> Quick summary of the past 2 weeks:
>
> 12 segmentation faults and 1 abort on the 8 Ubuntu 20.04 servers which
> are not running the patched packages.
>
> No segmentation faults on the 4 Ubuntu 20.04 servers which are running
> the patched packages.
>
> 4 Ubuntu 22.04 servers which are running similar configurations have not
> yet shown any problems, but for the most part they are on the quieter
> networks.
>
> No problems on any Ubuntu 18.04 server (but we only have 2 still in
> production). No Ubuntu 22.04 servers currently in production.
>
> I've still not found a way to trigger the problem, nor a common factor
> in which stock Ubuntu 20.04 servers are effected or not - beyond the
> obvious thing that the busier servers seem more likely to exhibit the
> problem.
>
> So the patched packages in the PPA look like they fix the problem to me,
> but I'm not enough of an expert on BIND to know if what they patch might
> cause a regression or problem elsewhere.
Thank you very much for the status update, John.
It does indeed look like the backported patch is the solution to this
problem. It would have been great if we could figure out the steps to
reproduce this problem, but sometimes it's just too hard to do it. I
think we can proceed with the SRU and inform that it will be tested by
the community.
Today is my last day before the EOY break, so I won't be able to make
any progress here until I'm back in the beginning of January.
Meanwhile, feel free to keep me informed in case something unexpected
happens.
On Friday, December 16 2022, John Edwards wrote:
> Quick summary of the past 2 weeks:
>
> 12 segmentation faults and 1 abort on the 8 Ubuntu 20.04 servers which
> are not running the patched packages.
>
> No segmentation faults on the 4 Ubuntu 20.04 servers which are running
> the patched packages.
>
> 4 Ubuntu 22.04 servers which are running similar configurations have not
> yet shown any problems, but for the most part they are on the quieter
> networks.
>
> No problems on any Ubuntu 18.04 server (but we only have 2 still in
> production). No Ubuntu 22.04 servers currently in production.
>
> I've still not found a way to trigger the problem, nor a common factor
> in which stock Ubuntu 20.04 servers are effected or not - beyond the
> obvious thing that the busier servers seem more likely to exhibit the
> problem.
>
> So the patched packages in the PPA look like they fix the problem to me,
> but I'm not enough of an expert on BIND to know if what they patch might
> cause a regression or problem elsewhere.
Thank you very much for the status update, John.
It does indeed look like the backported patch is the solution to this
problem. It would have been great if we could figure out the steps to
reproduce this problem, but sometimes it's just too hard to do it. I
think we can proceed with the SRU and inform that it will be tested by
the community.
Today is my last day before the EOY break, so I won't be able to make
any progress here until I'm back in the beginning of January.
Meanwhile, feel free to keep me informed in case something unexpected
happens.
Thanks,
--
Sergio
GPG key ID: E92F D0B3 6B14 F1F4 D8E0 EB2F 106D A1C8 C3CB BF14