[Bug 1503305] Re: iLab: power kvm client after svc node reboot error inject test, multipath output show missing path and missing storage name
Mathieu Trudel-Lapierre
mathieu.tl at gmail.com
Wed Nov 4 00:32:15 UTC 2015
** Description changed:
+ [Impact]
+ Users of multipath will see incomplete output for the vendor and model values for devices where the first path is the path that is failed.
+
+ [Test Cases]
+ 1) Experience a failed path:
+ echo 1 > /sys/devices/<path to device>/delete
+ (the above is an example for forcibly removing a path for SCSI via the /sys filesystem)
+ 2) run: sudo multipath -ll
+ 3) Verify the output, see below.
+
+ Storage name (vendor and model usually) should always be displayed after
+ dm-* on the header line for a multipath device.
+
+ [Regression Potential]
+ This is limited to output and is thus purely a cosmetic fix. Possible issues leading from that fix would include crashes due to an incorrect patch or the output for vendor/model info for the device to still be incorrect.
+
+ ----
== Comment: #0 - Yueh Chyong Jackson <idaj at us.ibm.com> - 2015-01-27 16:54:58 ==
---Problem Description---
power kvm client ubuntu ilp1fc85apA4 after error injection test (SVC node reboot), the multipath output show one path with #:#:#:# and did not show the storage name:
mpath6 (3600507680284066f1800000000000172) dm-6 , <---------(missing storage name)
size=3.0G features='1 queue_if_no_path' hwhandler='0' wp=rw
|-+- policy='round-robin 0' prio=50 status=active
| |- #:#:#:# - #:# failed faulty running <----------------------------------------
| `- 4:0:7:0 sdbr 68:80 active ready running
`-+- policy='round-robin 0' prio=10 status=enabled
- |- 4:0:9:0 sdbq 68:64 active ready running
- `- 3:0:7:0 sdby 68:192 active ready running
+ |- 4:0:9:0 sdbq 68:64 active ready running
+ `- 3:0:7:0 sdby 68:192 active ready running
this is a storage IBM svc storage
+ ---uname output---
+ power kvm server: [root at ilp1fc85a ~]# uname -a Linux ilp1fc85a.tuc.stglabs.ibm.com 3.10.42-2018.1.pkvm2_1_1.46.ppc64 #1 SMP Fri Oct 17 10:41:35 CDT 2014 ppc64 ppc64 ppc64 GNU/Linux,
+ client uname : root at ilp1fc85apA4:~# uname -a Linux ilp1fc85apA4.tuc.stglabs.ibm.com 3.13.0-24-generic #46-Ubuntu SMP Thu Apr 10 19:09:21 UTC 2014 ppc64le ppc64le ppc64le GNU/Linux
-
- ---uname output---
- power kvm server: [root at ilp1fc85a ~]# uname -a Linux ilp1fc85a.tuc.stglabs.ibm.com 3.10.42-2018.1.pkvm2_1_1.46.ppc64 #1 SMP Fri Oct 17 10:41:35 CDT 2014 ppc64 ppc64 ppc64 GNU/Linux,
- client uname : root at ilp1fc85apA4:~# uname -a Linux ilp1fc85apA4.tuc.stglabs.ibm.com 3.13.0-24-generic #46-Ubuntu SMP Thu Apr 10 19:09:21 UTC 2014 ppc64le ppc64le ppc64le GNU/Linux
-
---Additional Hardware Info---
4 adapters, coho, bluefin, houston,houston FCoE, with following storages:
DS8800,TMS, xiv gen2, svc V7k ,texan, texan FCoE, client lun that use for running tests(data lun) are passthrough
access to machines:
power kvm server: ilp1fc85a 9.11.56.126 (ssh root/B!gm0ney)
client: ilp1fc85apA4 9.11.56.72 (ssh root/B!gmoney)
- the logs are save at:
+ the logs are save at:
ip: 9.11.56.167(root/teamw0rk), under file:
/root/bflocal/CYCLE18/CheckHostStatus_V2/CYCLE18_226/scripts/9.11.56.72.hostcheck.CHS.svc_nr.stop.20150127073959
ip: 9.11.56.168 (root/teamw0rk), under file:
/PACKAGE/errlogs/Cycle18/SVC_NR/20150127081651/
c_85_g.ilp1fc85apA4.SVC_NR.Cycle18.27Jan15.081758.tar.gz
-
-
- Machine Type = IBM Power System S824, product: 8286-42A,serial: 10CD74T
-
+ Machine Type = IBM Power System S824, product: 8286-42A,serial: 10CD74T
+
---Debugger---
A debugger is not configured
-
- Contact Information = ida jackson/idaj at us.ibm.com, cwchen at us.ibm.com
-
+
+ Contact Information = ida jackson/idaj at us.ibm.com, cwchen at us.ibm.com
+
Stack trace output:
- no
-
+ no
+
Oops output:
- no
-
+ no
+
System Dump Info:
- The system is not configured to capture a system dump.
