmacct/programdashboard/pybal lint changes Use full names for class names, as relative names are not allowed in future Puppet versions Add trailing commas to abide by the Coding Style guidelines Bug: T93645 Change-Id: Ia2508c41d679d4679df86f82146a3e715f8dc29e 31 January 2017, 19:49:01 UTC
22d8465 Jaime Crespo 31 January 2017, 15:08:30 UTC mariadb: Add TLS support for tendril Bug: T111654 Change-Id: I47c171cb4b8c64f172c037d98d4e0ad16b60d9fd 31 January 2017, 17:16:46 UTC
64265ce RobH 31 January 2017, 17:06:38 UTC Revert "adding icinga cert monitoring for *.corp.wikimedia.org" the initial change broke icinga and cannot resolve directory.corp.wikimedia.org. not sure why yet, reverting and will determine why it wont resolve in the check, but will on the command line, without the host bieng in a downtime state on its config. This reverts commit 5ae578eee66d97d40eb13557362183fbc65e166f. Change-Id: I8d0609720753bf498527c74564b9479233e4ec0a 31 January 2017, 17:06:38 UTC
5ae578e RobH 31 January 2017, 01:33:41 UTC adding icinga cert monitoring for *.corp.wikimedia.org using the directory.corp.wikimedia.org service per Byron's suggestion for a public facing point to monitor this ssl certificate. Change-Id: If18ee070e28461993e8acf94e640f9ed133fb5a1 31 January 2017, 16:43:07 UTC
3476b4a Guillaume Lederrey 31 January 2017, 15:16:00 UTC elasticsearch - deploy the same administration scripts on Jessie and Trusty The es-tool script was not deployed on Jessie as we feared some incompatibilities with the version of the python-elasticsearch package available on Jessie. Testing shows that things are broken in the same way on Trusty and Jessie, so no need to have a special case. The part that we use regularly are working correctly. Relying on system packages for python-elasticsearch is problematic anyway as the API of elasticsearch changes over time. We should use a version of python-elasticsearch aligned with the version of elasticsearch that we have installed, or better, remove this dependency and talk directly to elasticsearch. The rewrite of es-tool will take some more time, let's align things first. Bug: T151326 Change-Id: I9c2ae2654c4ec70e3608623b3d4a26885af2f657 31 January 2017, 15:34:06 UTC
116e95c Andrew Otto 31 January 2017, 15:20:49 UTC Temporarily increase refined webrequest retention to 90 days This will allow us to have a little more time to compute Fundraising Banner Impressions stats for December. Bug: T155141 Change-Id: I893ee67f9f381ddc1e98c3a4d762a599f305f545 31 January 2017, 15:20:49 UTC
b651d5e Jaime Crespo 31 January 2017, 13:46:12 UTC sanitarium2: Enable TLS, disable Toku-specific config This is in preparation for a full restart & upgrade. Bug: T111654 Change-Id: Iac6a119b84a0221036ded3672be96f908ae6365c 31 January 2017, 14:24:27 UTC
c391ae6 Moritz Muehlenhoff 31 January 2017, 13:42:50 UTC Also remove springle from Icinga config Change-Id: I7617408cbaeac124ed13418f28d6707a65ed8145 31 January 2017, 14:05:10 UTC
793aa26 Guillaume Lederrey 31 January 2017, 13:25:13 UTC elasticsearch - ensure data directory with puppet Default data directory is created by elasticsearch .deb package. If we move data to /srv, the data directory needs to be created. Bug: T151328 Change-Id: Ie11729b6af6215106c4db202267efb7ac16fb6c7 31 January 2017, 13:31:24 UTC
deb7ad0 Moritz Muehlenhoff 31 January 2017, 12:43:25 UTC Remove access credentials for springle He confirmed that he doesn't need access any longer. Change-Id: I341c6a91c0c2a0e21704a123296665628ad8c61d 31 January 2017, 12:43:25 UTC
