[LU-9823] LNet fails to come up when using lctl but works with lnetctl

• Type: Bug
• Resolution: Not a Bug
• Priority: Critical
• Affects Version/s: Lustre 2.10.1, Lustre 2.11.0
• Labels: IPv6
• Environment: Seen on various systems with Lustre 2.10 and Lustre 2.11.
• Severity: 3
• Assignee: Amir Shehata (Inactive)
• Reporter: James A Simmons

Description

On several systems when attempting to bring a lustre system this is reported:

[188273.054578] LNet: Added LNI 10.0.1.22@tcp [8/256/0/180]
[188273.054724] LNet: Accept secure, port 988
[191295.504584] Lustre: Lustre: Build Version: 2.10.0_dirty
[191300.858629] Lustre: 22140:0:(client.c:2114:ptlrpc_expire_one_request()) @@@ Request sent has timed out for slow reply: [sent 1501789735/real 1501789735]  req@ffff800fb09cfc80 x1574740673167376/t0(0) o250->MGC128.219.141.4@tcp@128.219.141.4@tcp:26/25 lens 520/544 e 0 to 1 dl 1501789740 ref 1 fl Rpc:XN/0/ffffffff rc 0/-1
[191301.858634] LustreError: 22036:0:(mgc_request.c:251:do_config_log_add()) MGC128.219.141.4@tcp: failed processing log, type 1: rc = -5
[191330.858099] Lustre: 22140:0:(client.c:2114:ptlrpc_expire_one_request()) @@@ Request sent has timed out for slow reply: [sent 1501789760/real 1501789760]  req@ffff800fb4910980 x1574740673167424/t0(0) o250->MGC128.219.141.4@tcp@128.219.141.4@tcp:26/25 lens 520/544 e 0 to 1 dl 1501789770 ref 1 fl Rpc:XN/0/ffffffff rc 0/-1
[191332.858106] LustreError: 15c-8: MGC128.219.141.4@tcp: The configuration from log 'legs-client' failed (-5). This may be the result of communication errors between this node and the MGS, a bad configuration, or other errors. See the syslog for more information.
[191332.858399] Lustre: Unmounted legs-client
[191332.859241] LustreError: 22036:0:(obd_mount.c:1505:lustre_fill_super()) Unable to mount  (-5)

After investigation this is a symptom of the LNet layer communication failure. This occurs when LNet has been setup with lctl but if one uses lnetctl then this issue appears to go away.

Issue Links

• is related to: LU-9567 sptlrpc rules are not being updated
• is related to: LU-9086 obd_config.c:1258:class_process_config()) no device for:
• is related to: LU-10391 LNET: Support IPv6