[Gluster-users] "gluster volume heal datastore2 info" blocking, heal not completing

[Lindsay Mathieson]

On 14/04/2016 8:52 PM, Krutika Dhananjay wrote:
> OK here is one thing you could do if you are able to recreate this 
> hang easily on your setup.
> The next time you find heal-info hanging, could you do the following:
>
> # gluster volume statedump <VOL>
>
> and share the statedump logs? You will find them at /var/run/gluster/


Ok I have that statedump and the glfsheal-datastore2.log (from vng where 
the heal info was running)

reproduced by:
- killing glusterfsd, glusterd on vna
- ran a build process on a hosted vm
- number of entries as reported by 'gluster volume heal datastore2 
statistics heal-count' was at approx 5000
- restarted glusterd on vn, brick started healing
- ran the build and issued a 'gluster volume heal datastore2 info' on vng
- the build ran for a few tens of seconds and info started listing 
shards, then both froze.
- thats when I ran the state dump and retrieved the glfsheal-datastroe2 log

- killed the heal info and I/O on the vm resumed immediately.

Do you just want me to email yo the logs as attachements?, just a couple 
of MB.

-- 
Lindsay Mathieson