[lustre-discuss] strange errors on Lustre servers

[Lixin Liu]

Hi Zeeshan,

Thanks for the hint.

OPA works fine, but then I found someone brought up a misconfigured node which has
the conflict IP address on OPA interface. Fixing it and problem solved.

Lixin.


From: Zeeshan Ali Shah <javaclinic at gmail.com>
Date: Saturday, August 11, 2018 at 11:20 PM
To: Lixin Liu <liu at sfu.ca>
Cc: "lustre-discuss at lists.lustre.org" <lustre-discuss at lists.lustre.org>
Subject: Re: [lustre-discuss] strange errors on Lustre servers

What is output of opainfo ?
Sent from my iPhone

On 12 Aug 2018, at 04:04, Lixin Liu <liu at sfu.ca<mailto:liu at sfu.ca>> wrote:
Hi,

I am getting these errors on all our MDS and OSS servers (Lustre 2.10.1):

Aug 11 11:45:52 ndc-oss5b kernel: LNet: 24727:0:(o2iblnd_cb.c:2410:kiblnd_passive_connect()) Conn stale 172.19.142.119 at o2ib version 12/12 incarnation 1533927051163335/1533998625080752
Aug 11 11:55:52 ndc-oss5b kernel: LNet: 105990:0:(o2iblnd_cb.c:2410:kiblnd_passive_connect()) Conn stale 172.19.142.119 at o2ib version 12/12 incarnation 1533927051163335/1533998625080752
Aug 11 12:05:52 ndc-oss5b kernel: LNet: 105990:0:(o2iblnd_cb.c:2410:kiblnd_passive_connect()) Conn stale 172.19.142.119 at o2ib version 12/12 incarnation 1533927051163335/1533998625080752
Aug 11 12:15:52 ndc-oss5b kernel: LNet: 105990:0:(o2iblnd_cb.c:2410:kiblnd_passive_connect()) Conn stale 172.19.142.119 at o2ib version 12/12 incarnation 1533927051163335/1533998625080752
Aug 11 12:25:52 ndc-oss5b kernel: LNet: 105990:0:(o2iblnd_cb.c:2410:kiblnd_passive_connect()) Conn stale 172.19.142.119 at o2ib version 12/12 incarnation 1533927051163335/1533998625080752

This is a new node we brought online recently. Is it an indication that we have problem with
it OPA interface on the node? This machine has a 8160F CPU (OPA interface on chip).

Thanks,

Lixin Liu
High Performance Computing
Simon Fraser University

_______________________________________________
lustre-discuss mailing list
lustre-discuss at lists.lustre.org<mailto:lustre-discuss at lists.lustre.org>
http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lustre.org/pipermail/lustre-discuss-lustre.org/attachments/20180812/22302289/attachment-0001.html>