https://github.com/wikimedia/operations-puppet

sort by:
Revision Author Date Message Commit Date
642b9f0 cassandra: unblacklist 'max' metric Bug: T116913 Change-Id: I8807ed369c12274807c22a7a43e7ef509a3e36d3 28 October 2015, 16:41:10 UTC
482a176 reqstats: test on cp1065 (eqiad text) as well Should give us higher-rate test data to look at, the misc-cluster testing looks good so far! Change-Id: I3ab490e691d7d6ef38be905156312691ed81807e 28 October 2015, 16:37:24 UTC
3d7403c cassandra: fix HeapDumpPath and ErrorFile settings pointing HeapDumpPath to a directory will make the jvm pick a default filename containing the pid as well, also change ErrorFile name to its default Bug: T116814 Change-Id: I808f8978d0550a1014cfa6156a6fef6659f2d663 28 October 2015, 16:31:19 UTC
0c08cb4 Open up port 8000 for burrow (on krypton) Bug: T115669 Change-Id: I0065e1f7e5307c29f4af780dd29c791113b3120a 28 October 2015, 16:28:45 UTC
04c8e6a Labs: switch active labstore From labstore1002 to labstore1001 Bug: T107038 Change-Id: Iad709691b9f347f4b1adba9b4cd74867320eac43 28 October 2015, 15:42:46 UTC
6ee05fb contint: set Zuul URL based on server fqdn The zuul_url parameter is set by zuul-merger to indicate to the Jenkins jobs from where patches can be fetched from. Since we have the Zuul scheduler and merger cohosted on the same machine gallium.wikimedia.org, I used the service DNS entry 'zuul.eqiad.wmnet'. We are going to add a zuul-merger on scandium.eqiad.wmnet (T95046), so the zuul_url parameters needs to vary based on the host. Previously we had: ZUUL_URL=git://zuul.eqiad.wmnet That will now becomes: ZUUL_URL=git://gallium.wikimedia.org For scandium, it will be: ZUUL_URL=git://scandium.eqiad.wmnet Bug: T95046 Change-Id: I07323dfb5373be12d22cee4a1d4c36bf6e3ab3d6 28 October 2015, 15:08:32 UTC
b3b02fd Enable ferm on tin Change-Id: I934c1862dfd9c48e2069ad92a20ce63ae2cacf54 28 October 2015, 14:47:01 UTC
1c2692f burrow: Add new module for burrow Burrow(https://github.com/linkedin/burrow) is a consumer lag monitoring tool for Kafka, it will monitor the consumer groups on our Kafka clusters and send emails on their status. This patch installs and runs burrow on krytpon. Bug: T115669 Change-Id: Iaf6b2ee804df8f5e951fc7ced770b2565e3adb65 28 October 2015, 14:38:21 UTC
b20d822 role::statistics: fix file path Fixup for I32f74d9030 Change-Id: Id3622f1eba6da38a61b852fd6f4845dad0ce63a5 28 October 2015, 14:06:18 UTC
69138ca Increase the host and account size on P_S config As we have more hosts than usual connecting to mysql, 100 slots in performance_schema configuration are not enough to contain all hosts and users. Increasing it to 300, which should be enough for all our clients per datacenter. Change-Id: Ic2e1fef6e3ca4e24872e357761fc41426290450f References: T99485 28 October 2015, 14:00:07 UTC
9e24804 service_checker: correctly return an error in case of timeouts Bug: T116739 Change-Id: I16f32e57bfa6da0ff583080fc115f92faad2e413 28 October 2015, 12:43:06 UTC
b634798 Revert "cassandra: stop declaring service resource" it doesn't play well with nrpe::monitor_systemd_unit_state: Error: Failed to apply catalog: Could not find dependency Service[cassandra] for Nrpe::Monitor_systemd_unit_state[cassandra] at /etc/puppet/modules/cassandra/manifests/instance.pp:224 Change-Id: I58997603a03f2fb76a1ad4a13307d700042adcfd This reverts commit 2585d9cff1c295cff9246cb494bf7949a55c61fc. 28 October 2015, 12:13:28 UTC
2585d9c cassandra: stop declaring service resource this let's service_unit install the init/systemd files but don't otherwise manage the service Change-Id: Ib18f9a2be06c28c6cdc5fedfa2f99ca310300712 28 October 2015, 12:03:24 UTC
