<div dir="ltr"><div><div>Hi Russell,<br><br></div>Thanks for the IB subnet clues:<br><br>[root@pg-gpu01 ~]# ibv_devinfo<br>hca_id: mlx4_0<br> transport: InfiniBand (0)<br> fw_ver: 2.32.5100<br> node_guid: f452:1403:00f5:4620<br> sys_image_guid: f452:1403:00f5:4623<br> vendor_id: 0x02c9<br> vendor_part_id: 4099<br> hw_ver: 0x1<br> board_id: MT_1100120019<br> phys_port_cnt: 1<br> port: 1<br> state: PORT_ACTIVE (4)<br> max_mtu: 4096 (5)<br> active_mtu: 4096 (5)<br> sm_lid: 1<br> port_lid: 185<br> port_lmc: 0x00<br> link_layer: InfiniBand<br><br>[root@pg-gpu01 ~]# sminfo <br>sminfo: sm lid 1 sm guid 0xf452140300f62320, activity count 80878098 priority 0 state 3 SMINFO_MASTER<br><br></div>Looks like the rebooted node is able to connect/contact IB/IB subnetmanager<br><br><br><div><br></div></div><div class="gmail_extra"><br><div class="gmail_quote">On Mon, Apr 24, 2017 at 4:14 PM, Russell Dekema <span dir="ltr"><<a href="mailto:dekemar@umich.edu" target="_blank">dekemar@umich.edu</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">At first glance, this sounds like your Infiniband subnet manager may<br>
be down or malfunctioning. In this case, nodes which were already up<br>
when the subnet manager was working will continue to be able to<br>
communicate over IB, but nodes which reboot after the SM goes down<br>
will not.<br>
<br>
You can test this theory by running the 'ibv_devinfo' command on one<br>
of your rebooted nodes. If the relevant IB port is in state PORT_INIT,<br>
this confirms there is a problem with your subnet manager.<br>
<br>
Sincerely,<br>
Rusty Dekema<br>
<div><div class="h5"><br>
<br>
<br>
<br>
On Mon, Apr 24, 2017 at 9:57 AM, Strikwerda, Ger<br>
<<a href="mailto:g.j.c.strikwerda@rug.nl">g.j.c.strikwerda@rug.nl</a>> wrote:<br>
> Hi everybody,<br>
><br>
> Here at the university of Groningen we are now experiencing a strange Lustre<br>
> error. If a client reboots, it fails to mount the Lustre storage. The client<br>
> is not able to reach the MSG service. The storage and nodes are<br>
> communicating over IB and unitil now without any problems. It looks like an<br>
> issue inside LNET. Clients cannot LNET ping/connect the metadata and or<br>
> storage. But the clients are able to LNET ping each other. Clients which not<br>
> have been rebooted, are working fine and have their mounts on our Lustre<br>
> filesystem.<br>
><br>
> Lustre client log:<br>
><br>
> Lustre: Lustre: Build Version: 2.5.3-RC1--PRISTINE-2.6.32-<wbr>573.el6.x86_64<br>
> LNet: Added LNI 172.23.54.51@o2ib [8/256/0/180]<br>
><br>
> LustreError: 15c-8: MGC172.23.55.211@o2ib: The configuration from log<br>
> 'pgdata01-client' failed (-5). This may be the result of communication<br>
> errors between this node and the MGS, a bad configuration, or other errors.<br>
> See the syslog for more information.<br>
> LustreError: 3812:0:(llite_lib.c:1046:ll_<wbr>fill_super()) Unable to process<br>
> log: -5<br>
> Lustre: Unmounted pgdata01-client<br>
> LustreError: 3812:0:(obd_mount.c:1325:<wbr>lustre_fill_super()) Unable to mount<br>
> (-5)<br>
> LNetError: 2882:0:(o2iblnd_cb.c:2587:<wbr>kiblnd_rejected()) 172.23.55.212@o2ib<br>
> rejected: consumer defined fatal error<br>
> LNetError: 2882:0:(o2iblnd_cb.c:2587:<wbr>kiblnd_rejected()) Skipped 1 previous<br>
> similar message<br>
> Lustre: 3765:0:(client.c:1918:ptlrpc_<wbr>expire_one_request()) @@@ Request sent<br>
> has failed due to network error: [sent 1492789626/real 1492789626]<br>
> req@ffff88105af2cc00 x1565303228072004/t0(0)<br>
> o250->MGC172.23.55.211@o2ib@<wbr>172.23.55.212@o2ib:26/25 lens 400/544 e 0 to 1<br>
> dl 1492789631 ref 1 fl Rpc:XN/0/ffffffff rc 0/-1<br>