a2b25b3 juniorsys 26 January 2017, 10:34:28 UTC prometheus: Linting changes Use full names for class names, as relative names are not allowed in future Puppet versions Add trailing commas to abide by the Coding Style guidelines Bug: T93645 Change-Id: Id2c2fbc1a892708a057ce260c3696f517d1b3a73 31 January 2017, 12:13:09 UTC
3f82d4f Guillaume Lederrey 31 January 2017, 11:01:44 UTC wdqs - categorize wdqs1003 as a wdqs node in site.pp Bug: T152643 Change-Id: I1e3f3b7cc91b9bb5115d9643b9ac4c618bb1c5c1 31 January 2017, 11:03:58 UTC
5b54541 Guillaume Lederrey 31 January 2017, 09:29:43 UTC wdqs - configure new wdqs1003 node As we are moving toward the more standard use of /srv as a data partition, the default data location in the wdqs class has been updated. wdqs100[1-2] are treated as the exceptions, until they are reimaged with the new data location. Bug: T152643 Change-Id: I93ca1b32f8bfe56fb41387e060b8dcd7cf8b12d8 31 January 2017, 09:29:43 UTC
5555690 Andrew Bogott 31 January 2017, 03:58:04 UTC Revert "Keystone: Turn on caching of tokens and catalog" This didn't fix the one thing I wanted it to fix, and who the hell knows what else it might break... tonight keystone seemed to be handing expired or invalid tokens over to nova-api. This reverts commit 03f55287c8ee09f0f19d405f0a5dd63dcbe025d4. Change-Id: I38ee1c37a142fa1df7fab110d8b45ea2e8d2ca38 31 January 2017, 03:59:43 UTC
852b182 RobH 31 January 2017, 00:02:02 UTC decom db2015 old db host being decommissioned Bug:T149102 Change-Id: I8adae574ab7ba16f9d97cd1c577452c1ac469375 31 January 2017, 00:02:02 UTC
ddff9b8 Tyler Cipriani 20 January 2017, 16:31:17 UTC Scap: Bump version to 3.5.1-1 Update scap version and modify config for new version. Bug: T127762 Change-Id: I015350dca1ffc7ee5746d587c8370d428383af35 30 January 2017, 20:45:25 UTC
b61b14b Daniel Zahn 27 January 2017, 21:34:11 UTC admin: fix log file perms for dc-ops on jessie This admin group is supposed to have access to syslog, (just that until recently installservers (carbon) were on precise so the "syslog" line did the trick there. Now that install1001/2001 are installservers on jessie and have systemd that doesn't work there. So to fix this regression, add the standard journalctl line we use in many other admin groups as well. Otherwise it means papaul can't do things such as debug DHCP issues when installing servers and it slows us down. Do not remove the old line just yet until carbon is gone for real. Bug: T156529 Change-Id: I437f534326146b84941dc62764fca05b087c2537 30 January 2017, 19:02:54 UTC
87bfc35 YuviPanda 30 January 2017, 18:30:54 UTC labs: Fix novaproxy not working when use_ssl is false Change-Id: I952c49924c0aa6d2967423c9976cbdfde9a87894 30 January 2017, 18:30:54 UTC
f21ce59 Tim Landscheidt 18 December 2016, 07:52:06 UTC Tools: Disable automatic backups of aptly repositories Currently Puppet usually hits the timeout in puppet-run on tools-services-01 because backing up the aptly repositories to NFS takes an immense amount of time. On the other hand, the backups on every Puppet run are of no true value as meaningful changes to the repositories only happen once per weeks or months, and in those cases it is easy to manually trigger a backup. Thus this change disables the automatic backups of aptly repositories. Bug: T150726 Change-Id: I6e06e9e2e4c4be5b3f6555c22c0bb36181675318 30 January 2017, 15:23:37 UTC
85cd775 elukey 30 January 2017, 13:44:39 UTC Add aqs1007 to AQS's conftool data Bug: T155654 Change-Id: I17d8eb73e3aaa76d9755bbb4825af179e9a5bb90 30 January 2017, 15:02:16 UTC