73a1ab8 maintenance: amend hiera lookups Change-Id: I287fce7f3175bd553974a9833fd292d4e85e33d1 28 October 2015, 11:15:12 UTC
bbfdd02 Enabling performance schema experimentally on db1022 db1022 is currently depooled. Solify its configuration to enable performance_schema running on it. If the test works successfuly, we will enable performance_schema by default on all hosts, but we need to measure (performance) regressions first. Bug: T99485 Change-Id: I068dd8181f991dd5cb90a232f46595645fe086d9 28 October 2015, 11:00:52 UTC
a1972a9 Merge changes in the mariadb repo Change-Id: I64ca73402bc8b3214280b68fb42a0d7a5f28f734 28 October 2015, 10:13:27 UTC
58298d9 mw1152: convert to be the HAT maintenance host Bug: T116728 Change-Id: Id61a7310de5c1e878414ad5d689c183b67aab7b8 28 October 2015, 10:06:17 UTC
34603b7 cassandra: updated gc settings Updates GC settings for a 12G heap size, 8M region sizes, and an initiating occupancy threshold of 35%. See: https://phabricator.wikimedia.org/T106619#1604486 Note: This was applied once before as https://gerrit.wikimedia.org/r/#/c/236391/4, but reverted because the additional command arguments pushed the classname off the end, wreaking havoc with Icinga. Now that we are using systemd units everywhere, this should no longer be an issue. Bug: T106619 Change-Id: I9089354266e701be34faf2f9f066d0cb2c5e4224 28 October 2015, 09:42:59 UTC
f105652 mediawiki: group general monitoring scripts in a single role Change-Id: I013e19b1da9ce073405dd1da6389aa796e17f49f 28 October 2015, 08:58:39 UTC
643ef00 r::mw::maintenance: include role::mediawiki::common Bug: T116728 Change-Id: I080e553373cc23f71af7637987f2839c2a60cadb 28 October 2015, 08:40:52 UTC
197ade3 role::deployment::server: drop mod_dav Remove apache mod_dav because it's not used anywhere. Change-Id: I5fc994b0efd2e7b8e3b85c182556586200ca8a9b 28 October 2015, 07:21:58 UTC
cff5932 role::deployment: remove test role It was a useless one-purpose "role" that was used just by beta. Change-Id: I2d46fe778b26c821cf9efbc23579c71fc923b1b2 28 October 2015, 07:21:58 UTC
71086a7 deployment::mediawiki: rename wikitech::wiki::password class And remove the module in the process as it's useless Change-Id: If4a513f29dc84cf869de93bf775e3fc5de2784e2 28 October 2015, 07:21:58 UTC
79c8937 role::deployment: move things to deployment::mediawiki Change-Id: Ib57fcee46934b2e0e97c18d887bc0b8b20e4d7fb 28 October 2015, 07:21:58 UTC
efa625b role::deployment::server: reorganize code Change-Id: Iecbaa257349b013cf62aac5baeca6d2b15db316f 28 October 2015, 07:21:58 UTC
b28677b role::deployment: move to role module Change-Id: I32f74d90307244a36795f143c6d7369e4cbb7966 28 October 2015, 07:21:58 UTC
af0f588 labs restbase: Add en.wikivoyage.beta.wmflabs.org Change-Id: I5497bde80a3e7cfbd8b6cc04639acbbbdc87df8b 28 October 2015, 00:53:10 UTC
0acb62e labs kvm ssl cert monitoring: fix nrpe command The NRPE command needs the full path to the certfile, not just the cert name itself. Bug:T116332 Change-Id: I7c254f358c2929d8678730fa98c80aa0fe730880 28 October 2015, 00:11:32 UTC
35fcddf labs kvm ssl cert monitoring: fix it This is a follow-up to I83ba82a25e8e87 which didn't quite work yet because the plugin needs to be executed locally where the certs are, so we need to turn this into an NRPE check. And check_ssl.cfg is the config for check_ssl, a different plugin from check_ssl_certfile. Bug:T116332 Change-Id: I44fb35f362898d178d11f300c23ff227e59de5ba 28 October 2015, 00:00:26 UTC
b0a4762 smokeping: remove psw2-eqiad, decom'ed Bug: T115924 Change-Id: I0e7e3babecf716d6a835d83a0864bc00c76966f0 27 October 2015, 23:28:22 UTC
