infrastructure/ansible
1
0
Fork 0
Code Pull requests 7 Activity Actions
38658 commits 9 branches 0 tags 110 MiB
Commit graph

38658 commits

This branch
This branch
All branches
Author SHA1 Message Date
František Zatloukal
d04e46c8dc Blockerbugs: temp testing of releng requests on prod 2023-04-04 17:40:43 +02:00
Kevin Fenzi
36b489bce2 haproxy: adjust content size to 503 page 
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-04-04 07:50:02 -07:00
Kevin Fenzi
bbce51960d buildhw-x86: all these hosts are 10G and should use a 9000 mtu 
The last time they were re-installed they defaulted to 1500 and it's
causing some problems for some of them at least (01-08 for sure).

Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-04-03 14:47:31 -07:00
Kevin Fenzi
3e73c99574 public-db-copy: correct datanommer db name 
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-04-03 13:19:39 -07:00
Miroslav Suchý
843930005b partially revert aws: allow to use gp3 storage type for fedimg 
partially reverting 034cf953db
 2023-04-03 20:22:41 +02:00
Francois Andrieu
0ece2dfe06
use fedoraproject favicon everywhere it's needed 2023-04-03 13:35:55 +02:00
Francois Andrieu
a4570c90c2
websites: remove favicon alias on fedoraproject.org & fix 404 2023-04-03 13:35:52 +02:00
Francois Andrieu
c1816caed8 websites: add fedoraproject.org alias for Fedora Media Writer 2023-04-02 17:56:21 +00:00
Francois Andrieu
b85d421855
websites: drop fedoraproject.org/sponsors redirect 2023-04-02 16:31:39 +02:00
Kevin Fenzi
6030f92f4a db-datanommer: some more backups adjustments 
First we need to pipe stderr into the grep to filter out the timescaledb
warnings. So, |& does that.

Then, there's no reason to backup the staging database. Disable that.

Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-04-01 11:23:01 -07:00
Miroslav Suchý
034cf953db aws: allow to use gp3 storage type for fedimg 
this is cheaper than gp2 see
https://fedoraproject.org/wiki/Changes/CloudEC2gp3
 2023-04-01 10:29:16 +02:00
Kevin Fenzi
36eab9026b rkhunter: exclude the apache balancer shm files 
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-03-31 21:10:37 -07:00
Kevin Fenzi
87671c627b db-datanommer02: supress warnings from pg_dump 
db-datanommer02 uses timescaledb. When you do a pg_dump there's warnings
due to this, but according to upstream they are all completely harmless.
So, to avoid an email to everyone every day, lets just try and supress
these, but yet hopefully not supress real errors if they every occur.

Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-03-31 12:24:33 -07:00
Francois Andrieu
f5315b3ca4
websites: need to specify the dockerfile path 2023-03-31 19:12:15 +02:00
Francois Andrieu
be32dcebb4 websites: new main website deployment on prod 2023-03-31 16:58:12 +00:00
Francois Andrieu
6b596e84e2 websites: use upstream dockerfile 2023-03-31 16:58:12 +00:00
Kalev Lember
6d8bd22987 fedmsg / irc: use templates for Branched version 
This avoids having to update it again next cycle. Thanks for suggesting
it, Adam Williamson!

Signed-off-by: Kalev Lember <klember@redhat.com>
 2023-03-31 18:23:36 +02:00
Kalev Lember
28b03dbed0 fedmsg / irc: update #fedora-releng filtering for current Branched version 
Update the filter so that we get messages about current Branched, which
is 38 right now.

Signed-off-by: Kalev Lember <klember@redhat.com>
 2023-03-31 15:30:22 +00:00
Lukas Holecek
730a03992b waiverdb: Fix getting username from OIDC token 2023-03-31 15:12:06 +00:00
Aurélien Bompard
5c0ab4d9d0
Revert "messaging-bridge: leverage the rabbit/queue role and add monitoring to the queues" 
The messaging bridge queues have very specific setup, we can't use the
rabbit/queue role because it binds all queues to both amq.topic and
zmq.topic and we don't wan't that for the bridges.

This reverts commit 649eec104d.
 2023-03-31 14:59:02 +02:00
Michal Konečný
ae6c2b6508 [postgresql_server] Grant datanommer_ro access to datanommer2 db 
The datanommer_ro user was created in the task, but never got privilege to read
from datanommer2 db. This commit is fixing that.

Signed-off-by: Michal Konečný <mkonecny@redhat.com>
 2023-03-31 08:25:58 +00:00
Kevin Fenzi
8b3d4cd54e waiverdb: remove the playbook call to the now removed secrets 
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-03-30 13:33:19 -07:00
Lukas Holecek
c21081d0d5 waiverdb: Remove unused file 2023-03-30 20:27:33 +00:00
Kevin Fenzi
af2ddc228c db-datanommer02: switch backups to 02 and copy public db from there too 
We switched a while ago to db-datanommer02 which is a timescaledb from
the old db-datanommer01. However, backups were not working right until
recently. Now that they are, switch backups to backup 02 instead of 01
and also copy that one public. This should allow us to retire
db-datanommer01 entirely now.

Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-03-30 11:23:14 -07:00
Pavel Raiskup
76cf843c9b copr-backend: fixing ssh_config, take #4 (ownership) 2023-03-30 10:57:55 +02:00
Pavel Raiskup
c82af101e0 copr-backend: fixes in ssh_config, take #3 2023-03-30 10:56:53 +02:00
Pavel Raiskup
4c22f7011b copr-backend: fix whitespace errors in ssh_config, take #2 2023-03-30 10:53:15 +02:00
Pavel Raiskup
044d0bab36 copr-backend: fix newline in ssh_config 2023-03-30 10:49:38 +02:00
Pavel Raiskup
1bbb69261d copr-backend: ssh_config is actually a template 2023-03-30 10:44:40 +02:00
Pavel Raiskup
96f1d5184a copr-backend: prefer ssh over ipv4 to copr_hypervisor 2023-03-30 10:38:28 +02:00
Kevin Fenzi
3fc94c29fc bastion01: move to rhel9 
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-03-29 17:09:39 -07:00
Adam Williamson
29395d0c96 openqa: report queued 'results' on prod too 
It seems to work OK on staging, so let's try it out on prod and
see how it looks in Bodhi.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
 2023-03-29 08:47:14 -07:00
Kevin Fenzi
30ae0ea40b dl: move 04/03/02/01 all to rhel9 
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-03-28 17:06:36 -07:00
Kevin Fenzi
c14873010d ipsilon02: move to f37 
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-03-28 16:48:24 -07:00
Kevin Fenzi
4a90c1a152 Revert "batcave02: add a new batcave02 that is rhel9" 
This reverts commit a195bce0b3.

We need to get fedora-messaging into epel9 and then we can try this
again.
 2023-03-28 15:20:01 -07:00
Kevin Fenzi
52ec43c2a9 bodhi: set version correctly for openshift to 7.1.1 
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-03-28 15:02:53 -07:00
Kevin Fenzi
a9bb7f2254 batcave02: no ansible-collection-community-docker right now, so exclude it on rhel9 
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-03-28 14:32:44 -07:00
Kevin Fenzi
a195bce0b3 batcave02: add a new batcave02 that is rhel9 
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-03-28 14:16:22 -07:00
Kevin Fenzi
5143b86c51 ns05: fix resolv.conf 
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-03-28 13:38:25 -07:00
Aurélien Bompard
ca5eda3e3a
FMN: keep using Sendria on staging 
Signed-off-by: Aurélien Bompard <aurelien@bompard.org>
 2023-03-28 18:37:13 +02:00
Lukas Holecek
38a5336a2c waiverdb: Configure token issuer 
Fixes the following error (in stage):

    flask_oidc ERROR id_token issued by non-trusted issuer:
    https://id.stg.fedoraproject.org/openidc/
 2023-03-28 16:19:56 +00:00
Kevin Fenzi
c69fbc43f7 nagios: add monitoring for stg pagure ssl cert 
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-03-28 09:13:54 -07:00
Tomas Hrcka
597f85aa21 Retire epel8-modular 
Signed-off-by: Tomas Hrcka <thrcka@redhat.com>
 2023-03-28 15:09:06 +00:00
Jiri Kyjovsky
183fedafe1 copr-dist-git: specify default_branch option in dict-git.conf 
commit d9dfe63e34abd7d6fe3c9b6e4c70939f4932f250 in dist-git
introduces configuration of main branch name. Thus it should
be specified in dict-git.conf
 2023-03-28 15:01:39 +02:00
Dan Fandrich
078cde0378 [release-monitoring] Add more disto package links 
This time, drop Adélie Linux as it's the most likely reason that the
previous attempt failed.
 2023-03-28 12:36:58 +00:00
Aurélien Bompard
4dc57fbfa0
FMN: actually send emails 
Signed-off-by: Aurélien Bompard <aurelien@bompard.org>
 2023-03-28 11:22:50 +02:00
Kevin Fenzi
b85fc1665c bastion02: move to rhel9 
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-03-27 16:42:15 -07:00
Kevin Fenzi
9a26c66413 vhost_reboot: it is libvirt-sock, not libvirtd-sock 
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-03-27 15:24:31 -07:00
Kevin Fenzi
1b4443a8ee vhost_reboot: fix path 
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-03-27 14:51:10 -07:00
Kevin Fenzi
1eeb585f74 vhost_reboot: adjust libvirt pid to libvirt socket 
On newer libvirt there's no pid file, instead there's a socket that
listens and then the service is started. So, check that the socket is
there instead of looking for a broken pid file.

Signed-off-by: Kevin Fenzi <kevin@scrye.com>
 2023-03-27 14:27:55 -07:00