machine-0 takes up massive memory, slows down to unusable, "upgrade in progress" loops
Bug #1644396 reported by
Ryan Finnie
This bug report is a duplicate of:
Bug #1643795: juju.worker.dependency engine.go:539 "mgo-txn-resumer" manifold worker returned unexpected error: cannot resume transactions: cannot find transaction ObjectIdHex("stuff").
Edit
Remove
This bug affects 2 people
| Affects | Status | Importance | Assigned to | Milestone | |
|---|---|---|---|---|---|
| Canonical Juju |
New
|
Undecided
|
Unassigned | ||
Bug Description
We have an environment where the controller is making the environments unusable. If jujud-machine-0 is restarted, it spams with 'login for "unit-nrpe-0" blocked because upgrade in progress', even though all agents seemed to be on the latest (2.0.2).
If all of the agents on the associated model are stopped, machine 0's non-machine agents are stopped, just leaving jujud-machine-0, it looks ok, but jujud will be at 200% CPU, slowly gaining memory over minutes/hours (over 8GB RSS), and will eventually stop responding to anything (juju status, etc). See attached log.
mgopurge does not have an effect.
Revision history for this message
| Ryan Finnie (fo0bar) wrote | #1 |
| description: | updated |
| description: | updated |
Revision history for this message
| Junien Fridrick (axino) wrote | #2 |
To post a comment you must log in.
This is actually juju catching up for https:/