Bug#848115: nfs-utils / more details?

[Daniel Pocock]

On 14/12/16 19:25, Andrey Rahmatullin wrote:
> On Wed, Dec 14, 2016 at 07:03:18PM +0100, Daniel Pocock wrote:
>> Can you clarify, did you manually mask the proc-fs-nfsd.mount
>> service and you expect dpkg to ignore that error and leave the
>> service as it is?
> I don't think I did anything to the systemd units, only disabled
> the sysvinit service when there was no systemd. In any case, I
> expect it to be disabled and I expect that this does not break the
> package upgrade :)
> 


I would agree with that - postinst and similar scripts should not fail
when a service restart fails.  The maintainer scripts need to be
looked at more closely, I made a similar fix in the ganglia package
scripts some time ago.

I don't think this is a regression with 1.3.4, it simply hasn't been
noticed before.

There is at least one other bug about dpkg failing to install this
package:

https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=640222

Regards,

Daniel

Bug#848115: nfs-utils / more details?

[Andrey Rahmatullin]

On Wed, Dec 14, 2016 at 07:03:18PM +0100, Daniel Pocock wrote:
> Can you clarify, did you manually mask the proc-fs-nfsd.mount service
> and you expect dpkg to ignore that error and leave the service as it is?
I don't think I did anything to the systemd units, only disabled the
sysvinit service when there was no systemd. In any case, I expect it to be
disabled and I expect that this does not break the package upgrade :)

-- 
WBR, wRAR


signature.asc
Description: PGP signature

Bug#848115: nfs-utils / more details?

[Daniel Pocock]

Thanks for the bug report

Can you clarify, did you manually mask the proc-fs-nfsd.mount service
and you expect dpkg to ignore that error and leave the service as it is?

Or you believe the package is doing something really bad and
proc-fs-nfsd.mount should not be in that state at all?