a7a350e logstash: fix double quoted strings & alignments fixes 21 puppet-lint warnings of the types: WARNING: indentation of => is not properly aligned WARNING: double quoted string containing no variables Change-Id: I22b7d6976006bc0233766435207cb50db7bee8e9 27 October 2015, 22:53:56 UTC
0f3f2a5 Fix mwscriptwikiset dblists paths Bug: T113721 Change-Id: Ie7d37b1b70e824a5383fb664452b5f2c2728fd45 27 October 2015, 22:44:20 UTC
76a97ce Add MAC entries for ms-be20[1-2][0-6] Bug:T114712 Change-Id: Ief86db404780debb8e9cd67f983dd0d912cc757c 27 October 2015, 22:12:06 UTC
151fa24 tools: Make main page check in icinga Can be made paging after testing to make sure it works Also move the NFS check to a more appropriate place Bug: T116925 Change-Id: I35c5bfe8941a509a2c92283e82208868e19e0e87 27 October 2015, 21:25:14 UTC
cf07906 Typo fix in labvirt-ssd.cfg Change-Id: I780c9802ad107e84bb273d8a5297b4f0fc980b72 27 October 2015, 21:16:33 UTC
2a4025b xenon: sort files entries in descending date order Change-Id: I4855f7c342ddc42d6a0e03626d003e45a46892a9 27 October 2015, 21:10:35 UTC
66f5485 Modify labvirt-ssd to more closely resemble db.cfg I'm just trusting that db.cfg actually works. Change-Id: I3e539071723b834a1fa5f44e1a530ce82cfe70f8 27 October 2015, 09:21:17 UTC
cefb1f9 Revert "Revert "dynamicproxy: Make blocked user agents configurable"" This reverts commit f64ff16ed666ccef2e8b14d38c58563922bf4464. Change-Id: I5a2999abb687174b70187c0999a5265840517899 27 October 2015, 20:29:54 UTC
1f6a1e7 contint: install rake on Nodepool instances On Debian Jessie, neither 'ruby2.1' nor 'ruby' provides the 'rake' command unlike Ubuntu. The Jenkins job rake-jessie ended up using 'rake2.1'. The reason is that Nodepool instances do not include contint::packages which provides 'rake'. Move Package['rake'] to contint::packages::ruby so it get installed on Nodepool. Change-Id: I06f5fb66966e70bbf16fcc1d9d928e063e1e728c 27 October 2015, 20:26:38 UTC
99508b9 Add monitoring for the kvm ssl cert, labvirt-star Bug: T116332 Change-Id: I83ba82a25e8e877de0cb745b0d768e34fb784310 27 October 2015, 20:22:51 UTC
b58dcbd Reenable DSH checks for codfw, uncomment hosts that are up It was disabled when that DC was still very green and caused scap failures. Now, it's mature and the lack of this check results in icinga alerts due to code being out of date. Need to fix that. To avoid immediate alerts, reenable all hosts that are green in icinga. Change-Id: I324134438955c742bfbfd398d2f3373036c7c87a 27 October 2015, 20:20:00 UTC
a2a5a58 Harden grafana setup grafana is configured to authenticate users based on the value of the X-WEBAUTH-USER. The header is set in Apache: for grafana-admin.wikimedia.org, it is set to the LDAP CN of the user by mod_authnz_ldap; for grafana.wikimedia.org, it is hard-coded to 'Anonymous'. Since in both cases the request header is unconditionally set, any X-WEBAUTH-USER coming from the user would get overwritten, which is what we want. But we can reject user-supplied X-WEBAUTH-USER headers a little more explicitly by doing it on the varnish layer as well and by explicitly unsetting it in the Apache config before setting it to its final value. Change-Id: I78683567240a355e1b44a6aecbc8a04e9b459c2a 27 October 2015, 19:56:50 UTC
4ced387 Publish the new pageviews dataset to dumps For now, I'm not adding this to the html page that describes what's available. It will be usable by people who know about it, but I think we have to think more about how to explain this to users. We have 6 different sources of data available and just adding one more link might break the proverbial camel's back :) Change-Id: Ie6ace1d3bb84e261e6407b9a9b4fc1c32e346bc5 27 October 2015, 19:12:09 UTC
