lxc

lxc-start failure on vivid

Bug #1434737 reported by Eric Snow
8
This bug affects 1 person
Affects Status Importance Assigned to Milestone
lxc
New
Undecided
Unassigned
lxc (Ubuntu)
Confirmed
Critical
recep

Bug Description

When I run the following (after populating the template directory with 2 files) it fails:

    sudo lxc-start --daemon --name juju-vivid-lxc-template --logfile /var/lib/juju/containers/juju-vivid-lxc-template/container.log --logpriority WARN --console-log /var/lib/juju/containers/juju-vivid-lxc-template/console.log

I'm trying to determine if this is an issue in juju or in lxc. The two files are a cloud-init script and lxc.conf. The conf file contains the following:

# network config
# interface "eth0"
lxc.network.type = veth
lxc.network.link = lxcbr0
lxc.network.flags = up
lxc.network.mtu = 9001

After running the command I find a new container.log with a bunch of logging and an empty console.log (also new). In the container.log file everything looks fine until the following lines right after a bunch of "cgroup" messages:

      lxc-start 1426888101.167 ERROR lxc_apparmor - lsm/apparmor.c:apparmor_process_label_set:183 - No such file or directory - failed to change apparmor profile to lxc-container-default
      lxc-start 1426888101.167 ERROR lxc_sync - sync.c:__sync_wait:51 - invalid sequence number 1. expected 4
      lxc-start 1426888101.167 ERROR lxc_start - start.c:__lxc_start:1157 - failed to spawn 'juju-vivid-lxc-template'

This is followed by a bunch of other error messages, but I figured this is the key bit. If necessary I can attach the whole log.

Revision history for this message
Eric Snow (ericsnowcurrently) wrote :

If I run in foreground mode, this is what gets printed out:

lxc-start: lsm/apparmor.c: apparmor_process_label_set: 183 No such file or directory - failed to change apparmor profile to lxc-container-default
lxc-start: sync.c: __sync_wait: 51 invalid sequence number 1. expected 4
lxc-start: start.c: __lxc_start: 1157 failed to spawn 'juju-vivid-lxc-template'
lxc-start: cgmanager.c: cgm_remove_cgroup: 518 call to cgmanager_remove_sync failed: invalid request
lxc-start: cgmanager.c: cgm_remove_cgroup: 520 Error removing all:lxc/juju-vivid-lxc-template-21
lxc-start: lxc_start.c: main: 344 The container failed to start.
lxc-start: lxc_start.c: main: 348 Additional information can be obtained by setting the --logfile and --logpriority options.

Revision history for this message
Eric Snow (ericsnowcurrently) wrote :

This *may* be related to https://bugs.launchpad.net/ubuntu/+source/lxc/+bug/1347020.

Revision history for this message
Eric Snow (ericsnowcurrently) wrote :

LXC was installed with "sudo apt-get install lxc" on vivid.

$ lxc-ls --version
1.1.0

I'm not exactly sure which vivid revision we are running other than than it's from at most within the last week. If I recall correctly we're updating it nightly, but I'll need to double-check.

$ lsb_release -a
No LSB modules are available.
Distributor ID: Ubuntu
Description: Ubuntu Vivid Vervet (development branch)
Release: 15.04
Codename: vivid

Revision history for this message
Eric Snow (ericsnowcurrently) wrote :

FWIW, I don't think this is related to lp:1347020.

Revision history for this message
Curtis Hovey (sinzui) wrote :

Per strikov, this might be bug 1432683. I ran
     sudo /etc/init.d/apparmor reload
on the vivid-slave-b machine that is reporting the issue. Then reran the juju local deploy test and it failed. This is a different kind of issue.

Revision history for this message
Eric Snow (ericsnowcurrently) wrote :

FTR:

$ sudo aa-status
apparmor module is loaded.
11 profiles are loaded.
11 profiles are in enforce mode.
   /sbin/dhclient
   /usr/bin/lxc-start
   /usr/lib/NetworkManager/nm-dhcp-client.action
   /usr/lib/NetworkManager/nm-dhcp-helper
   /usr/lib/connman/scripts/dhclient-script
   /usr/lib/libvirt/virt-aa-helper
   /usr/sbin/libvirtd
   /usr/sbin/tcpdump
   lxc-container-default
   lxc-container-default-with-mounting
   lxc-container-default-with-nesting
0 profiles are in complain mode.
21 processes have profiles defined.
21 processes are in enforce mode.
   /usr/bin/lxc-start (31271)
   /usr/sbin/libvirtd (6819)
   lxc-container-default (578)
   lxc-container-default (850)
   lxc-container-default (918)
   lxc-container-default (935)
   lxc-container-default (1083)
   lxc-container-default (1092)
   lxc-container-default (1093)
   lxc-container-default (1109)
   lxc-container-default (1184)
   lxc-container-default (1248)
   lxc-container-default (1250)
   lxc-container-default (7676)
   lxc-container-default (7678)
   lxc-container-default (12780)
   lxc-container-default (12815)
   lxc-container-default (16641)
   lxc-container-default (17358)
   lxc-container-default (31290)
   lxc-container-default (32719)
0 processes are in complain mode.
0 processes are unconfined but have a profile defined.

Revision history for this message
Eric Snow (ericsnowcurrently) wrote :

The root cause in lp-1432683 sounds likely for this issue as well. I have verified that "/lib/init/apparmor-profile-load" does not exist on the host.

Revision history for this message
Serge Hallyn (serge-hallyn) wrote :

Sorry for the mess - I was thinking this was the cgmangaer bug. yes, this is due to not having cgroup-bin package installed

Changed in lxc (Ubuntu):
status: New → Confirmed
importance: Undecided → Critical
assignee: nobody → Serge Hallyn (serge-hallyn)
summary: - lxc-start failure on vivid
+ lxc-start failure (systemd vs cgmanager) on vivid
Changed in cgmanager (Ubuntu):
status: New → Confirmed
importance: Undecided → Critical
summary: - lxc-start failure (systemd vs cgmanager) on vivid
+ lxc-start failure on vivid
no longer affects: cgmanager (Ubuntu)
Changed in lxc (Ubuntu):
assignee: Serge Hallyn (serge-hallyn) → recep (rbasak)
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.
Subscribing...

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.