8da5703 Guillaume Lederrey 30 January 2017, 13:51:13 UTC maps - increase replication frequency to each minute on maps-test cluster We have a hard to reproduce bug related to OSM synchronization (T145534). Increasing the sync frequency might help reproduce the bug more often and help understand (and fix) it. Note: when deploying this, the state file (/srv/osmosis/state.txt) needs to be updated accordingly with minutely state.txt (eg. https://planet.openstreetmap.org/replication/minute/002/295/440.state.txt). Bug: T145534 Change-Id: Iee5e10a3644d8a6585e72347b86af748b4cf8aa9 30 January 2017, 13:56:02 UTC
94666b0 Moritz Muehlenhoff 30 January 2017, 12:17:11 UTC Remove elukey from analytics-admins I've written a script to run consistency checks on group memberships in the WMF cluster. Remove elukey from analytics-admins since he has cluster- wide full sudo anyway. Bug: T142836 Change-Id: I66851b03d010286e89770176d76dc12a8005477b 30 January 2017, 13:40:29 UTC
71f3adf Alexandros Kosiaris 27 January 2017, 15:31:22 UTC ores::redis: Enable diamond collector Enable the diamond collector for redis in the ORES redis environment. That should allow us to create graph to monitor the databases more closely Change-Id: Idf752db74219df64ad892a23cf4e25ef7cdb42ad 30 January 2017, 12:26:06 UTC
fe81792 Moritz Muehlenhoff 30 January 2017, 11:40:03 UTC More email addresses Change-Id: I7f41368b0e7429f28647c63fd115d8f796f342ac 30 January 2017, 11:40:03 UTC
c7fd39d Ariel T. Glenn 30 January 2017, 09:42:08 UTC Turn off centralauth table dumps. Bug: T153633 Change-Id: Idc2b085c33f2d76c67fec32a965d4da368226385 30 January 2017, 10:56:22 UTC
4ba86fd Emanuele Rocca 25 January 2017, 11:11:49 UTC cache: remove varnish_version4 from hiera and salt We have been running Varnish 4 for a while now, no reason to specify the version in hiera/salt. Change-Id: I539fc29f634bd4d3a9b3e092f9d6eca23d54d80e 30 January 2017, 10:40:04 UTC
b5ca759 Gilles Dubuc 20 January 2017, 10:42:23 UTC Revert "Remove broken Thumbor IP throttling from configuration" This reverts commit 82e128d167dc4b563236faf95226a28e11a620c8. Change-Id: I851705e4fd97a0d0fc7680f9df4e5317ea9f5d33 Depends-On: Id2c3cd2800e632e10974cdf189c3f627b3eb5d37 30 January 2017, 09:48:45 UTC
c4a746c elukey 28 January 2017, 10:19:09 UTC Enable AQS aqs1007-b cassandra instance Bug: T155654 Change-Id: I0eae04930d8d3e0cd8ffae2d86c1df531576d69a 30 January 2017, 09:16:18 UTC
a76d098 Moritz Muehlenhoff 19 January 2017, 10:06:09 UTC Switch app servers in codfw to systemd-timesyncd Test-drive systemd-timesyncd on some selected servers for a month before enabling it in general; already enabled on cp4* and swift servers in esams. Bug: T150257 Change-Id: I0ddb7fba1331eaec2cce2a11f6272bf548028384 30 January 2017, 07:35:51 UTC
c39665b andrewbogott 27 January 2017, 23:12:12 UTC Keystone: Define some stopped services on the spare host uwsgi-keystone-admin and uwsgi-keystone-public services are referred to in varions refresh lines, so they need to be defined on labcontrol1002 even though the services don't run there. Change-Id: Ic49b51ca7c17fa93968918e572d4d74ebc45fc57 27 January 2017, 23:13:10 UTC
016335d Daniel Zahn 27 January 2017, 00:59:30 UTC aptrepo: add second rsync module for entire /srv/ To also sync things like /srv/junos, /srv/megacli and others that are not inside /srv/wikimedia. The question is if these technically belong in the aptrepo module but at least sync them once before shutting down carbon. Bug: T132757 Change-Id: Ief81c69d8d2c473e6e35ac9efb2a1346e6076e22 27 January 2017, 22:19:26 UTC
