Infrastructure/ansible
1
0
Fork 0
Code Issues Pull requests 6 Projects Releases Packages Wiki Activity Actions
38447 commits 9 branches 0 tags 111 MiB
Commit graph

38447 commits

This branch
This branch
All branches
Author SHA1 Message Date
Mikolaj Izdebski
8593bea6ae Koschei: remove rewind-unapplied-changes cronjob 
This job was not functional as it needs to be ran as db admin (due to
its use of ALTER DDL command), but was ran as ordinary db user.
 2023-02-13 09:17:26 +01:00
Mikolaj Izdebski
cbfa7d32e2 Koschei: fix mixed staging/prod db passwords 2023-02-13 08:58:12 +01:00
David Kirwan
34a70319ca fas2discourse: change git checkout of operator to discard local changes 
Signed-off-by: David Kirwan <dkirwan@redhat.com>
 2023-02-13 16:08:13 +09:00
Pavel Raiskup
af2507f1bf copr-fe: nagios -> prometheus: WARNING state == 50% 2023-02-10 15:22:40 +01:00
Pavel Raiskup
001f60b8bd copr-fe: prometheus: get the data from nagios more frequently 
But do not fail immediately if we can not get the results from Nagios.
 2023-02-10 14:58:44 +01:00
Pavel Raiskup
6d82fba602 copr-fe: more obvious script name than just "monitoring.py" 
The cron job doesn't need to "mv" stuff here and there, the script
itself now directly writes to the right file.
 2023-02-10 14:46:10 +01:00
Pavel Raiskup
cacaec9c8e copr-fe: prometheus monitoring to gather the actual state 
Not the "historical" info which is calculated into the percentage
of time service spent in UP state.
 2023-02-10 14:44:58 +01:00
Pavel Raiskup
39070dc595 copr-fe: catch exceptions while checking uptime robot 2023-02-10 14:44:58 +01:00
Kevin Fenzi
17a83f33e2 robosignatory: fix the boneheaded mistake that took me hours to find 
When I pushed this config, I forgot to change fedora-39 to fedora-38.
This amusingly caused signing to fail in a completely non obvious way.

Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-02-09 15:42:35 -08:00
Kevin Fenzi
a1463e1b3e robosignatory: fix up config for f38/f39/eln 
Added some comments, made sure all releases had the needed config.

Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-02-09 11:56:56 -08:00
Francois Andrieu
f450173400 ocp_monitoring: enable alerting by default 2023-02-09 17:57:39 +00:00
Pavel Raiskup
073d940775 copr-backend*: copr-ping: nicer info message 2023-02-09 17:01:53 +01:00
Pavel Raiskup
0613f46b41 copr-backend*: copr-ping: work with rounded seconds 2023-02-09 17:00:19 +01:00
Pavel Raiskup
6e9d9a5bd9 copr-be*: copr-ping: dump to the nagios web-UI elapsed time 2023-02-09 16:58:08 +01:00
Pavel Raiskup
e37d20a00d copr-be-dev: run the ping script on devel with the same cadence 
That script is a no-op, so it doesn't hurt the performance nor the cloud
budget.
 2023-02-09 16:55:53 +01:00
Pavel Raiskup
b8f9517b4b copr: fix the copr-ping checker 
Go "green" as soon as possible.  Consider only the last build.
Fail also if no builds are submitted for a long time.

Fixes: https://github.com/fedora-copr/copr/issues/2355
 2023-02-09 16:47:04 +01:00
Ryan Lerch
732e303232 Properly get the plugin logging messages to httpd errorlog 
Signed-off-by: Ryan Lerch <rlerch@redhat.com>
 2023-02-09 12:25:18 +10:00
Adam Williamson
c3974fc8c5 openqa: test reporting non-complete 'results' on lab 
Signed-off-by: Adam Williamson <awilliam@redhat.com>
 2023-02-08 15:55:18 -08:00
Kevin Fenzi
f391d8dd45 robosignatory: fix typo and add resigning 
There was a typo and we were resigning eln with f38 ima key instead of
f39 one. ;( Fixed

Also added f38/f39 retag signing so we can get everything signed with
the right keys.

Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-02-08 13:29:47 -08:00
Tomas Hrcka
cbd4238536 Fix pkgdb gnome entries 
Signed-off-by: Tomas Hrcka <thrcka@redhat.com>
 2023-02-08 21:23:26 +01:00
Tomas Hrcka
ce8a92b311 Uncomment f37 tags 
Update f39 key

Signed-off-by: Tomas Hrcka <thrcka@redhat.com>
 2023-02-08 20:53:04 +01:00
Samyak Jain
549e5d3ace Branch Fedora 38 from rawhide 
Signed-off-by: Samyak Jain <samyak.jn11@gmail.com>
 2023-02-08 20:30:05 +01:00
Michal Konečný
7e8bf26944 [the-new-hotness] Sync the staging changes with production 
With version 1.2.2 deployed on production this needs to be done for bugzilla
ticket to work.

Signed-off-by: Michal Konečný <mkonecny@redhat.com>
 2023-02-08 16:10:29 +01:00
David Kirwan
154f97555a fas2discourse: Deploy operator task 
Signed-off-by: David Kirwan <dkirwan@redhat.com>
 2023-02-08 16:21:53 +09:00
David Kirwan
b7e4fdf852 fas2discourse: Clean whitespace, fix permissions on keytab 
Signed-off-by: David Kirwan <dkirwan@redhat.com>
 2023-02-08 15:08:07 +09:00
Kevin Fenzi
337285c85f autosign: switch eln to sign things already in it's tag with f39 key 
This will allow us to resign everything with this new key.
Normal builds shouldn't be using it as they come via bodhi.

Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-02-07 16:49:02 -08:00
Kevin Fenzi
0801e0fdd5 pkgs01: sync ansible up with reality on memory and switch procs to 8 to try and get a bit better performance 
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-02-07 16:47:08 -08:00
Adam Williamson
24d1f467f3 greenwave: prepare for critical-path-base group split 
See https://pagure.io/fedora-comps/pull-request/797 - we need the
greenwave policy ready for that change before we land it, so updates
are gated on the correct tests and Bodhi doesn't crash because
there is no policy for the new contexts.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
 2023-02-07 15:58:46 -08:00
Kevin Fenzi
e9a8583570 security.txt: move expires out 
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-02-07 13:56:13 -08:00
Kevin Fenzi
d2c0673c1c linux-system-roles/seconday01: need user=root to find the right cron 
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-02-07 13:50:04 -08:00
Jiri Kucera
f49f4a7ffd lsr-logs-clean: Add MAILTO to cron file 
Set MAILTO for the particular cron file to the email address
of the Linux system roles community so any output from the log
pruning job going to stderr is reported to them. Send stdout to
/dev/null since it is not important.

Signed-off-by: Jiri Kucera <jkucera@redhat.com>
 2023-02-07 16:39:46 +01:00
Jiri Kucera
e207a4cdf9 lsr-logs-clean: Normalize image names 
Normalize RHEL image names to RHEL-X.Y. Since RHEL images are
produced quite frequently it is sufficient to keep just one
RHEL <major>.<minor> after the log expires.

Also add contacts to Linux system roles maintainers.

Signed-off-by: Jiri Kucera <jkucera@redhat.com>
 2023-02-07 16:39:46 +01:00
Jiri Kucera
dc8c275f09 Clean /srv/pub/alt/linuxsystemroles/logs 
Add cron task that prune /srv/pub/alt/linuxsystemroles/logs.

Related: RHELPLAN-104620

Sign-off-by: Jiri Kucera <jkucera@redhat.com>
 2023-02-07 16:39:46 +01:00
Ryan Lerch
5fd4fc49a9 kojihub staging: fix my linting issues 
Signed-off-by: Ryan Lerch <rlerch@redhat.com>
 2023-02-07 20:08:56 +10:00
Ryan Lerch
a91b34f9b9 kojihub staging: use the new fm plugin and schemas 
Signed-off-by: Ryan Lerch <rlerch@redhat.com>
 2023-02-07 19:58:20 +10:00
Kevin Fenzi
29dcdd6ce5 proxies: new *.id.fedoraproject.org cert 
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-02-06 11:47:26 -08:00
Kalev Lember
9853012f58 fedmsg / irc: drop #fedora-flatpaks bot 
We have tested it for a week now and it turned out that the bot's
verbosity was effectively killing communication in the channel.

Signed-off-by: Kalev Lember <klember@redhat.com>
 2023-02-06 17:24:52 +01:00
Pavel Raiskup
2b4e3c2f0e copr: incremental backup: preserve hardlinks 
Relates: https://github.com/fedora-copr/copr/pull/2526
 2023-02-06 10:34:04 +01:00
David Kirwan
f65c1808de fas2discourse: Correct host target 
Signed-off-by: David Kirwan <dkirwan@redhat.com>
 2023-02-06 14:06:34 +09:00
David Kirwan
a21dfeea4f fas2discourse: structure to deploy the fas2discourse-operator 
Signed-off-by: David Kirwan <dkirwan@redhat.com>
 2023-02-06 13:22:22 +09:00
Francois Andrieu
953f2e5c4a ocp_monitoring: fix cronjob alert rule 2023-02-05 22:52:10 +00:00
Francois Andrieu
d0704659a7 docsbuilding: set alert_users list 2023-02-05 22:49:44 +00:00
Francois Andrieu
1304f93460 ocp_monitoring: allows a separate list of alert recipients 2023-02-05 22:49:44 +00:00
Francois Andrieu
7e979cb2f7 communishift: set default resource limits for pods 
Once a resource quota is set for a namespace, kube will refuse to
schedule any pod without limits set, including build pod.
This can be difficult to figure out unless you know where to look, and
can be challenging for new openshift/kubernetes users.
Setting a default limit would, at least, avoid the non-schedulable
issue.
 2023-02-05 22:46:55 +00:00
Pavel Raiskup
57065bd9b0 copr: incremental backups: documentation 2023-02-05 12:32:12 +01:00
Pavel Raiskup
182dce43e8 copr: incremental backup: tweak periods once more 2023-02-05 12:15:58 +01:00
Pavel Raiskup
9ffbaef878 copr: incremental backup: do backup twice a week 2023-02-05 12:12:41 +01:00
Pavel Raiskup
b771f93a53 copr: incremental backup: don't spam /var/log/crond 2023-02-05 11:43:46 +01:00
Pavel Raiskup
f146b38572 copr: incremental backup: don't eat too much I/O 2023-02-05 11:34:47 +01:00
Pavel Raiskup
6dac8b3a1b copr: incremental backup: more readable rsync progress 2023-02-05 11:27:28 +01:00