Hi Chris,
Don't you have things setup such that the xrootd will be restarted
automatically? (I see that someone posted a message that the servers
were done.)
In general you should set things up such that (a) core files from
crashed xrootd/olbd are created/saved and (b) that the servers are
restarted automatically if they crash. This will minimize the impact
for the users and also give Andy something to look at.
thanks,
Pete
On Fri, Oct 15, 2004 at 09:13:43AM +0100, Brew, CAJ (Chris) wrote:
> Hi,
>
> I came in this morning to find the xrootd processes on csflnx108 (our
> redirector) and csfnfs41 (one of the data servers) had died.
>
> As far as I know nothing had has changed on these servers recently.
>
> Looking in the xrdlogs I see:
>
> pure virtual method called
> 041012 18:58:17 11981 oss_XrdOssMSS_Gateway: parent process terminated;
> terminat
> ing gateway
>
> on csfnfs41, and:
>
> 05:26:29 3639 oss_XrdOssMSS_Gateway: parent process terminated; terminating
> gateway
>
> on csflnx108.
>
> We're running version xrootd-20040901-0613.
>
> Yours,
> Chris.
>
+-----------------------------------------------------------------------+
I'm away from 8 Oct - 4 Nov (inclusive) with only occasional access
to email, so you should expect some delay in my responses.
-------------------------------------------------------------------------
Peter Elmer E-mail: [log in to unmask] Phone: +41 (22) 767-4644
Address: CERN Division PPE, Bat. 32 2C-14, CH-1211 Geneva 23, Switzerland
-------------------------------------------------------------------------
