<div dir="ltr"><div dir="ltr"><div class="gmail_default" style="font-size:small">Yeah, I thought about that. Both the client and servers are using the defaults for ko2iblnd-</div><div class="gmail_default" style="font-size:small"><br></div><div class="gmail_default" style="font-size:small"> tunables:<br> peer_timeout: 180<br> peer_credits: 8<br> peer_buffer_credits: 0<br> credits: 256<br> lnd tunables:<br> peercredits_hiw: 4<br> map_on_demand: 0<br> concurrent_sends: 8<br> fmr_pool_size: 512<br> fmr_flush_trigger: 384<br> fmr_cache: 1<br> ntx: 512<br> conns_per_peer: 1<br></div><div class="gmail_default" style="font-size:small"><br></div><div class="gmail_default" style="font-size:small">Thanks,</div><div class="gmail_default" style="font-size:small">Kevin<br></div></div></div><br><div class="gmail_quote"><div dir="ltr">On Fri, Jan 11, 2019 at 5:17 PM Mohr Jr, Richard Frank (Rick Mohr) <<a href="mailto:rmohr@utk.edu">rmohr@utk.edu</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Is it possible you have some incompatible ko2iblnd module parameters between the 2.8 servers and the 2.10 clients? If there was something causing LNet issues, that could possibly explain some of the symptoms you are seeing.<br>
<br>
--<br>
Rick Mohr<br>
Senior HPC System Administrator<br>
National Institute for Computational Sciences<br>
<a href="http://www.nics.tennessee.edu" rel="noreferrer" target="_blank">http://www.nics.tennessee.edu</a><br>
<br>
<br>
> On Jan 10, 2019, at 4:23 PM, Kevin M. Hildebrand <<a href="mailto:kevin@umd.edu" target="_blank">kevin@umd.edu</a>> wrote:<br>
> <br>
> I've got a RHEL6 Lustre installation where the servers are running 2.8.0, that I'd prefer not to upgrade. <br>
> We've been running 2.8.0 on RHEL6 clients as well and everything's been working fine. However, I just updated the Linux release on the RHEL6 clients to 6.10, and Lustre 2.8.0 will no longer compile on the latest kernel. I've built and installed 2.10.6 on these clients, and the kernel modules load fine, but on first contact with any lustre server, I get a bunch of timeouts before I can get a valid connection. The Lustre network in this case is Infiniband, using Mellanox OFED on the clients.<br>
> 'lctl ping' hangs for a few seconds and returns 'failed to ping 192.168.64.70@o2ib1: Input/output error'. An IPoIB ping of the server IP address works fine.<br>
> At the same time I get a message in syslog that says 'LNet: 8778:0:(o2iblnd_cb.c:3192:kiblnd_check_conns()) Timed out tx for 192.168.64.70@o2ib1: 4296292 seconds'<br>
> Nothing shows up in the logs on the server side.<br>
> <br>
> If I repeat the 'lctl ping' a few times, after 30-60 seconds or so, 'lctl ping' succeeds.<br>
> This happens for each of my Lustre servers, and once I get a successful ping back, it seems to be fully functional up until the next reboot, or until the Infiniband modules are reloaded.<br>
> <br>
> If I try to mount the filesystem without doing the pings, I'll get timeouts contacting the MDS for the same 30-60 seconds, and then once the MDSes are reachable, I get timeouts to the OSSes for a while, until they become reachable, and once they're all talking, all seems to be fine.<br>
> <br>
> Any ideas on what could be wrong?<br>
> <br>
> Thanks,<br>
> Kevin<br>
> <br>
> --<br>
> Kevin Hildebrand<br>
> University of Maryland<br>
> _______________________________________________<br>
> lustre-discuss mailing list<br>
> <a href="mailto:lustre-discuss@lists.lustre.org" target="_blank">lustre-discuss@lists.lustre.org</a><br>
> <a href="http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org" rel="noreferrer" target="_blank">http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org</a><br>
<br>
<br>
</blockquote></div>