This change simplifies the scale down logic a bit. It also make sure we
don't accidentally remove runners that are in the process of being created
when we try to consolidate.
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
This change adds some more cache helper functions, additional tests,
vastly improves memory usage when loading instances and cleans up some
code.
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
Remove calls to github after creating a runner. It just adds overhead
for very little benefit.
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
This change adds a number of fixes for scale sets:
* Reset last message ID when we need to recreate the scale set in GitHub.
Message ID gets reset in github when this happens and we end up ignoring
messages because we see that they are older than we have recorded.
* Clean up deleted instances from state scale set state
* Properly stop instance handler in the provider worker when an update
operation comes in that signals that an instance has been marked as "deleted"
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
The debug-log command now supports log level filtering and attribute
filtering. The log level filtering will only be able to set the minimum
log level as low as the server is configured to stream. If the server has
its log level set as INFO, then setting the log level in the CLI to DEBUG
will have no effect.
But anything above what the server sends, is within the control of the client
to filter. This is all done client side.
Attribute filters are useful if you need to watch the logs for a particular
worker, entity, etc.
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
On github, attempt to use the scaleset API to list all runners without
pagination. This will avoid missing runners and accidentally removing them.
Fall back to paginated API if we can't use the scaleset API.
Add ability to retrieve all instances from cache, for an entity.
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
Github will remove inactive scale sets after 7 days. This change
ensures the scale set exists in github before spinning up the listener.
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
This change adds a single page application front-end to GARM. It uses
a generated REST client, built from the swagger definitions, the websocket
interface for live updates of entities and eager loading of everything
except runners, as users may have many runners and we don't want to load
hundreds of runners in memory.
Proper pagination should be implemented in the API, in future commits,
to avoid loading lots of elements for no reason.
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
Filter out gitea tools to only consider archived downloads. This
should help in situations where bandwidth is more important than
CPU time used to unarchive the tools.
Also a drive by fix for scale sets cleanup.
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
There seems to be a change in the scale set message. It now includes
a jobID and sets the runner request ID to 0. This change adds separate
job ID fields for workflow jobs and scaleset jobs.
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
Do not look for a name when composing the scale set. Preload may not
have been called on an entity, but we still have the ID, which is the
only thing needed when GetEntity() is called.
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
* time.NewTicker will panic if the duration is 0. Make it return
early if duration is 0.
* Return a pre-closed channel in Wait() instead of nil. Ensures receiver
will not block forever.
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
Use JoinPath() in newActionsRequest() and make sure we pass relative
paths to it. This should fix scale sets on GHES.
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
This change adds the ability to filter the list of entities returned
by the API by entity owner, name or endpoint, depending on the entity
type.
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
* We were passing the wrong type to GORM for events
* We now expose entity events in the API and CLI
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
This change renames a lot of variables, types and functions to be more
generic. The goal is to allow GARM to add more forges in the future.
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
This change adds a loop that keeps a cache of credentials rate limits
as reported by the github API. The cache is updated every 30 seconds
and is purely informational for the user.
This change also adds some caching improvements. Functions that return
values from the cache as lists, will now sort by ID or creation date.
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
This change adds an update routine in the cache worker, for github tools
downloads.
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
* Split the main function into a couple of more functions
* Add credentials, entity, pool and scaleset cache
* add credentials worker that updates the cache
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
Adds a periodic cleanup function that cross checks runners between github,
the provider and the GARM database. If an inconsistency is found, GARM will
attempt to fix it.
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
We will need to run various checks against the runners that are managed
by a scale set. The runners are updated by the DB watcher, so we should
always have an up to date view of their state. We can then confidently
monitor them without needing to query the DB.
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