> Lustre: 3765:0:(client.c:1918:ptlrpc_<wbr>expire_one_request()) Skipped 1<br>
> previous similar message<br>
> LustreError: 3826:0:(client.c:1083:ptlrpc_<wbr>import_delay_req()) @@@ send limit<br>
> expired req@ffff882041ffc000 x1565303228071996/t0(0)<br>
> o101->MGC172.23.55.211@o2ib@<wbr>172.23.55.211@o2ib:26/25 lens 328/344 e 0 to 0<br>
> dl 0 ref 2 fl Rpc:W/0/ffffffff rc 0/-1<br>
> LustreError: 3826:0:(client.c:1083:ptlrpc_<wbr>import_delay_req()) Skipped 2<br>
> previous similar messages<br>
> LustreError: 15c-8: MGC172.23.55.211@o2ib: The configuration from log<br>
> 'pghome01-client' failed (-5). This may be the result of communication<br>
> errors between this node and the MGS, a bad configuration, or other errors.<br>
> See the syslog for more information.<br>
> LustreError: 3826:0:(llite_lib.c:1046:ll_<wbr>fill_super()) Unable to process<br>
> log: -5<br>
><br>
> LNetError: 2882:0:(o2iblnd_cb.c:2587:<wbr>kiblnd_rejected()) 172.23.55.212@o2ib<br>
> rejected: consumer defined fatal error<br>
> LNetError: 2882:0:(o2iblnd_cb.c:2587:<wbr>kiblnd_rejected()) Skipped 1 previous<br>
> similar message<br>
> LNet: 3755:0:(o2iblnd_cb.c:475:<wbr>kiblnd_rx_complete()) Rx from<br>
> 172.23.55.211@o2ib failed: 5<br>
> LNetError: 2882:0:(o2iblnd_cb.c:2587:<wbr>kiblnd_rejected()) 172.23.55.211@o2ib<br>
> rejected: consumer defined fatal error<br>
> LNetError: 2882:0:(o2iblnd_cb.c:2587:<wbr>kiblnd_rejected()) Skipped 1 previous<br>
> similar message<br>
> LNet: 2882:0:(o2iblnd_cb.c:2072:<wbr>kiblnd_peer_connect_failed()) Deleting<br>
> messages for 172.23.55.211@o2ib: connection failed<br>
> LNet: 2882:0:(o2iblnd_cb.c:2072:<wbr>kiblnd_peer_connect_failed()) Deleting<br>
> messages for 172.23.55.212@o2ib: connection failed<br>
> LNet: 3754:0:(o2iblnd_cb.c:475:<wbr>kiblnd_rx_complete()) Rx from<br>
> 172.23.55.212@o2ib failed: 5<br>
> LNet: 3754:0:(o2iblnd_cb.c:475:<wbr>kiblnd_rx_complete()) Skipped 17 previous<br>
> similar messages<br>
> LNet: 2882:0:(o2iblnd_cb.c:2072:<wbr>kiblnd_peer_connect_failed()) Deleting<br>
> messages for 172.23.55.211@o2ib: connection failed<br>
> LNet: 3754:0:(o2iblnd_cb.c:475:<wbr>kiblnd_rx_complete()) Rx from<br>
> 172.23.55.212@o2ib failed: 5<br>
> LNet: 2882:0:(o2iblnd_cb.c:2072:<wbr>kiblnd_peer_connect_failed()) Deleting<br>
> messages for 172.23.55.212@o2ib: connection failed<br>
><br>
> LNET ping of a metadata-node:<br>
><br>
> [root@pg-gpu01 ~]# lctl ping 172.23.55.211@o2ib<br>
> failed to ping 172.23.55.211@o2ib: Input/output error<br>
><br>
> LNET ping of the number 2 metadata-node:<br>
><br>
> [root@pg-gpu01 ~]# lctl ping 172.23.55.212@o2ib<br>
> failed to ping 172.23.55.212@o2ib: Input/output error<br>
><br>
> LNET ping of a random compute-node:<br>
><br>
> [root@pg-gpu01 ~]# lctl ping 172.23.52.5@o2ib<br>
> 12345-0@lo<br>
> 12345-172.23.52.5@o2ib<br>
><br>
> LNET to OST01:<br>
><br>
> [root@pg-gpu01 ~]# lctl ping 172.23.55.201@o2ib<br>
> failed to ping 172.23.55.201@o2ib: Input/output error<br>
><br>
> LNET to OST02:<br>
><br>
> [root@pg-gpu01 ~]# lctl ping 172.23.55.202@o2ib<br>
> failed to ping 172.23.55.202@o2ib: Input/output error<br>
><br>
> 'normal' pings (on ip level) works fine:<br>
><br>
> [root@pg-gpu01 ~]# ping 172.23.55.201<br>
> PING 172.23.55.201 (172.23.55.201) 56(84) bytes of data.<br>
> 64 bytes from <a href="http://172.23.55.201" rel="noreferrer" target="_blank">172.23.55.201</a>: icmp_seq=1 ttl=64 time=0.741 ms<br>
><br>
> [root@pg-gpu01 ~]# ping 172.23.55.202<br>
> PING 172.23.55.202 (172.23.55.202) 56(84) bytes of data.<br>
