swm: package -- if source is invalid we cannot check test flavours
The data needed to generate the meta mapping is stored in kernel-series.
If we have no valid kernel-series source pointer we will crater trying
to generate it. Return a null list. We will catch the bad source
linkage elsewhere and put the whole tracker into a crank-failure so it
will never be consumed in anger.
Signed-off-by: Andy Whitcroft <email address hidden>
With the introduction of LRG and LRS we have introduces the ancillary
PPAs (ppa-ps) associated with each build PPA. These are private and
allow the LRG package to remain embargoed. Add monitoring of the
`build-private` route so we can monitor these packages. Also fix the
private PPA subscription handling so that we do not have to manually
generate subscription keys for new PPAs.
Acked-by: Stefan Bader <email address hidden>
Signed-off-by: Andy Whitcroft <email address hidden>
Add support to swm-instantiate to add `kernel-block-derivatives` on all
primary kernels (those without a parent). Also improve reporting of
`kernel-block-derivative` tagged kernels so that we only report them
when something is actually being blocked.
Acked-by: Stefan Bader <email address hidden>
Signed-off-by: Andy Whitcroft <email address hidden>
iswm-instantiate: use kernel-sru-workflow task Triaged to track New trackers
When a tracker is first scanned a hard failure (such as a crank-error)
can leave the tracker missing from status.yaml. If it does not appear
swm-publishing will lose sight of it and it will never be scanned again.
Switch swm-instantiate to livening the kernel-sru-workflow task to
Triaged. swm-publishing then can unconditionally scan any tracker in
Triaged. swm can move the tracker to In Progress once it has recorded
the tracker in status.yaml; closing the loop.
Acked-by: Stefan Bader <email address hidden>
Signed-off-by: Andy Whitcroft <email address hidden>
During publication of new records we supersede any previous publication.
This is not however atomic and so we can find multiple Published
records. Take the newest as the effective publication.
Acked-by: Stefan Bader <email address hidden>
Signed-off-by: Andy Whitcroft <email address hidden>
kernel-series: Drop i386 from trusty:lts-xenial testing
The architecture was dropped from support when Xenial moved into ESM
mode. So there is no longer an i386 image being produced for the HWE
kernel either.
Signed-off-by: Stefan Bader <email address hidden>
swm-maintenance: elide positive records by default
Now that we are showing postive records (currently building etc) we need
to hide positive records by default. Also check the actual status not
the reported one when checking. Use --all to include them.
Signed-off-by: Andy Whitcroft <email address hidden>
swm-publishing: check for changes in buildstate and trigger
Now that we keep maintenance records for all builds, check them against
the current state in launchpad and trigger a scan when they do not match.
This lets us detect the end of builds which result in failure.
Signed-off-by: Andy Whitcroft <email address hidden>
Merged
into
~canonical-kernel/+git/kteam-tools:master
at
revision b596e4daf14e11300a29f461059b509688c2ab9e
