There was a placeholder for pagure user in scm_request_processor configuration.
Let's change this to correct user.
Signed-off-by: Michal Konečný <mkonecny@redhat.com>
Problem: Fedora Infrastructure has used various DNS servers in
PHX2. Those no longer exist. This is causing breakage in DNS and many
other internal services.
Fix: Work with Red Hat IT
- to get new DNS servers which are infoblox devices.
- get firewall rules fixed to that
- update named.conf to have new ips
- remove old ips which no longer exist.
Signed-off-by: Stephen Smoogen <ssmoogen@redhat.com>
systemd-oomd seems to be a bit eager and kills builds before the kernel
OOM would have. Disable it for now and see if it helps memory hungry
builds any.
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
distgit_packager_bugzilla_sync toddler doesn't work well on staging, because the
accounts from staging dist_git are not properly synced to staging bugzilla.
which is causing this toddler to run for too long and blocking the queue.
Signed-off-by: Michal Konečný <mkonecny@redhat.com>
They aren't encrypted, and it causes networking config problems.
We'd like to solve the underlying problem but we don't know how,
this is good enough for now. Also drop the workaround things
again because they don't seem to help.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
nirik did some stuff in the virthost and buildhw groups to try
and deal with a problem where network configuration created
during the initrd phase for the nbde_client role is activated
by NM during the regular system boot phase, which results in
the network configuration not being the one we actually want
and carefully set up. However, he didn't add this stuff to the
openqa-workers group playbook even though that uses the
nbde_client role too, and we sure are having the same problem
on the openQA workers. Adding it now to see if it helps.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
This reverts commit 81c4dd3a18.
Revert "flask-oidc-dev: Changing layout of client secrets configuration"
This reverts commit 81d27e668b.
Signed-off-by: David Kirwan <dkirwan@redhat.com>
We changed the global value here to none because we are behind proxies,
but some applications expected the duplicate headers. This seems to be
another one that does.
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
Upstream implemented a feature that we can use to do the same
thing using just a test variable, so we're switching to that.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
