------- Comment From [email protected] 2016-06-07 08:23 EDT------- ===================================END=================================== State: Verify by: cde00 on 31 May 2016 03:43:19 ====
== Comment: #1 - Application Cdeadmin <[email protected]> - 2016-03-21 15:55:11 ====== State: Verify by: cde00 on 31 May 2016 04:07:26 ==== ==== State: Verify by: byrneadw on 01 June 2016 11:03:58 ==== I loaded the test packages and can now successfully run HTX, I am not seeing EEH errors anymore but I do still see these "FLOGI failure Status:x3/x103 TMO:x14" errors. 2) from #1 execute ssh root@rcx2c360 (password is PASSW0RD) ==== State: Verify by: byrneadw on 01 June 2016 11:07:41 ==== I loaded the test packages and can now successfully run HTX, I am not seeing EEH errors anymore but I do still see these "FLOGI failure Status:x3/x103 TMO:x14" errors. I see a comment earlier that suggests it is normal ( update #31 from Guilherme ). I'm wondering if this is another event to add to our ignore list. In addition to the comment from Guilherme I can see a very similar event already in our ignore list due to feedback we received on SW315535 - event in that case was "FLOGI failure Status:x3/x103 TMO:x4". I'm not sure what the difference between TMO:x4 vs TMO:x14 is Is it ok to add "FLOGI failure Status:x3/x103 TMO:x14" events to our ignore list also or is more debug required ? root@rcx2c360:/tmp# dmesg -T --level=alert,crit,err [Wed Jun 1 13:26:03 2016] lpfc 0000:01:00.0: 0:1303 Link Up Event x1 received Data: x1 x0 x80 x0 x0 x0 0 [Wed Jun 1 13:26:03 2016] lpfc 0000:01:00.0: 0:(0):2858 FLOGI failure Status:x3/x103 TMO:x14 Data x1800 x0 [Wed Jun 1 13:26:03 2016] lpfc 0000:01:00.0: 0:(0):0100 FLOGI failure Status:x3/x103 TMO:x14 [Wed Jun 1 13:26:04 2016] lpfc 0000:01:00.1: 1:(0):2858 FLOGI failure Status:x3/x103 TMO:x14 Data x1800 x0 [Wed Jun 1 13:26:04 2016] lpfc 0000:01:00.1: 1:(0):0100 FLOGI failure Status:x3/x103 TMO:x14 ===>> If required, access to system: 1) Telnet rchd08e0.rchland.ibm.com ( login with userid=dlth1025, password=tim2fish ) 2) from #1 execute ssh root@rcx2c360 (password is PASSW0RD) ==== State: Verify by: byrneadw on 02 June 2016 17:11:41 ==== considering TMO:x4 and TMO:x14 are timeout values it suggests to me this is the same error we hit before with SW315535. The root cause of SW315535 was the mfg usage of wrap plugs on the Fibre ports for the purpose of running HTX. It resulted in the FLOGI message because a port cannot login to itself. The TMO values must have changed with Ubuntu 16.04 or new drivers as you mentioned above. This is the first system with a Bluefin running with 16.04 we've had. In the past all our systems with Bluefin were running in Habanero boxes with Ubuntu 14.04.03 In SW315535 Dan Eisenhauer commented : "That "error" message means the link came up, so I am conjecturing that there is a wrap plug installed, The FLOGI failed messages would be expected in that case since a port cannot login to itself. So, all those messages are expected and indicate that a wrap plug is installed and the adapters are functioning. Those can all be ignored." I removed the wrap plugs on our Garrison system and was able to boot many times without hitting this error. I think that matches the results of SW315535. I'll confirm with our HTX guy that we need to continue using these wraps for HTX. If we do need them I can ignore this message per Dan's analysis. I can change the current entry in our ignore list so that the regex doesn't include the TMO value as that might change and catch us out again. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1587316 Title: STC840.20:Alpine:alp7fp1:Ubuntu 16.04, BlueFin (SAN) EEH 6 times during boot then disabled SRC BA188002:b0314a_1612.840 To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1587316/+subscriptions -- ubuntu-bugs mailing list [email protected] https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
