<div dir="ltr"><div dir="ltr">Take a look at this: <a href="https://jira.whamcloud.com/browse/LU-11840">https://jira.whamcloud.com/browse/LU-11840</a></div><div>Let me know if this is the same issue you're seeing.<br></div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Tue, 5 Mar 2019 at 14:04, Amir Shehata <<a href="mailto:amir.shehata.whamcloud@gmail.com">amir.shehata.whamcloud@gmail.com</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"><div dir="ltr"><div>Hi Riccardo,</div><div><br></div><div>It's not LNet Health. It's Dynamic Discovery. What's happening is that 2.12 is discovering all the interfaces on the peer. That's why you see all the interfaces in the peer show.</div><div><br></div><div>Multi-Rail doesn't enable o2ib. It just sees it. If the node doing the discovery has only tcp, then it should never try to connect over the o2ib. <br></div><div><br></div><div>Are you able to do a "lnetctl ping 172.21.48.250@tcp" from the MDS multiple times? Do you see the ping failing intermittently?</div><div><br></div><div>What should happen is that when the MDS (running 2.12) tries to talk to the peer you have identified, then it'll discover its interfaces. But then should realize that it can only reach it on the tcp network, since that's the only network configured on the MDS.</div><div><br></div><div>It might help, if you just configure LNet only, on the MDS and the peer and run a simple</div><div>lctl set_param debug=+"net neterror"<br></div><div> lnetctl ping <></div><div>lctl dk >log</div><div><br></div><div>If you can share the debug output, it'll help to pinpoint the problem.</div><div><br></div><div>thanks</div><div>amir<br></div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Tue, 5 Mar 2019 at 12:30, Riccardo Veraldi <<a href="mailto:Riccardo.Veraldi@cnaf.infn.it" target="_blank">Riccardo.Veraldi@cnaf.infn.it</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">I think I figured out the problem.<br>
My problem is related to Lnet Network Health feature: <br>
<a href="https://jira.whamcloud.com/browse/LU-9120" rel="noreferrer" target="_blank">https://jira.whamcloud.com/browse/LU-9120</a><br>
the lustre MDS and the lsutre client having same version 2.12.0 <br>
negotiate a Multi-rail peer connection while this does not happen with <br>
the other clients (2.10.5). So what happens is that both IB and tcp are <br>
being used during transfers.<br>
tcp is only for connecting to the MDS, IB only to connect to the OSS <br>
anyway Multi-rail is enabled by default between the MDS,OSS and client. <br>
This messes up the situation. the MDS has only one TCP interface and <br>
cannot communicate by IB but in the "lnetctl peer show" a NID @o2ib <br>
shows up and it should not. At this point the MDS tries to connect to <br>
the client using IB and it will never work because there is no IB on the <br>
MDS.<br>
MDS Lnet configuration:<br>
<br>
net:<br>
- net type: lo<br>
local NI(s):<br>
- nid: 0@lo<br>
status: up<br>
- net type: tcp<br>
local NI(s):<br>
- nid: 172.21.49.233@tcp<br>
status: up<br>
interfaces:<br>
0: eth0<br>
<br>
but if I look at lnetctl peer show I See<br>
<br>
- primary nid: 172.21.52.88@o2ib<br>
Multi-Rail: True<br>
peer ni:<br>
- nid: 172.21.48.250@tcp<br>
state: NA<br>
- nid: 172.21.52.88@o2ib<br>
state: NA<br>
- nid: 172.21.48.250@tcp1<br>
state: NA<br>
- nid: 172.21.48.250@tcp2<br>
state: NA<br>
<br>
there should be no o2ib nid but Multi-rail for some reason enables it.<br>
I do not have problems with the other clients (non 2.12.0)<br>
