[lxc-users] lxc-ls -f problem

[Dave Birch]

Just seen this happen again on a different machine and may have worked 
out when it started happening.

It looks like my scripts had just cloned a container from my template 
container using overlayfs, then tried to start the new container.


I now not run the following commands whilst referencing this container:
lxc-start
lxc-destroy

Any attempt to run them hangs and I get very similar trace level debug 
for both commands when trying to start either the template container or 
the clone:

sudo lxc-start -n ubuntutemplate -l trace -o startTemplate.out
<left it for a few minutes>  
^C

cat startTemplate.out: 
      lxc-start 1432292499.352 INFO     lxc_start_ui - 
lxc_start.c:main:265 - using rcfile /var/lib/lxc/ubuntutemplate/config
      lxc-start 1432292499.352 WARN     lxc_log - log.c:lxc_log_init:316 
- lxc_log_init called with log already initialized

Only difference is the destroy log doesn't show the "using rcfile" 
message, only the log warning.

The previously reported error occurs when I run lxc-ls with any of the 
optional flags such as -f, --running etc. Plain lxc-ls with no flags 
seems to return without error.

Also, lxc commands on other containers other than my template and clone 
seem to be fine.

My issue is that without being able to lxc-destroy the container, I have 
no idea how to get rid of the problem containers to resolve my issue. I 
will do some more investigation and see if I find anything else, the big 
thing to know would be how to manually remove an lxc container (if it's 
possible).