--- Comment From heinz-werner_se...@de.ibm.com 2017-10-11 03:36 EDT---
IBM Bugzilla status -> closed, Fix Released within Xenial
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/16706
--- Comment From jac...@de.ibm.com 2017-09-26 07:41 EDT---
Ok, just checked kernel 4.4.0-97 ... that looks much better:
root@x:~# uname -a
Linux mclint 4.4.0-97-generic #120-Ubuntu SMP Tue Sep 19 17:27:01 UTC 2017
s390x s390x s390x GNU/Linux
root@x:~# systool -v -m scsi_mod
Module
--- Comment From jac...@de.ibm.com 2017-09-22 04:36 EDT---
Just checked kernel 4.4.0-96. blk_mq() is still activated as default:
root@x:~# systool -v -m dm_mod
Module = "dm_mod"
Attributes:
uevent =
Parameters:
reserved_bio_based_ios= "16"
reserved_rq_based_ios= "256"
s
--- Comment From jac...@de.ibm.com 2017-09-11 10:46 EDT---
Ok, I think then the discussion is back at the point I addressed with my
comment from June 6th (comment 36 or 38)! If the Multi-Queue feature doesn't
work with Kernel 4.4 yet, then you must not deliver it (that's exactly why SUSE
--- Comment From heinz-werner_se...@de.ibm.com 2017-09-11 09:52 EDT---
No specific requirement for s390 know, for enabled this config options.
They can be set to N
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu
--- Comment From heinz-werner_se...@de.ibm.com 2017-09-04 04:52 EDT---
no specific cpu (un)plug tests where executed
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1670634
Titl
--- Comment From heinz-werner_se...@de.ibm.com 2017-08-30 05:15 EDT---
This function worked with SLES 12 SP2 kernel 4.4.74.
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1670634
--- Comment From heinz-werner_se...@de.ibm.com 2017-08-22 03:42 EDT---
@Juerg. Now you should have access to the box folder...
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1670634
--- Comment From jac...@de.ibm.com 2017-07-17 12:10 EDT---
@jsalisbury:
To get your point: If 4.11rc8 was a "good kernel" and 4.11rc7 the last "bad
kernel" you could try to get that very diff that fixes this problem and could
apply it to the stable 4.4-Kernel line, I agree that's a feasi
--- Comment From jac...@de.ibm.com 2017-07-13 10:26 EDT---
Just wanted to try the 4.11rc7 kernel and my DKMS-OpenAFS packages don't build
on it. I already experienced this once with a package from the kernel-ppa
repository:
root@mclint:~# file
/usr/src/linux-headers-4.11.0-041100rc7-gen
--- Comment From jac...@de.ibm.com 2017-07-12 10:45 EDT---
@jsalisbury:
Pardon to object in some respect, but I can't get the point why I should try
another 4.11rc-Kernel and what you refer to with a "bad kernel" and a "good
kernel".
The 4.11rc8 kernel seemed to be fine and I just got in
--- Comment From jac...@de.ibm.com 2017-07-10 04:34 EDT---
@jsalisbury:
I tried the kernel and didn't experience a hang as the system ran, but I ran
into a hang when I shut the system down after my test. The dump of that very
hang was uploaded from me to our Box account and Frank Heimes
--- Comment From jac...@de.ibm.com 2017-06-08 10:28 EDT---
I shut down the test system I used for the repros here yesterday. And all of a
sudden during the shutdown process the system ran into a hang scenario again. I
don't know whether that hang is related to the problem addressed here
--- Comment From jac...@de.ibm.com 2017-06-07 07:59 EDT---
Hello Benjamin,
you're right and let me rephrase in more detail:
The multiqueue feature is questioned here, but the feature has an impact to
multipathing, because Ubuntu-Xenial boots up and has the multiqueue feature
turned on a
--- Comment From bbl...@de.ibm.com 2017-06-06 12:46 EDT---
@jac...@de.ibm.com
Just to prevent confusion, its probably just a typo, but I think you
mean multi-queue (blk-mq and scsi-mq) - thats a different feature.
Multipathing (dm-multipath) should certainly work regardless.
--
You recei
--- Comment From jac...@de.ibm.com 2017-06-06 12:31 EDT---
@jsalisbury
Sorry for the late late late answer, but finally I've found time and
resources to test the 4.11-rc8 Kernel. And yes, this one looks promising
:-)
I was able to build OpenAFS with the pbuilder environment and also tryin
--- Comment From jac...@de.ibm.com 2017-04-06 11:53 EDT---
Same procedure as every kernel, /var/log/syslog:
Apr 6 16:17:29 mclint multipathd[881]: mpatha: sda - tur checker timed out
Apr 6 16:17:29 mclint multipathd[881]: 8:0: reinstated
Apr 6 16:17:29 mclint multipathd[881]: mpatha: sd
--- Comment From jac...@de.ibm.com 2017-03-30 08:50 EDT---
New Kernel, new hang ...
? 961.242228! INFO: task kworker/1:1:38 blocked for more than 120 seconds.
? 961.242235! Not tainted 4.4.0-71-generic #92-Ubuntu
? 961.242236! "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" dis
--- Comment From jac...@de.ibm.com 2017-03-28 12:23 EDT---
Just tried Kernel 4.4.0-70 ->
/var/log/syslog:
Mar 28 18:07:46 mclint multipathd[888]: mpatha: sda - tur checker timed out
Mar 28 18:07:46 mclint multipathd[888]: 8:0: reinstated
Mar 28 18:07:46 mclint multipathd[888]: mpatha: sdb
--- Comment From jac...@de.ibm.com 2017-03-14 12:07 EDT---
I was able to run into the hand now also without openafs, /var/log/syslog:
Mar 14 15:10:46 mclint multipathd[887]: mpatha: sda - tur checker timed out
Mar 14 15:10:46 mclint multipathd[887]: 8:0: reinstated
Mar 14 15:10:46 mclint m
--- Comment From jac...@de.ibm.com 2017-03-13 09:57 EDT---
Next iteration: I was able to make OpenAFS for the proposed kernel fixed and so
I was able to start the next pdebuild job and again I run into a hang
scenario:
Mar 13 14:30:32 mclint multipathd[881]: mpatha: sda - tur checke
--- Comment From jac...@de.ibm.com 2017-03-09 11:28 EDT---
Ok, I just tested the kernel from http://kernel.ubuntu.com/~rtg/lp1670634/ and
so far this looks good! I was able to Make the OpenAFS Ubuntu package three
times with pdebuild without running into a hang, and this was a very good
--- Comment From jac...@de.ibm.com 2017-03-08 05:56 EDT---
Just tried the newest Kernel 4.4.0-66, and I'm still running into the hang.
Here the final statements in /var/log/syslog (the lines, that never make it out
onto the disk):
Mar 8 11:26:31 mclint multipathd[955]: mpatha: sdb - tur
--- Comment From heinz-werner_se...@de.ibm.com 2017-03-07 11:44 EDT---
Please provide Debug-Info to this IBM_BOX folder
https://ibm.box.com/s/y10o4u7bcvc6nk7rgk2gfmk039ii5d1i
After Debugging this folder will be deleted.
--
You received this bug notification because you are a member of Ker
24 matches
Mail list logo
