snap using a lot of CPU inside containers
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
snapd |
Fix Committed
|
Undecided
|
Unassigned |
Bug Description
We are having an issue with our Juju environment - snap based, where our instances have state DOWN.
Juju services are running with no issues. We have a Jenkins instance on the default model - it is still working (despite of its DOWN state):
```
Model Controller Cloud/Region Version SLA Timestamp
controller localhost-localhost localhost/localhost 2.9.35 unsupported 14:44:55-04:00
Machine State Address Inst id Series AZ Message
0 down 10.5.194.201 juju-32e0d8-0 bionic Running
```
Checking the container where Jenkins is running, snap seems to be using a lot of CPU:
```
Default model - jenkins instance:
Machine State Address Inst id Series AZ Message
2 down 10.5.194.200 juju-5ff7db-2 focal Running
```
Jenkins instance: `juju-5ff7db-2`:
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
419130 root 20 0 2717504 80948 15704 S 105.9 0.1 0:54.78 snapd
We have ZFS as the the backing file system.
also, while trying to run `'df -h` it got stuck on the container