[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Condor-users] GCB log errors



I have the same problem.

If you find a solution, please post it here

Thanks



El Thursday 12 July 2007 22:04:23 David Atkinson escribió:
> Hello all,
>
>
>
> I am configuring the head node on a cluster of ours with GCB so that we can
> flock to it from our main condor pool.  This head node is the central
> manager for the cluster's condor pool.  I keep seeing a number of the
> following two types of errors in the GCB_RelayServerLogs and I was
> wondering if anyone could shed some light on them for me.
>
>
>
> These were the first big grouping of continuous errors:
>
>
>
> 7/10 15:30:07 [relay_server.C:1019]     bind - Address already in use
>
> 7/10 15:31:37 [relay_server.C:1018] handleNewClient: bind(proxy) to
> <ip.addr.of.centralmanager:portnum>
>
>
>
> In the past couple days these made an appearance as well:
>
>
>
> 7/12 10:19:49 [relay_server.C:676] Relaying UDP packet from client: fd[23]
>
> 7/12 10:19:49 [relay_server.C:677]      receiving from the client:
> Connection refused
>
> 7/12 10:19:49 [relay_server.C:773] Deleting fd[23](client side) <-> <
> ip.addr.of.centralmanager:portnum >
>
>
>
>
>
> I'm aware the manual discourages having a central manager and GCB running
> on the same server but since this was the head node for the cluster we were
> wanting to communicate with it made the best sense.  Could that be the
> cause of these errors?
>
>
>
> Thanks in advance,
>
> David Atkinson