31882dc halfak 27 January 2017, 18:20:16 UTC ores:Adds aspell-ro to ores base. Change-Id: Ic06ab72ddb095dc3a3352121e88ce823a7ea318b 27 January 2017, 22:12:26 UTC
a62f64c andrewbogott 27 January 2017, 19:28:33 UTC Keystone: use uwsgi::app instead of service::uwsgi Bug: T156337 Change-Id: I6285d053eb1c445a9d8feb7a3f65058dd65ea7b3 27 January 2017, 21:40:45 UTC
56d79f2 Daniel Zahn 25 January 2017, 22:12:41 UTC ssl: delete stream.wikimedia.org cert This cert has expired in 2016 and stream.wikimedia.org now resolves to misc-web varnish, so this should not be used anymore. openssl x509 -noout -dates < stream.wikimedia.org.crt notBefore=Jun 23 00:00:00 2014 GMT notAfter=Jun 27 12:00:00 2016 GMT Bug: T134361 Bug: T155524 Change-Id: I4136168a21e7fd1013ced88c397db130dee0d7c5 27 January 2017, 19:35:26 UTC
03f5528 andrewbogott 26 January 2017, 16:54:28 UTC Keystone: Turn on caching of tokens and catalog Bug: T156337 Change-Id: If9677640253afc8d856ad4888c7808c5f14e2635 27 January 2017, 17:22:02 UTC
e620086 Moritz Muehlenhoff 27 January 2017, 16:27:01 UTC Two more email addresses for ISI Foundation researchers Also move to organisational email addresses instead. Change-Id: Ia027f168ae323465b1930b9772fb99be09223b3c 27 January 2017, 16:27:01 UTC
5038439 Moritz Muehlenhoff 27 January 2017, 16:09:55 UTC Two more researchers Change-Id: I7c390f3a2831a6624496df089b28f502f8adc21e 27 January 2017, 16:09:55 UTC
28c1ddb Moritz Muehlenhoff 27 January 2017, 15:55:00 UTC Two special cases; account expiry dates for research fellows Change-Id: Ib9553f7bd90d49485befe860ed8f238a5dc7e9a8 27 January 2017, 15:55:00 UTC
d2566a3 Moritz Muehlenhoff 27 January 2017, 15:39:23 UTC More email addresses Change-Id: Id3acd9ce21de75b08bb1596b0ca6ffd2b45904c1 27 January 2017, 15:39:23 UTC
35b63e0 Moritz Muehlenhoff 27 January 2017, 14:50:43 UTC Add more researchers (email and expiries) Change-Id: I1bb8bf9183ea4197aaa204022b9826941d4af9e1 27 January 2017, 14:52:54 UTC
27119b9 Emanuele Rocca 20 January 2017, 02:01:12 UTC Text VCL: consolidate mobile hostname rewrite regex Extend the mobile hostname rewrite regex to work with all hostnames, not only language-specific ones such as en.m.wikipedia.org. The two cases left out are m.mediawiki.org and m.wikidata.org, which currently need www in the rewritten hostname. Bug: T155504 Change-Id: I5fcb334639c4b7bbb6ab82704cfd3d543b21af4c 27 January 2017, 14:41:34 UTC
86a20e9 Moritz Muehlenhoff 27 January 2017, 14:25:02 UTC Add more email addresses and account expiration dates Change-Id: Icdb64139d5cd79d571662e045bd3816223cfd8eb 27 January 2017, 14:29:36 UTC
f767240 pt1979 27 January 2017, 02:23:43 UTC DHCP: Add DHCP entries for mc2019-mc2036 Bug:T155755 Change-Id: Iccad560a18199ae849d713f7fe9e5366155cc269 27 January 2017, 14:16:12 UTC
3dfa92b Faidon Liambotis 27 January 2017, 14:00:24 UTC labs: remove keystone alerts Clearly noone is paying attention to them: $ egrep -c '(has roles in every|has only observer|admin and observer)' icinga.log* icinga.log:1555 icinga.log.1:3108 icinga.log.2:2790 icinga.log.3:2277 icinga.log.4:2774 icinga.log.5:3200 icinga.log.6:4471 icinga.log.7:3645 Running the check results in an error once every 20 times or so, with a backtrace that ends with the following error: keystoneauth1.exceptions.connection.ConnectFailure: Unable to establish connection to http://labcontrol1001.wikimedia.org:35357/v3/roles Change-Id: I421fbb11019db3a155217e9fba1130b97a08f95c 27 January 2017, 14:05:09 UTC
back to top