-
- *Additional Instructions for ida jackson/idaj at us.ibm.com, cwchen at us.ibm.com:
+ The system is not configured to capture a system dump.
+
+ *Additional Instructions for ida jackson/idaj at us.ibm.com, cwchen at us.ibm.com:
-Post a private note with access information to the mach
- ine that the bug is occuring on.
+ ine that the bug is occuring on.
-Attach sysctl -a output output to the bug.
please let me know if other info is needed. thanks for looking into the
problem
== Comment: #9 - Vaishnavi Bhat <vaish123 at in.ibm.com> - 2015-02-11
04:11:40 ==
-
== Comment: #17 - Andrea M. Davis <amdavis at us.ibm.com> - 2015-03-03 17:57:51 ==
Please update the bug with the specific Ubunutu release please Ida or Chang.
== Comment: #18 - Andrea M. Davis <amdavis at us.ibm.com> - 2015-03-03 17:57:56 ==
Please update the bug with the specific Ubunutu release please Ida or Chang.
== Comment: #19 - Mauricio Faria De Oliveira <mauricfo at br.ibm.com> - 2015-03-04 08:17:23 ==
This is Ubuntu 14.04 GA (14.04.0), using the ISO/DVD install (kernel 3.13.0-24).
> client uname : root at ilp1fc85apA4:~# uname -a Linux
ilp1fc85apA4.tuc.stglabs.ibm.com 3.13.0-24-generic #46-Ubuntu SMP Thu
Apr 10 19:09:21 UTC 2014 ppc64le ppc64le ppc64le GNU/Linux
This version is way old (2014/04) for a bug reported on 2015/1, in
Ubuntu's fast pace.
Do the guest has any* other updates applied? (i.e., apt-get update && apt-get upgrade)
It seems there are no kernel upgrades as well?(i.e., apt-get update && apt-get dist-upgrade)
Any reason for not testing 14.04.1 or 14.04.2? (you can bump to it w/
dist-upgrade)
== Comment: #20 - Brian J. King <bjking1 at us.ibm.com> - 2015-03-04 10:11:57 ==
(In reply to comment #19)
> This is Ubuntu 14.04 GA (14.04.0), using the ISO/DVD install (kernel
> 3.13.0-24).
-
- Also keep in mind that you need at *least* 14.04.1 in order to be able to do PCI passthrough with PowerKVM.
+ Also keep in mind that you need at *least* 14.04.1 in order to be able
+ to do PCI passthrough with PowerKVM.
== Comment: #29 - Gabriel Krisman Bertazi <gbertazi at br.ibm.com> - 2015-09-30 15:38:23 ==
The attached patch should fix this issue. Mauricio, please tell me what you think.
== Comment: #30 - Mauricio Faria De Oliveira <mauricfo at br.ibm.com> - 2015-10-02 09:52:01 ==
Gabriel,
Looks good to me.
Good discussion yesterday while traveling.
Let's hear on dm-devel.
Thanks!
== Comment: #32 - Gabriel Krisman Bertazi <gbertazi at br.ibm.com> - 2015-10-06 08:28:24 ==
Patch is applied to Christophe's tree. <https://www.redhat.com/archives/dm-devel/2015-October/msg00021.html>.
Now, let's get it downstream into Ubuntu.
== Comment: #34 - Gabriel Krisman Bertazi <gbertazi at br.ibm.com> - 2015-10-06 08:55:40 ==
- Christophe Varoqui applied this patch to his multipath-tools tree <https://www.redhat.com/archives/dm-devel/2015-October/msg00022.html>.
+ Christophe Varoqui applied this patch to his multipath-tools tree <https://www.redhat.com/archives/dm-devel/2015-October/msg00022.html>.
We would like it to be backported to Ubuntu to close this bug.
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to multipath-tools in Ubuntu.
https://bugs.launchpad.net/bugs/1503305
Title:
iLab: power kvm client after svc node reboot error inject test,
multipath output show missing path and missing storage name
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/multipath-tools/+bug/1503305/+subscriptions
More information about the Ubuntu-server-bugs
mailing list