Application fails to connect to controller during install, connection reset by peer
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
| juju |
Medium
|
Unassigned |
Bug Description
There are instances of a subordinate application is failing to connect to the juju controller during install and shutting down:
$ cat unit-telegraf-
2020-10-03 16:16:31 ERROR juju.worker.
2020-10-03 16:16:31 DEBUG install ERROR codec.ReadHeader error: error receiving message: read tcp 10.244.
2020-10-03 16:16:31 ERROR juju.worker.
2020-10-03 16:16:31 ERROR juju.worker.
2020-10-03 16:16:31 ERROR juju.worker.
2020-10-03 16:16:31 ERROR juju.worker.
2020-10-03 16:16:31 ERROR juju.worker.
2020-10-03 16:16:35 DEBUG install ERROR connection is shut down
2020-10-03 16:16:44 DEBUG install ERROR connection is shut down
2020-10-03 16:16:44 ERROR juju.worker.
2020-10-03 16:16:44 ERROR juju.worker.
It is a subordinate application relating to an application in one model for placement, and relaying information to a 2nd model by way of CMR.
Full logs can be found at:
https:/
Alexander Balderson (asbalderson) wrote : | #1 |
Pete Vander Giessen (petevg) wrote : | #2 |
Triaged as medium and dropped into 2.8.6 milestone. This looks like the sort of sporadic issue that we should fix so that our Solutions QA tests will run more consistently.
Changed in juju: | |
status: | New → Triaged |
importance: | Undecided → Medium |
milestone: | none → 2.8.6 |
Changed in juju: | |
milestone: | 2.8.6 → 2.8-next |
All future test runs which hit this bug can be found at /solutions. qa.canonical. com/bugs/ bugs/bug/ 1898576
https:/