Hmm...good point. However, I should clarify that userspace logs are being
corrected for timezone, but kernel logs are not. For userspace apps the
timestamping is done in the glibc syslog() call, so now I need to figure out
where it's done for the kernel.
> > Can anyone point me to the right place to deal with this?
>
> Restart syslog so that it notices the new timezone, or something else, I
> don't know. IIRC, you are the one doing strange things with syslog.
> Are you doing network syslog logging now? Are both of your hosts running
> with the same timezone?
I always was logging remotely, but we wanted to log to NFS-mounted files as well
without hanging the userspace apps when NFS borked.
What I ended up doing was to fork syslog and set up a sysV message queue between
the parent and child. The child does all writing to the NFS-mounted
filesystem. Thus, if NFS dies for whatever reason the child blocks and the
parent just dumps messages into the queue. Eventually (1024 messages or 16KB of
data) the queue fills up and we start to lose messages (the parent uses
non-blocking writes), but in practice we don't hit the limit before NFS is
regained.
Chris
-- Chris Friesen | MailStop: 043/33/F10 Nortel Networks | work: (613) 765-0557 3500 Carling Avenue | fax: (613) 765-2986 Nepean, ON K2H 8E9 Canada | email: cfriesen@nortelnetworks.com - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/