I can't remember if this was on mlx4 or mlx5 driver cards, but we found we had to use LINKDELAY=20 when using bonding for Ethernet.
Simon ________________________________________ From: [email protected] [[email protected]] on behalf of J. Eric Wonderley [[email protected]] Sent: 20 December 2017 20:37 To: gpfsug main discussion list Subject: Re: [gpfsug-discuss] more than one mlx connectx-4 adapter in same host Just plain tcpip. We have dual port connectx4s in our nsd servers. Upon adding a second connectx4 hba...no links go up or show "up". I have one port on each hba configured for eth and ibv_devinfo looks sane. I cannot find anything indicating that this should not work. I have a ticket opened with mellanox. On Wed, Dec 20, 2017 at 3:25 PM, Knister, Aaron S. (GSFC-606.2)[COMPUTER SCIENCE CORP] <[email protected]<mailto:[email protected]>> wrote: We’ve done a fair amount of VPI work but admittedly not with connectx4. Is it possible the cards are trying to talk IB rather than Eth? I figured you’re Ethernet based because of the mention of Juniper. Are you attempting to do RoCE or just plain TCP/IP? On December 20, 2017 at 14:40:48 EST, J. Eric Wonderley <[email protected]<mailto:[email protected]>> wrote: Hello: Does anyone have this type of config? The host configuration looks sane but we seem to observe link-down on all mlx adapters no matter what we do. Big picture is that we are attempting to do mc(multichassis)-lags to a core switch. I'm somewhat fearful as to how this is implemented in the juniper switch we are about to test. _______________________________________________ gpfsug-discuss mailing list gpfsug-discuss at spectrumscale.org<http://spectrumscale.org> http://gpfsug.org/mailman/listinfo/gpfsug-discuss _______________________________________________ gpfsug-discuss mailing list gpfsug-discuss at spectrumscale.org http://gpfsug.org/mailman/listinfo/gpfsug-discuss
