| Author |
Message
|
| Gideon |
 Posted: Tue Aug 20, 2013 12:07 pm Post subject: Multi-Instance Qmgr status |
 |
|
Chevalier
Joined: 18 Aug 2009
Posts: 403
|
I created a multi instance queue manager named mqha with crtmqm and addmqinf
I started both instnaces with strmqm -x
On the standby instance I showed the following status:
| Code: |
-bash-4.1$ dspmq -x -o standby -o status
QMNAME(WQM) STATUS(Ended immediately) STANDBY(Not applicable)
QMNAME(WQM1) STATUS(Status not available) STANDBY(Not applicable)
QMNAME(mqha) STATUS(Running as standby) STANDBY(Permitted)
INSTANCE(btvm1) MODE(Active)
INSTANCE(btvm2) MODE(Standby) |
From the documentation, this appears correct behavior
So I try to display the status on the active instance, as follows:
| Code: |
-bash-4.1$ dspmq -x -o standby -o status
QMNAME(WQM) STATUS(Ended immediately) STANDBY(Not applicable)
QMNAME(WQM1) STATUS(Status not available) STANDBY(Not applicable)
|
From this point it just hangs there. If I hit the ENTER key, it will go to th next line, sort of like runmqsc. I eventually have to use the kill command on the dspmq to kill it
According to the documentation, this should be correct syntax, in fact it is the same syntax I use successfully on the standby instance.
How can I resolve this ? |
|
| Back to top |
|
 |
| JosephGramig |
| Posted: Tue Aug 20, 2013 12:44 pm Post subject: |
|
|
Grand Master
Joined: 09 Feb 2006
Posts: 1244
Location: Gold Coast of Florida, USA
|
So, this is interesting to look at... From the looks of this, you have installed WMQ Server twice on the same box. Instances btvm1 and btvm2.
You are trying to run MI of a Qmgr on one box between two WMQ instances?
What is the point of this? Not HA.
I would be surprised if this configuration is supported. |
|
| Back to top |
|
|
| Gideon |
| Posted: Tue Aug 20, 2013 1:02 pm Post subject: |
|
|
Chevalier
Joined: 18 Aug 2009
Posts: 403
|
On Machine A I created a multi-instance qmgr, as follows
| Code: |
| crtmqm -md /mqha/qmgrs -ld /mqha/logs mqha |
with /mqha being an external RAID
On machine B (also with /mqha as the same xternal RAID, I added an instance to this qmgr
| Code: |
| addmqinf -v Name=mqha -v Directory=mqha -v Prefix=/var/mqm -v DataPath=/mqha/qmgrs/mqha |
My understanding is that this creates a multi instance qmgr that shares the RAID, but not installing the same qmgr twice. Please correct me if I am wrong |
|
| Back to top |
|
|
| Gideon |
| Posted: Tue Aug 20, 2013 1:07 pm Post subject: |
|
|
Chevalier
Joined: 18 Aug 2009
Posts: 403
|
Now when I tried to end the qmgr, to fail over to the other instance, I get the following
| Code: |
-bash-4.1$ endmqm -s mqha
^C^C^C
|
It just sits there for a half hour, just like the dspmq command
Whatever command I use on the mqha, it just hangs
Any idea on how I should resove this ?
Or what could be the cause ? |
|
| Back to top |
|
|
| mqjeff |
| Posted: Tue Aug 20, 2013 1:13 pm Post subject: |
|
|
Grand Master
Joined: 25 Jun 2008
Posts: 17447
|
Did you confirm that /mqha is fully mounted and readable/writeable on both machines at all times?
Did you run the mq command to verify that the file system met the necessary requirements for an MI qmgr support? |
|
| Back to top |
|
|
| Gideon |
| Posted: Wed Aug 21, 2013 9:38 am Post subject: |
|
|
Chevalier
Joined: 18 Aug 2009
Posts: 403
|
I mounted as nfs4
| Code: |
| mount -t nfs4 10.14.12.23:/ /mqha |
The POSIX test completed successfully on both machines
The File lock test completed successfully.