90e21b8 install/deploy scandium as zuul merger (ci) server adding to site.pp for updates T95046 Change-Id: I7d68afda55bf5b64e898f1898def539be1a97642 27 October 2015, 19:04:43 UTC
d8f2073 Move dsh code into scap where it belongs Nothing uses dsh except scap and our dsh setup is scap-specific anyway. Removes the top-level dsh module. dsh::config is now scap::dsh (for icinga). Removed the other class since all it did was install dsh in one place...just do it in scap::master itself Change-Id: Ib66c162504e5eb7949e63e5fe2e4f67646905d18 27 October 2015, 18:59:47 UTC
f525ee2 Update camus runs Add run parameter to match script update Add check parameter to webrequest job Bug: T113252 Change-Id: I5db0f95859ca52ee2cf4f40ead88f68c66607e5a 27 October 2015, 18:50:13 UTC
60d6de2 setting scandium install params scandium slated for use as zuul merger (ci) server T95046 Change-Id: I67bb07c816c20d224f40c0fc5c0ff80b6a64e94b 27 October 2015, 18:33:58 UTC
1902657 Improved the 'buggy kernel' error message for labvirt nodes. Change-Id: I181e31844e368f29ca2b948d8faf286062f037bc 27 October 2015, 06:12:36 UTC
aab1db0 Make labvirt1011 a compute node. Change-Id: Ib9fa42b44b6f5f8f94fe815fa3b7894077200bdf 27 October 2015, 17:43:02 UTC
90fb6c4 Use force => true for /usr/share/diamond/collectors/$name diamond collector This will allow puppet to remove collector directories that have ensure => absent Change-Id: I92a8b326e3d5589f72003ff8a4af7e54ff339713 27 October 2015, 17:37:00 UTC
5f9893d Use Requires, not Require, for varnishreqstats systemd Change-Id: Idb3b50074979b1d704e7c19b46dc33ea1fc44b32 27 October 2015, 17:34:37 UTC
edc3324 Add env python at top of varnishreqstats script Change-Id: Ibb4fd8a99aa9c5832421016228bfe347c8bb79a5 27 October 2015, 17:28:58 UTC
e668a43 varnishreqstats: fix instance_name corner case stuff In the long run we'll escape these problems by having uniquely defined non-empty instance names for all varnish daemons. For now, using the empty-string in place of "undef" to get at the default varnish service instance works correctly even if it's ugly. With the "undef" usage before, undef triggered defaulting of $instance_name back to $name as 'frontend' in the misc-cluster case. Change-Id: If71625a7727ed3a140208a22ef558f77ce6c09b7 27 October 2015, 17:23:35 UTC
f4814be monitoring: append sms to contact groups, don't override with admins,sms Change-Id: I9652ada7eaee2e6604cc6ba07dd5676fb593f286 27 October 2015, 17:15:59 UTC
5ab78ad lawrencium uses h710 controller hw raid swapping the h310 means the h710 can just do a hw raid1, changing the partman selection used T116645 Change-Id: I6b6e9b47bc43df3ecb7799503980229318cb9d3f 27 October 2015, 17:09:38 UTC
5e2d93c openstack: cleanup up old repo setups We are solid on trusty/kilo now Change-Id: Ice2e75e49887148f09296e012c0d418a9d35c5a9 27 October 2015, 17:05:26 UTC
d95c2af setting lawrencium dns entries lawrencium will become a new salt master in eqiad T116645 This reverts commit e3b810bc175dc2ad6dca0519012ff0b51f1bf0dd. Change-Id: I84a6609d13e49062557f67510785f489b71382c8 27 October 2015, 16:53:57 UTC
f64ff16 Revert "dynamicproxy: Make blocked user agents configurable" Despite appearances, this sets the UA check regex to an empty string. This reverts commit edb5baa8d906e16ef878791569c78d5c4298c7ba. Change-Id: Ib56f3cbed1da6de0267579a12ef83808ebc18c82 27 October 2015, 16:49:56 UTC
c7d9dc9 dynamicproxy: Empty data from initial-data.db Gets rid of a load of pmtpa references. I think this should contain the table schema only. Change-Id: I17fd303f9ea59d5f909a97a9ab79d466b216a172 27 October 2015, 16:42:04 UTC
