master promoter script looping and failing since 12/9
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
tripleo |
Fix Released
|
Critical
|
Unassigned |
Bug Description
It appears that the promoter script has been looping since 12/9, and is so unresponsive that serving logs is very slow and sometimes simply fails. The behavior started on 12/9.
The script runs every 10 minutes, and since 12/9 has been failing on the check for another promoter instance running. This is typically seen while uploading new container images upon promotion, as it can take > 30 mins to do so.
- http://
2017-12-09 00:51:48,436 32059 DEBUG promoter rdo-master-
2017-12-09 00:51:48,436 32059 INFO promoter Skipping promotion of current-tripleo-rdo to current-
2017-12-09 00:51:48,436 32059 INFO promoter new hash found for {'timestamp': 1512517091, 'promote_name': 'current-
2017-12-09 01:01:01,402 32149 ERROR promoter Another promoter process is running
2017-12-09 01:11:01,758 32244 ERROR promoter Another promoter process is running
2017-12-09 01:21:01,902 32320 ERROR promoter Another promoter process is running
2017-12-09 01:31:01,370 32397 ERROR promoter Another promoter process is running
2017-12-09 01:41:01,782 32479 ERROR promoter Another promoter process is running
2017-12-09 01:51:01,384 32557 ERROR promoter Another promoter process is running
2017-12-09 02:01:01,428 32651 ERROR promoter Another promoter process is running
2017-12-09 02:11:02,020 32741 ERROR promoter Another promoter process is running
...
- http://
- http://
- http://
The last link promoted for master was
- https:/
on 12/7
tags: | added: alert |
Changed in tripleo: | |
status: | New → Triaged |
tags: | added: quickstart |
Changed in tripleo: | |
status: | Triaged → Fix Released |
Access to the promoter instance is afaik controlled by
https:/ /github. com/openstack- infra/tripleo- ci/blob/ master/ scripts/ tripleo- cd-admins