[Debian-ha-maintainers] Bug#933398: Bug#933398: resource-agents: ZFS resource agent contains a bashism and fails

[Valentin Vidić]

On Tue, Jul 30, 2019 at 12:52:20PM +0200, Koen de Jonge wrote:
> while trying to use the ocf:heartbeat:ZFS resource from the resource-agents package on debian buster I ran in to a problem where 
> 
> In /var/log/daemon.log I got this:
>  pacemaker-controld[759]:  warning: Action 3 (zfs_monitor_0) on node1 failed (target: 7 vs. rc: 0): Error
> 
> The result was that directly after the zfs pool was imported by pacemaker (i.e. the resource was started) the resource got stopped again since the monitor script did not report the HEALTH of the pool well. 
> 
> The check for health changed on 17 oct 2018 see: https://github.com/ClusterLabs/resource-agents/commit/2bdeee4b063f73b5e1804b4bec4d1ad43733a9f7#diff-2f9687bda2dc6253e000b30aaea222d9
> 
> This change, that subsequently went in to debian buster works fine in bash, but not in dash (which /bin/sh is).
> 
> I have reported the problem upstream and created a fix here: https://github.com/ClusterLabs/resource-agents/pull/1372
> 
> Can you fix this in debian stable? 

Thanks for the report and the fix.  We will definitely update
unstable/testing and backports when a new version is released
by the upstream. The updated for stable is a bit more tricky
because:

https://www.debian.org/doc/manuals/developers-reference/ch05.html#upload-stable
"Basically, a package should only be uploaded to stable if one of the following
happens:

* a truly critical functionality problem
* the package becomes uninstallable
* a released architecture lacks the package"

-- 
Valentin