99d4ba1 cassandra: switch to monitor_systemd_unit_state cassandra has been switched to systemd units, switch also service monitoring Change-Id: I2c909cd1dd46b7d62952d4d07be82a5b6162c356 27 October 2015, 16:41:54 UTC
46f2b0d Set instance name specifically for varnish misc Change-Id: I735c066a83d6041c32e0c5324e33a2afdc0cb67d 27 October 2015, 16:34:09 UTC
af41c53 Fix varnishreqstats dependency on varnish service Change-Id: I9477be7c9be98b2e72646f002f35fbcdd9e827c6 27 October 2015, 16:29:35 UTC
35be29e phabricator: Set security ext tag for labs Installs security extension in labs by specifying git tag. Bug: T104904 Change-Id: I4773cac123557b827a0ad368de32a03e7ab7e617 27 October 2015, 16:24:00 UTC
dd90787 wikitech: make private settings file writable by owner We set the permissions for this file to be equal to the ones of the other private files we have in /srv/mediawiki/private, so that they can be formally synced by scap when working in multi-master mode. Bug: T87036 Change-Id: I87f21996b106797f89964bbf2335f4ce9e54252a 27 October 2015, 16:21:30 UTC
fd03063 Rename mediawiki::web::sites to mediawiki::web::prod_sites to make room for a new generic sites.pp Bug: T86644 Change-Id: I2978b697e3c0145b2d331802532b2bd3ee43bbd5 27 October 2015, 16:19:34 UTC
edb5baa dynamicproxy: Make blocked user agents configurable Untested Bug: T90844 Change-Id: Ia739e81e25873095f47751a57fb7abfd889b5f89 27 October 2015, 16:12:56 UTC
408d481 Fix param name for reqstats Change-Id: I18b270231b32f5aeefc5b75f63cb2a4120f15c20 27 October 2015, 16:00:20 UTC
0c2c477 The varnish reqstats diamond collector does not work, emit to statsd directly instead This also adds a main statsd hiera variable for general use. Bug: T83580 Change-Id: Ife12f04c2b12f582a2338cc5791de1e75ef9114a 27 October 2015, 15:53:56 UTC
7562ddd otrs: omit the --force argument to otrs.Daemon.pl It is not a valid argument in start action and does not really do much in stop apart from reducing a timeout before supposedly SIGKILLing the module daemons Change-Id: Ia7503484541fcdf4c6975f1e3aa602c276c9a896 27 October 2015, 15:31:38 UTC
845f25d otrs: Ship systemd unit file for OTRS Daemon OTRS 5.0.1 expanded more on the concept of the OTRS scheduler and improved into an all generic Daemon. Remove the scheduler systemd unit file present in OTRS 4.0.x and replace it with the OTRS daemon. The daemon is also responsible for all cron jobs done in the past by the various cronjobs so remove all of those as well Change-Id: I789dd446efc5d5e20199d3cc50c9b0058b2863ed 27 October 2015, 15:25:44 UTC
9a5471e HTCP: split multicast for cache clusters This assigns our traditional singular 239.128.0.112 to the text/mobile clusters (no-op for them in practice) and splits out .113 and .114 for upload and maps respectively. This configures maps to listen only to its new address, as it had no working HTCP before. Upload is configured to listen both .112 and .113 for now, so that it works through the transition of the MW side of things (after which we'll remove .112 from its list). Bug: T112836 Bug: T116752 Change-Id: I003dee670cf87d35b9b39104af6a28f5055871dd 27 October 2015, 14:46:57 UTC
490b194 Setting max_allowed_package to 32MB because otrs requires it Change-Id: I331bc2e3001a073d2af69aa7b6a5785a3b299e1c 27 October 2015, 14:42:33 UTC
cedd4e3 cassandra: fix restbase-test2003-b JMX port Change-Id: I14a08ebb78f8b09cd29e3eedd365ebeb328dded3 27 October 2015, 14:40:37 UTC
f789c42 vhtcpd: refac args template, allow multiple mc addrs Bug: T116752 Change-Id: Icc0497e2fe8330ed5a451ecc384bd78e719816e0 27 October 2015, 14:33:41 UTC
0332151 cassandra: add restbase-test2003-b instance Change-Id: Ic787fbd9e44fb886f897c1682bd57449584b03ce 27 October 2015, 14:26:31 UTC
