[Lustre-discuss] performance issues in simple lustre setup

[Isaac Huang]

On Tue, Jun 03, 2008 at 08:32:09AM -0400, Murray Smigel wrote:
>    Some additional information on the problem.  I tried disconnecting the
>    ethernet connection to
>    the server machine (192.168.1.94) and tried running a disk test on the
>    client (192.168.1.156 via ethernet), writing to
>    what I thought was the IB mounted file system (mount -t lustre
>    192.168.3.50 at o2ib:/lusty /slut, where 192.168.3.50 is
>    the ip of ib0 on the client machine).
>    Looking at /var/log/syslog I saw:
>    Jun  3 08:18:59 nasnu3 kernel: Lustre:
>    3625:0:(router.c:167:lnet_notify()) Ignoring prediction from
>    192.168.1.156 at tcp of 192.168.1.94 at tcp down 23987983711 seconds in the
>    future

Known bug, and already fixed:
https://bugzilla.lustre.org/show_bug.cgi?id=14300

> ......
>    lusty-MDT0000-mdc-ffff810226207000: Connection restored to service
>    lusty-MDT0000 using nid 192.168.1.94 at tcp.

It seemed to me that Lustre was actually using the @tcp network, which
could be a Lustre configuration issue.

Isaac