> 64 bytes from <a href="http://172.23.55.202" rel="noreferrer" target="_blank">172.23.55.202</a>: icmp_seq=1 ttl=64 time=0.704 ms<br>
><br>
> lctl on a rebooted node:<br>
><br>
> [root@pg-gpu01 ~]# lctl dl<br>
><br>
> lctl on a not rebooted node:<br>
><br>
> [root@pg-node005 ~]# lctl dl<br>
> 0 UP mgc MGC172.23.55.211@o2ib 94bd1c8a-512f-b920-9a4e-<wbr>a6aced3d386d 5<br>
> 1 UP lov pgtemp01-clilov-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 4<br>
> 2 UP lmv pgtemp01-clilmv-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 4<br>
> 3 UP mdc pgtemp01-MDT0000-mdc-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 5<br>
> 4 UP osc pgtemp01-OST0001-osc-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 5<br>
> 5 UP osc pgtemp01-OST0003-osc-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 5<br>
> 6 UP osc pgtemp01-OST0005-osc-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 5<br>
> 7 UP osc pgtemp01-OST0007-osc-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 5<br>
> 8 UP osc pgtemp01-OST0009-osc-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 5<br>
> 9 UP osc pgtemp01-OST000b-osc-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 5<br>
> 10 UP osc pgtemp01-OST000d-osc-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 5<br>
> 11 UP osc pgtemp01-OST000f-osc-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 5<br>
> 12 UP osc pgtemp01-OST0011-osc-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 5<br>
> 13 UP osc pgtemp01-OST0002-osc-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 5<br>
> 14 UP osc pgtemp01-OST0004-osc-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 5<br>
> 15 UP osc pgtemp01-OST0006-osc-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 5<br>
> 16 UP osc pgtemp01-OST0008-osc-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 5<br>
> 17 UP osc pgtemp01-OST000a-osc-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 5<br>
> 18 UP osc pgtemp01-OST000c-osc-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 5<br>
> 19 UP osc pgtemp01-OST000e-osc-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 5<br>
> 20 UP osc pgtemp01-OST0010-osc-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 5<br>
> 21 UP osc pgtemp01-OST0012-osc-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 5<br>
> 22 UP osc pgtemp01-OST0013-osc-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 5<br>
> 23 UP osc pgtemp01-OST0015-osc-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 5<br>
> 24 UP osc pgtemp01-OST0017-osc-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 5<br>
> 25 UP osc pgtemp01-OST0014-osc-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 5<br>
> 26 UP osc pgtemp01-OST0016-osc-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 5<br>
> 27 UP osc pgtemp01-OST0018-osc-<wbr>ffff88206906d400<br>
> 281c441f-8aa3-ab56-8812-<wbr>e459d308f47c 5<br>
> 28 UP lov pgdata01-clilov-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 4<br>
> 29 UP lmv pgdata01-clilmv-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 4<br>
> 30 UP mdc pgdata01-MDT0000-mdc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 31 UP osc pgdata01-OST0001-osc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 32 UP osc pgdata01-OST0003-osc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 33 UP osc pgdata01-OST0005-osc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 34 UP osc pgdata01-OST0007-osc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 35 UP osc pgdata01-OST0009-osc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 36 UP osc pgdata01-OST000b-osc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 37 UP osc pgdata01-OST000d-osc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 