bge0 watchdog timeout?

No one on the freebsd-current mailing list seems to know
how to avoid the following:

May 28 16:39:59 node13 kernel: bge0: watchdog timeout -- resetting
May 28 16:39:59 node13 kernel: bge0: link state changed to DOWN
May 28 16:40:02 node13 kernel: bge0: link state changed to UP

This is causing all sorts of havoc on a cluster of FreeBSD
systems that is used for multi-day MPI computations.

Anyone have any insight in fixing bge0 to no timeout?

--
Steve
_______________________________________________
freebsd-net@xxxxxxxxxxx mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "freebsd-net-unsubscribe@xxxxxxxxxxx"

Relevant Pages

  • Re: Full Text is locking up
    ... Its unlikely that the cluster is causing the problem. ... Looking for a SQL Server replication book? ... on SP1. ...
    (microsoft.public.sqlserver.fulltext)
  • Cancelling jobs in print queue
    ... the print queue ("24 documents waiting"). ... This is causing ... havoc whenever I want to print something. ...
    (microsoft.public.word.printingfonts)
  • Re: Winfax Pro
    ... Winfax Pro seems to be causing all sorts of problems. ... For example, the cursor disappears suddenly, for absolutely no reason. ...
    (microsoft.public.windowsxp.general)
  • Re: Newbie - Quorum on SAN
    ... the quorum device at the same...which is causing both nodes to lose the ... of time to arbitrate for the quorum disk without the interference of the ... other cluster nodes. ... Cluster service has terminated. ...
    (microsoft.public.windows.server.clustering)
  • Re: nodes will not shut down
    ... than your cluster configuration. ... You also mention that the private NIC didn't have a DNS record, ... >> are running a configuration that is likely causing your problems. ...
    (microsoft.public.windows.server.clustering)