We don't have permissions to write to /run when running on OpenShift so let's
just use /tmp and change the filename to prevent any conflicts.
Signed-off-by: Ondřej Budai <ondrej@budai.cz>
Currently liveness and readiness was treated the same. However, their
behaviour at shutdown is meant to be different. When a service is not read
no new connections are made to it, and when a service is not live it can be
cleaned up.
By considering our service live if and only if it listens to HTTP requests we
don't have the opportunity to clean up after we stop listening to new requests.
Leave readiness probes as they are, and instead use a file in the filesystem to
indicate when the service is live. It is created before composer is spawned and
deleted once composer exits.
This will allow us to use the service accounts which work against
identity.api.openshift.com. These are much easier to manage, especially
with the new multi-tenancy, as there's a single page to create/expire
them across an account.
They also have the added benefit of not expiring automatically when
they're not used like offline tokens, and immediate expiration when
desired.
Tidy up the queries for the composer dashboard
and making them more readable in grafana. Additionally
add some fallback values for when empty query results
are returned from prometheus.
The decision logic which jobs to run is quite confusing but that's how we
roll for now:
Jenkins builds RHEL images only on main
Schutzbot builds RHEL images only in PRs
Schutzbot builds Fedora images on both PRs and on main
To achieve this, the commit re-enables running Packer on main on Schutzbot.
Signed-off-by: Ondřej Budai <ondrej@budai.cz>
Arch was easy.
For passing the repository distribution and osbuild_commit (it can be
different for each distro), I decided to go in the way of ansible
inventory directories. It adds a bit of structure but I think it's
the most clean solution.
Signed-off-by: Ondřej Budai <ondrej@budai.cz>
So you don't have to pass these if packer is supposed to find them
on its own (instance profile, local profile).
Signed-off-by: Ondřej Budai <ondrej@budai.cz>
This should save us some money. 10 GB is the size of the underlying
RHEL 8.5 AMI so this should be the minimum.
Signed-off-by: Ondřej Budai <ondrej@budai.cz>
The prometheus queries have been updated with
the correct namepsace for the job metrics
Additionally, this commit fixes some of the
queries to add fallback values when the
query results are returned empty.
Because of the way the gcp secrets are stored for the workers, and how
the mapping from vault to openshift works (unable to map a multiple key
secret into a single json file), there's a bit of juggling required to
get the gcp credentials in the right format.