38 UP osc pgdata01-OST000f-osc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 39 UP osc pgdata01-OST0002-osc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 40 UP osc pgdata01-OST0004-osc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 41 UP osc pgdata01-OST0006-osc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 42 UP osc pgdata01-OST0008-osc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 43 UP osc pgdata01-OST000a-osc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 44 UP osc pgdata01-OST000c-osc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 45 UP osc pgdata01-OST000e-osc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 46 UP osc pgdata01-OST0010-osc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 47 UP osc pgdata01-OST0013-osc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 48 UP osc pgdata01-OST0015-osc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 49 UP osc pgdata01-OST0017-osc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 50 UP osc pgdata01-OST0014-osc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 51 UP osc pgdata01-OST0016-osc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 52 UP osc pgdata01-OST0018-osc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 53 UP osc pgdata01-OST0019-osc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 54 UP osc pgdata01-OST001a-osc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 55 UP osc pgdata01-OST001b-osc-<wbr>ffff88204bab6400<br>
> 996b1742-82eb-281c-c322-<wbr>e244672d5225 5<br>
> 56 UP lov pghome01-clilov-<wbr>ffff88204bb50000<br>
> 9ae8f2a9-1cdf-901f-160c-<wbr>66f70e4c10d1 4<br>
> 57 UP lmv pghome01-clilmv-<wbr>ffff88204bb50000<br>
> 9ae8f2a9-1cdf-901f-160c-<wbr>66f70e4c10d1 4<br>
> 58 UP mdc pghome01-MDT0000-mdc-<wbr>ffff88204bb50000<br>
> 9ae8f2a9-1cdf-901f-160c-<wbr>66f70e4c10d1 5<br>
> 59 UP osc pghome01-OST0011-osc-<wbr>ffff88204bb50000<br>
> 9ae8f2a9-1cdf-901f-160c-<wbr>66f70e4c10d1 5<br>
> 60 UP osc pghome01-OST0012-osc-<wbr>ffff88204bb50000<br>
> 9ae8f2a9-1cdf-901f-160c-<wbr>66f70e4c10d1 5<br>
><br>
> Please help, any clues/advice/hints/tips are appricated<br>
><br>
> --<br>
><br>
> Vriendelijke groet,<br>
><br>
> Ger Strikwerda<br>
> Chef Special<br>
> Rijksuniversiteit Groningen<br>
> Centrum voor Informatie Technologie<br>
> Unit Pragmatisch Systeembeheer<br>
><br>
> Smitsborg<br>
> Nettelbosje 1<br>
> 9747 AJ Groningen<br>
> Tel. 050 363 9276<br>
><br>
> "God is hard, God is fair<br>
> some men he gave brains, others he gave hair"<br>
><br>
><br>
</div></div>> ______________________________<wbr>_________________<br>
> lustre-discuss mailing list<br>
> <a href="mailto:lustre-discuss@lists.lustre.org">lustre-discuss@lists.lustre.<wbr>org</a><br>
> <a href="http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org" rel="noreferrer" target="_blank">http://lists.lustre.org/<wbr>listinfo.cgi/lustre-discuss-<wbr>lustre.org</a><br>
><br>
</blockquote></div><br><br clear="all"><br>-- <br><div class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><pre cols="72"><span style="font-family:arial,helvetica,sans-serif"><font size="2">Vriendelijke groet, <br><br>Ger Strikwerda
</font></span><span style="font-family:arial,helvetica,sans-serif"><font size="2">Chef Special
Rijksuniversiteit Groningen
Centrum voor Informatie Technologie
Unit Pragmatisch Systeembeheer
Smitsborg
Nettelbosje 1
9747 AJ Groningen
Tel. 050 363 9276
<br>"God is hard, God is fair<br> some men he gave brains, others he gave hair"<br></font></span></pre></div></div></div></div></div></div></div></div></div></div></div></div>
</div>