On Monday I'll be upgrading the database that backs toolforge's Harbor service. Any attempts to download new container images during this window will fail.
Existing, running tools will not be affected, and periodically scheduled jobs should also be fine, but any brand new jobs may fail, depending on what is or isn't cache at the moment.
The outage should be brief; if you see a cryptic failure around the scheduled time just take a 10 minute break and try again.
-Andrew
_______________________________________________ Cloud-announce mailing list -- cloud-announce@lists.wikimedia.org List information: https://lists.wikimedia.org/postorius/lists/cloud-announce.lists.wikimedia.o...
This is done. If you were unlucky enough to see a failure during this very short window, try again now and all should be well.
-Andrew
On 7/23/24 10:00 AM, Andrew Bogott wrote:
On Monday I'll be upgrading the database that backs toolforge's Harbor service. Any attempts to download new container images during this window will fail.
Existing, running tools will not be affected, and periodically scheduled jobs should also be fine, but any brand new jobs may fail, depending on what is or isn't cache at the moment.
The outage should be brief; if you see a cryptic failure around the scheduled time just take a 10 minute break and try again.
-Andrew
_______________________________________________ Cloud-announce mailing list -- cloud-announce@lists.wikimedia.org List information: https://lists.wikimedia.org/postorius/lists/cloud-announce.lists.wikimedia.o...