<br>
How can I disable Multi-rail on 2.12.0 ??<br>
<br>
thank you<br>
<br>
<br>
<br>
On 3/5/19 12:14 PM, Patrick Farrell wrote:<br>
> Riccardo,<br>
><br>
> Since 2.12 is still a relatively new maintenance release, it would be helpful if you could open an LU and provide more detail there - Such as what clients were doing, if you were using any new features (like DoM or FLR), and full dmesg from the clients and servers involved in these evictions.<br>
><br>
> - Patrick<br>
><br>
> \ufeffOn 3/5/19, 11:50 AM, "lustre-discuss on behalf of Riccardo Veraldi" <<a href="mailto:lustre-discuss-bounces@lists.lustre.org" target="_blank">lustre-discuss-bounces@lists.lustre.org</a> on behalf of <a href="mailto:Riccardo.Veraldi@cnaf.infn.it" target="_blank">Riccardo.Veraldi@cnaf.infn.it</a>> wrote:<br>
><br>
> Hello,<br>
> <br>
> I have quite a big issue on my Lustre 2.12.0 MDS/MDT.<br>
> <br>
> Clients moving data to the OSS occur into a locking problem I never met<br>
> before.<br>
> <br>
> The clients are mostly 2.10.5 except for one which is 2.12.0 but<br>
> regardless the client version the problem is still there.<br>
> <br>
> So these are the errors I see on hte MDS/MDT. When this happens<br>
> everything just hangs. If I reboot the MDS everything is back to<br>
> normality but it happened already 2 times in 3 days and it is disrupting.<br>
> <br>
> Any hints ?<br>
> <br>
> Is it feasible to downgrade from 2.12.0 to 2.10.6 ?<br>
> <br>
> thanks<br>
> <br>
> Mar 5 11:10:33 psmdsana1501 kernel: Lustre:<br>
> 7898:0:(client.c:2132:ptlrpc_expire_one_request()) @@@ Request sent has<br>
> failed due to network error: [sent 1551813033/real 1551813033]<br>
> req@ffff9fdcbecd0300 x1626845000210688/t0(0)<br>
> o104-><a href="mailto:ana15-MDT0000@172.21.52." target="_blank">ana15-MDT0000@172.21.52.</a>87@o2ib:15/16 lens 296/224 e 0 to 1 dl<br>
> 1551813044 ref 1 fl Rpc:eX/0/ffffffff rc 0/-1<br>
> Mar 5 11:10:33 psmdsana1501 kernel: Lustre:<br>
> 7898:0:(client.c:2132:ptlrpc_expire_one_request()) Skipped 50552576<br>
> previous similar messages<br>
> Mar 5 11:13:03 psmdsana1501 kernel: LustreError:<br>
> 7898:0:(ldlm_lockd.c:682:ldlm_handle_ast_error()) ### client (nid<br>
> 172.21.52.87@o2ib) failed to reply to blocking AST (req@ffff9fdcbecd0300<br>
> x1626845000210688 status 0 rc -110), evict it ns: mdt-ana15-MDT0000_UUID<br>
> lock: ffff9fde9b6873c0/0x9824623d2148ef38 lrc: 4/0,0 mode: PR/PR res:<br>
> [0x2000013a9:0x1d347:0x0].0x0 bits 0x13/0x0 rrc: 5 type: IBT flags:<br>
> 0x60200400000020 nid: 172.21.52.87@o2ib remote: 0xd8efecd6e7621e63<br>
> expref: 8 pid: 7898 timeout: 333081 lvb_type: 0<br>
> Mar 5 11:13:03 psmdsana1501 kernel: LustreError: 138-a: ana15-MDT0000:<br>
> A client on nid 172.21.52.87@o2ib was evicted due to a lock blocking<br>
> callback time out: rc -110<br>
> Mar 5 11:13:03 psmdsana1501 kernel: LustreError:<br>
> 5321:0:(ldlm_lockd.c:256:expired_lock_main()) ### lock callback timer<br>
> expired after 150s: evicting client at 172.21.52.87@o2ib ns:<br>
> mdt-ana15-MDT0000_UUID lock: ffff9fde9b6873c0/0x9824623d2148ef38 lrc:<br>
> 3/0,0 mode: PR/PR res: [0x2000013a9:0x1d347:0x0].0x0 bits 0x13/0x0 rrc:<br>
> 5 type: IBT flags: 0x60200400000020 nid: 172.21.52.87@o2ib remote:<br>
> 0xd8efecd6e7621e63 expref: 9 pid: 7898 timeout: 0 lvb_type: 0<br>
> Mar 5 11:13:04 psmdsana1501 kernel: Lustre: ana15-MDT0000: Connection<br>
> restored to 59c5a826-f4e9-0dd0-8d4f-08c204f25941 (at 172.21.52.87@o2ib)<br>