The test on the concurrent writes failed on the second instance machine
Machine 1:
| Code: |
-bash-4.1$ cd /opt/mqm/bin
-bash-4.1$
-bash-4.1$ ls -la /mqha
total 8
drwxrwxr-x 7 mqm mqm 4096 Aug 21 2013 .
dr-xr-xr-x. 30 root root 4096 Aug 21 06:44 ..
drwxr-xr-x 2 nobody nobody 256 Aug 20 14:32 craig
drwxrwxr-x 3 mqm mqm 256 Aug 20 14:16 logs
drwxr-xr-x 2 nobody nobody 256 Aug 8 14:49 lost+found
drwxrwxr-x 3 mqm mqm 256 Aug 20 14:16 qmgrs
drwxr-xr-x 2 nobody nobody 256 Aug 19 18:53 temp |
-bash-4.1$ ./amqmfsck -c /mqha
Start a second copy of this program with the same parameters on another server.
Writing to test file. This will normally complete within about 60 seconds.
.
The tests on the directory completed successfully.
-bash-4.1$
Machine 2:
| Code: |
-bash-4.1$ cd /opt/mqm/bin
-bash-4.1$
-bash-4.1$ ls -la /mqha
total 8
drwxrwxr-x. 7 mqm mqm 4096 Aug 21 2013 .
dr-xr-xr-x. 27 root root 4096 Aug 20 11:25 ..
drwxr-xr-x. 2 nobody nobody 256 Aug 20 14:32 craig
drwxrwxr-x. 3 mqm mqm 256 Aug 20 14:16 logs
drwxr-xr-x. 2 nobody nobody 256 Aug 8 14:49 lost+found
drwxrwxr-x. 3 mqm mqm 256 Aug 20 14:16 qmgrs
drwxr-xr-x. 2 nobody nobody 256 Aug 19 18:53 temp
-bash-4.1$ ./amqmfsck -c /mqha
Writing to test file. This will normally complete within about 60 seconds.
AMQ5667: Error executing system call 'fcntl(F_SETLKW)[c]' on file
'/mqha/amqmfsck.wrt', errno 5 ('Input/output error').
-bash-4.1$
-bash-4.1$ ls -la /mqha
total 8
drwxrwxr-x. 7 mqm mqm 4096 Aug 21 2013 .
dr-xr-xr-x. 27 root root 4096 Aug 20 11:25 ..
drwxr-xr-x. 2 nobody nobody 256 Aug 20 14:32 craig
drwxrwxr-x. 3 mqm mqm 256 Aug 20 14:16 logs
drwxr-xr-x. 2 nobody nobody 256 Aug 8 14:49 lost+found
drwxrwxr-x. 3 mqm mqm 256 Aug 20 14:16 qmgrs
drwxr-xr-x. 2 nobody nobody 256 Aug 19 18:53 temp
-bash-4.1$ |
After reading a previous thread, I provided the ID of the mqm's:
| Code: |
-bash-4.1$ id mqm
uid=495(mqm) gid=501(mqm) groups=501(mqm)
-bash-4.1$ id mqm
uid=495(mqm) gid=501(mqm) groups=501(mqm) |
I am having difficutly determining why the mounted drive can not support concurrent writes, since they are both mounted as nfs4.
Any ideas on how I can resolve ?
Thanks |
|
| Back to top |
|
|
| JosephGramig |
| Posted: Wed Aug 21, 2013 10:04 am Post subject: |
|
|
Grand Master
Joined: 09 Feb 2006
Posts: 1244
Location: Gold Coast of Florida, USA
|
Yes, when creating IDs and Groups on UNIX systems, make the UID and GID exactly the same for any one ID on all systems. Same for groups/GIDs.
Do you see where you went wrong here?
[EDIT] I see you do have the UIDs and GIDs the same (I'm blind).
Looks like you need a PMR. |
|
| Back to top |
|
|
|
|
