Cluster image download gives up and logs an IOError too soon
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
MAAS |
Fix Released
|
Critical
|
Mike Pontillo |
Bug Description
This is error #1 in bug #1457788; original description:
When restarting the two main MAAS services (maas-regiond and maas-clusterd), a bunch of stacktraces are printed in the logs. Ultimately, the services start okay.
service maas-regiond restart && service maas-clusterd restart
Most of these stacktraces seems to be caused by races in how the various services are started and although it is not critical, it's really painful to see that MAAS doesn't deal more gracefully with this.
This is the combination of both regiond.log and clusterd.log: http://
2015-05-22 08:39:28+0000 [-] Downloading images failed.
Traceback (most recent call last):
File "/usr/lib/
self.
File "/usr/lib/
self.
File "/usr/lib/
current.result = callback(
File "/usr/lib/
_inlineCal
--- <exception caught here> ---
File "/usr/lib/
result = result.
File "/usr/lib/
return g.throw(self.type, self.value, self.tb)
File "/usr/lib/
yield self._start_
File "/usr/lib/
result = result.
File "/usr/lib/
return g.throw(self.type, self.value, self.tb)
File "/usr/lib/
get_
File "/usr/lib/
result = context.call(ctx, function, *args, **kwargs)
File "/usr/lib/
return self.currentCon
File "/usr/lib/
return func(*args,**kw)
File "/usr/lib/
return func(*args, **kwargs)
File "/usr/lib/
boot_
File "/usr/lib/
image_
File "/usr/lib/
source['url'], keyring=
File "/usr/lib/
dumper.
File "/usr/lib/
content, payload = reader.
File "/usr/lib/
raw = self.source(
File "/usr/lib/
self.open()
File "/usr/lib/
self.fd = self._open()
File "/usr/lib/
raise myerr
exceptions.
This is the cluster trying to sync from the region before the region is up.
Related branches
- Blake Rouse (community): Approve
-
Diff: 63 lines (+30/-0)2 files modifiedsrc/provisioningserver/import_images/download_descriptions.py (+9/-0)
src/provisioningserver/import_images/tests/test_download_descriptions.py (+21/-0)
Changed in maas: | |
importance: | Undecided → High |
status: | New → Triaged |
summary: |
- Cluster service gives up and logs an IOError too soon + Cluster image download gives up and logs an IOError too soon |
description: | updated |
Changed in maas: | |
status: | Triaged → Fix Committed |
Changed in maas: | |
status: | Fix Committed → Fix Released |
Raising this to critical since this is a crash and we shouldn't tolerate them even if they seem inconsequential.