> Mar 5 11:15:34 psmdsana1501 kernel: LustreError:<br>
> 7898:0:(ldlm_lockd.c:682:ldlm_handle_ast_error()) ### client (nid<br>
> 172.21.52.142@o2ib) failed to reply to blocking AST<br>
> (req@ffff9fde2d393600 x1626845000213776 status 0 rc -110), evict it ns:<br>
> mdt-ana15-MDT0000_UUID lock: ffff9fde9b6858c0/0x9824623d2148efee lrc:<br>
> 4/0,0 mode: PR/PR res: [0x2000013ac:0x1:0x0].0x0 bits 0x13/0x0 rrc: 3<br>
> type: IBT flags: 0x60200400000020 nid: 172.21.52.142@o2ib remote:<br>
> 0xbb35541ea6663082 expref: 9 pid: 7898 timeout: 333232 lvb_type: 0<br>
> Mar 5 11:15:34 psmdsana1501 kernel: LustreError: 138-a: ana15-MDT0000:<br>
> A client on nid 172.21.52.142@o2ib was evicted due to a lock blocking<br>
> callback time out: rc -110<br>
> Mar 5 11:15:34 psmdsana1501 kernel: LustreError:<br>
> 5321:0:(ldlm_lockd.c:256:expired_lock_main()) ### lock callback timer<br>
> expired after 151s: evicting client at 172.21.52.142@o2ib ns:<br>
> mdt-ana15-MDT0000_UUID lock: ffff9fde9b6858c0/0x9824623d2148efee lrc:<br>
> 3/0,0 mode: PR/PR res: [0x2000013ac:0x1:0x0].0x0 bits 0x13/0x0 rrc: 3<br>
> type: IBT flags: 0x60200400000020 nid: 172.21.52.142@o2ib remote:<br>
> 0xbb35541ea6663082 expref: 10 pid: 7898 timeout: 0 lvb_type: 0<br>
> Mar 5 11:15:34 psmdsana1501 kernel: Lustre: ana15-MDT0000: Connection<br>
> restored to 9d49a115-646b-c006-fd85-000a4b90019a (at 172.21.52.142@o2ib)<br>
> Mar 5 11:20:33 psmdsana1501 kernel: Lustre:<br>
> 7898:0:(client.c:2132:ptlrpc_expire_one_request()) @@@ Request sent has<br>
> failed due to network error: [sent 1551813633/real 1551813633]<br>
> req@ffff9fdcc2a95100 x1626845000222624/t0(0)<br>
> o104-><a href="mailto:ana15-MDT0000@172.21.52." target="_blank">ana15-MDT0000@172.21.52.</a>87@o2ib:15/16 lens 296/224 e 0 to 1 dl<br>
> 1551813644 ref 1 fl Rpc:eX/2/ffffffff rc 0/-1<br>
> Mar 5 11:20:33 psmdsana1501 kernel: Lustre:<br>
> 7898:0:(client.c:2132:ptlrpc_expire_one_request()) Skipped 23570550<br>
> previous similar messages<br>
> Mar 5 11:22:46 psmdsana1501 kernel: LustreError:<br>
> 7898:0:(ldlm_lockd.c:682:ldlm_handle_ast_error()) ### client (nid<br>
> 172.21.52.87@o2ib) failed to reply to blocking AST (req@ffff9fdcc2a95100<br>
> x1626845000222624 status 0 rc -110), evict it ns: mdt-ana15-MDT0000_UUID<br>
> lock: ffff9fde86ffdf80/0x9824623d2148f23a lrc: 4/0,0 mode: PR/PR res:<br>
> [0x2000013ae:0x1:0x0].0x0 bits 0x13/0x0 rrc: 3 type: IBT flags:<br>
> 0x60200400000020 nid: 172.21.52.87@o2ib remote: 0xd8efecd6e7621eb7<br>
> expref: 9 pid: 7898 timeout: 333665 lvb_type: 0<br>
> Mar 5 11:22:46 psmdsana1501 kernel: LustreError: 138-a: ana15-MDT0000:<br>
> A client on nid 172.21.52.87@o2ib was evicted due to a lock blocking<br>
> callback time out: rc -110<br>
> Mar 5 11:22:46 psmdsana1501 kernel: LustreError:<br>
> 5321:0:(ldlm_lockd.c:256:expired_lock_main()) ### lock callback timer<br>
> expired after 150s: evicting client at 172.21.52.87@o2ib ns:<br>
> mdt-ana15-MDT0000_UUID lock: ffff9fde86ffdf80/0x9824623d2148f23a lrc:<br>
> 3/0,0 mode: PR/PR res: [0x2000013ae:0x1:0x0].0x0 bits 0x13/0x0 rrc: 3<br>
> type: IBT flags: 0x60200400000020 nid: 172.21.52.87@o2ib remote:<br>
> 0xd8efecd6e7621eb7 expref: 10 pid: 7898 timeout: 0 lvb_type: 0<br>
> <br>
> <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>
> _______________________________________________<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>
_______________________________________________<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>
</blockquote></div>
</blockquote></div>