Can't deploy nested focal LXDs on an impish host
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Canonical Juju |
Triaged
|
Low
|
Unassigned |
Bug Description
Using the lxd/localhost provider on an impish amd64 host, running the latest Juju snap:
https:/
I've only tested focal guests on impish, but can test other series if that's helpful.
Syslog is full of repeated apparmor DENIED messages like these, seems to be the same targets each time (netstat and snmp):
Oct 22 14:53:50 x1c9 kernel: [100445.847161] audit: type=1400 audit(163488922
Oct 22 14:53:50 x1c9 kernel: [100445.847170] audit: type=1400 audit(163488922
Changed in juju: | |
status: | New → Triaged |
importance: | Undecided → Medium |
milestone: | none → 2.9-next |
Changed in juju: | |
milestone: | 2.9-next → none |
Decided to work around this by starting up a multipass VM, manually adding that to my model as machine 0, then deploying LXDs onto that, but I see the same issue:
0 started 10.214.6.65 manual:10.214.6.65 focal Manually provisioned machine
0/lxd/0 pending juju-789e09-0-lxd-0 focal Container started
0/lxd/1 pending juju-789e09-0-lxd-1 focal Container started
So this isn't necessarily LXD-specific on the provider side after all - updated bug title.