100% CPU / wedge with 9.8.3-P4 & RPZ?
p.mayers at imperial.ac.uk
Sat Mar 16 19:16:17 UTC 2013
On 03/16/2013 06:46 PM, Vernon Schryver wrote:
>> From: Phil Mayers <p.mayers at imperial.ac.uk>
>> It's unfortunate I wasn't able to obtain one; gdb wasn't installed on
>> the box, and I couldn't get the package installed because DNS was down.
> Depending on the flavor of the system and its configuration, adding
> lines to /etc/hosts can be effective for working around local or
> remote DNS problems.
Thanks. I do know what /etc/hosts does!
To be honest, I was focussed on restoring service. Grabbing a quick core
dump would have been fine. Spending 10 minutes getting packages
installed was not (To put in a local hosts entry, I would first have had
to logged into our IPAM system to find the IP address of the webserver
running our yum repo. Since DNS was out, login was tricky because the
Kerberos SRV records could not be resolved).
> I use `kill -6` or another fatal signal.
> Some systems disable by default the creation of core files. This
Having tested after the event, kill -6 doesn't create a core on my
system for named. Whether that's permissions or what I don't know and do
not have time to investigate.
> irritating, simplistic misconfiguration (that's only my view) is
> especially common in the Linux world. See
Erm... thanks. I think. But I am broadly familar with the concept of
using a search engine to find out information...
> Gdb would have been handy for looking at named without creating a
> core file or disturbing the process by more than what it would see
> as a jump in time.
Or for running "gcore `pgrep named`" which has the advantages of
sidestepping all the crap about core file output locations, permissions,
ulimit and so on.
More information about the bind-users