eaad9c4 Revert "Revert "Replicate pt-heartbeat table to labs. Stop replicating msg_resource"" This reverts commit 2320791be293856849e58709570305f400ac48b2. Change-Id: Ida36bafa2d7a61a545631b5caee7a7af5b99655f 27 October 2015, 13:48:19 UTC
9397064 rm files/misc/apt-security-updates Does not appear to be used by anything. "misc"-cruft? Change-Id: I0aa7e16b1219b0f0744f7cc0cb6882a6ada28e02 27 October 2015, 13:16:46 UTC
2320791 Revert "Replicate pt-heartbeat table to labs. Stop replicating msg_resource" Was never puppet-merged. This reverts commit 97cd5c5a53b062c36eae12490ffabe424b1d14a1. Change-Id: I1dc8cdd13d82f6d0fe0e3705c6c4eafc20cedf44 27 October 2015, 13:13:03 UTC
d66608a Add missing Hiera data Change-Id: Ib5b4c71da827ce9c142b58368a1e185ba25648bb 27 October 2015, 12:34:39 UTC
97cd5c5 Replicate pt-heartbeat table to labs. Stop replicating msg_resource The heartbeat database was filtered on sanitarium. I have applied manually this changes to allow the pt-heartbeat table (heartbeat. heartbeat) to reach sanitarium. I have also allowed it on labs. I have changed other configuration changes, like increasing the max_allowed_package, activating the buffer_pool dump feature, and I have set msg_resource as a private table. This requires a followup change on labs to fix the script that creates the views on labs- creating a custom view for heartbeat and deleting the one for msg_resource. Bug: T116720 Bug: T71463 Change-Id: Idd65d4cc18f43a35c6d2bfa24560f3d271279724 References: T114752 27 October 2015, 12:22:48 UTC
bd74d1c maps: Tune replication parameters There has been an instance of a query breaking postgres replication. This is covered in T116553. Using the numbers from that and extrapolating to using around 30GB for WAL segments for slaves to use we set checkpoint_segments and wal_keep_segments to 768. This should cover us even in the case were really problematic queries are being run on the cluster at the cost of keeping around extra WAL files. We do follow a similar approach in MySQL keeping around a lot of extra WAL files. This adds the extra benefit of allowing even quite out of date slaves to catch up to the master after a period of extended downtime. That being said, this should not be considered a silver bullet Bug: T116553 Change-Id: Ic13cd1366662fdc17c49abadf61ffaa72c119fdf 27 October 2015, 11:59:12 UTC
73429e8 cassandra: add restbase-test2003 instances restbase-test2003-b will come up later Change-Id: If4dc551937fedf3b3fd54953d84dda967db185fe 27 October 2015, 11:46:42 UTC
534deab maps: some puppet-lint fixes Change-Id: Ic0bb14d824985923eceafb1d6e2f66d9195d0d60 WARNING: class not documented on line 4 WARNING: class not documented on line 45 WARNING: class not documented on line 90 WARNING: double quoted string containing no variables on line 1 WARNING: double quoted string containing no variables on line 1 WARNING: double quoted string containing no variables on line 2 WARNING: double quoted string containing no variables on line 2 27 October 2015, 08:36:56 UTC
ae4135d add apps guidelines to legal info page on datasets Change-Id: If217c430e4f6be30bf29007bb2ba8155838a0a33 27 October 2015, 05:35:03 UTC
6283cf7 add apps development guidelines to legal text for dumps Change-Id: I100f2f5a32d8c45fa25009dd96d74c5553a4e4bb 27 October 2015, 05:20:14 UTC
a4d4508 mailman: remove check for out queue [+data cron] Provides nothing useful, regular increases make it impossible to accurately detect an increase. in, bounces and virgin sufficiently handles the need anyway. Remove the cron on fermium too from root crontab. Change-Id: I5c25b2f00087ce07457452233ffc0cdb90a9851c 27 October 2015, 00:35:23 UTC
42e0d28 ssh: Allow customizing authorized_keys_command Bug: T113979 Change-Id: Iee5197ef33abde4c6c0002c0129f0931bc9ab1f5 27 October 2015, 00:04:59 UTC
d8f1b8f ssh: Allow direct login as servicegroups Can be turned on with ssh config if necessary. Isn't actually turned on anywhere atm, since it requires ssh 6.9 to be able to specify parameters Simplifies direct user lookups too Bug: T113979 Change-Id: Iec3b944cfd0cf095cbc22554b5a8ba99d1262510 26 October 2015, 23:49:33 UTC
20bd8ec tilerator/k10n: add trailing * to journalctl sudo Add a trailing * to the sudo permissions line that allows reading logfiles with journalctl. The intention was that it ends in a wildcard to allow variations of the command. We just wanted to limit on the user. Compare to other admin groups for other services with a journcalctl line. quote from Yurik: " does not allow for any other arguments, such as -a, -r, --no-pager, -x, -o, -n, -f, -e, and possibly others. Which means that it opens a pager at Aug 13th, and if I try to jump to the end with "shift+G", it hangs." and not having this was an oversight. Bug:T115067 Change-Id: I8919a96db3f2f0e74412571555d6e48076dcf1ca 26 October 2015, 23:32:29 UTC
e3b810b Revert "setting lawrencium install params" System has an H310 which Jessie wont work with, returning to spares. This reverts commit 90d95a2c7a2cf82be7b55fb75d6421ef62d82f0d. Change-Id: I5fe10dd43fb36f933b0f2072a6d20c73840dd07a 26 October 2015, 23:07:28 UTC
9b052ab Enable the wdqs service per default No point in not having that. Bug: T116673 Change-Id: I6b82a5f0b057a77f7ff031d112d442e3af43a8fb 26 October 2015, 22:17:14 UTC
d6f96f4 admin: add papaul to datacenter ops group Change-Id: I2a64c98902f45e1d02074ae607a533aff7131526 26 October 2015, 21:03:35 UTC
782e34f admin: add datacenter-ops to palladium Originally i wanted to use a role to add this group to nodes (Id553cc38820e6d7b), but that doesn't work (Ib75956186d9c0a5). So for now i'm just adding the group by hostname. And access on iron wasn't existing anyways, so this is actually correct. Bug:T115718 Change-Id: Iee4eeb4d10644ab1a0d72256ecf84a592b05cd38 26 October 2015, 20:45:18 UTC
5997a2c interface: do not 'ensure latest', use require_package Do not let Puppet automatically update these fundamental networking-related packages. While at it, use require_package and drop the singleton classes. Bug: T115348 Change-Id: Ifa690317111326841bca1376cc7e1fcfd08ab639 26 October 2015, 20:33:18 UTC
bf42025 puppet: do not 'ensure latest' Do not let puppet automatically upgrade itself. Bug: T115348 Change-Id: Ic4b14952b3d5f3847bf3588539f83c02cde2ce12 26 October 2015, 20:21:17 UTC
2668e86 Mark rubidium as spare Was formerly used as a DNS server, now unused. Change-Id: I5e020ccc0997f019a578ca0db7a0c0e3b7a9740f 26 October 2015, 20:18:38 UTC
3525a2f Use testsystem role for ruthenium Change-Id: I10950fc761617be5ebbc10972f50de1cdeedf9f9 26 October 2015, 20:17:19 UTC
90d95a2 setting lawrencium install params lawrencium to become eqiad salt-master T116645 Change-Id: Ifa2b16b4df72f2075cc10fb32021931aa805d421 26 October 2015, 19:32:24 UTC
2708748 openstack: Remove havana/icehouse files These directories contain pmtpa references which will no longer work, plus I don't think they're used anymore. Change-Id: I0d046de0237cc80510f4cb784bd1e5a2c0906499 26 October 2015, 18:09:15 UTC
7a80749 adding in globalsign to procurement approved vendors adding globalsigns email domain to allowed direct emails into procurement tasks Change-Id: I4f41b21ffc3a240d77655412a17028e8c17f0ba9 26 October 2015, 17:35:10 UTC
back to top