Author |
Message
|
MQAltaf |
Posted: Wed Oct 11, 2006 12:51 am Post subject: Unable to start queue manager |
|
|
 Centurion
Joined: 10 Feb 2005 Posts: 119
|
Hi,
I am receiving the following error when I try to start a queue manager on our solaris server
----- amqxfdcx.c : 805 --------------------------------------------------------
10/11/06 09:37:04 - Process(23137.1) User(mqm) Program(amqzxma0_nd)
AMQ6119: An internal WebSphere MQ error has occurred (Failed to get memory
segment: shmget(0x01b80068, 3301376) [rc=-1 errno=28] No space left on device)
------------------------------------------------------------------------------------
The follwing FDC is also being generated: -
+-----------------------------------------------------------------------------+
| |
| WebSphere MQ First Failure Symptom Report |
| ========================================= |
| |
| Date/Time :- Wednesday October 11 09:37:14 BST 2006 |
| Host Name :- hfxrp7 (SunOS 5. |
| PIDS :- 5724H7203 |
| LVLS :- 6.0.1.1 |
| Product Long Name :- WebSphere MQ for Solaris (SPARC platform) |
| Vendor :- IBM |
| Probe Id :- RM185002 |
| Application Name :- MQM |
| Component :- rrmInitRepos |
| SCCS Info :- cmd/remote/amqrrmfa.c, 1.414.1.13 |
| Line Number :- 3165 |
| Build Date :- May 5 2006 |
| CMVC level :- p600-101-060504 |
| Build Type :- IKAP - (Production) |
| UserID :- 00000125 (mqm) |
| Program Name :- amqrrmfa |
| Addressing mode :- 64-bit |
| Process :- 23141 |
| Thread :- 1 |
| QueueManager :- QRFSDV08 |
| ConnId(2) QM :- 9 |
| Major Errorcode :- rrcE_MQCONN_FAILED |
| Minor Errorcode :- OK |
| Probe Type :- MSGAMQ9508 |
| Probe Severity :- 2 |
| Probe Description :- AMQ9508: Program cannot connect to the queue manager. |
| FDCSequenceNumber :- 0 |
| Arith1 :- 2 2 |
| Arith2 :- 2059 80b |
| Comment1 :- QRFSDV08 |
| Comment2 :- MQRC_Q_MGR_NOT_AVAILABLE |
| |
+-----------------------------------------------------------------------------+
The queue manager has been running fine for over a week. Only started failing yesterday.
Kernel setting as follows: -
* vxfs_END
* Oracle specific semaphores
set shmsys:shminfo_shmmax=4294967295
set shmsys:shminfo_shmmin=1
set shmsys:shminfo_shmmni=100
set semsys:seminfo_semmsl=550
set semsys:seminfo_semopm=100
* End Oracle IPC paramters
*For mq 6
set semsys:seminfo_semmni=1024
set semsys:seminfo_semaem=16384
set semsys:seminfo_semvmx=32767
set semsys:seminfo_semmns=16384
set semsys:seminfo_semmnu=16384
set semsys:seminfo_semume=256
set shmsys:shminfo_shmseg=1024
set semsys:seminfo_semmap=1026
set rlim_fd_cur=10000
set rlim_fd_max=10000
Any help would be greatly appreciated
Thanks in advance |
|
Back to top |
|
 |
Vitor |
Posted: Wed Oct 11, 2006 1:01 am Post subject: Re: Unable to start queue manager |
|
|
 Grand High Poobah
Joined: 11 Nov 2005 Posts: 26093 Location: Texas, USA
|
MQAltaf wrote: |
Hi,
I am receiving the following error when I try to start a queue manager on our solaris server
----- amqxfdcx.c : 805 --------------------------------------------------------
10/11/06 09:37:04 - Process(23137.1) User(mqm) Program(amqzxma0_nd)
AMQ6119: An internal WebSphere MQ error has occurred (Failed to get memory
segment: shmget(0x01b80068, 3301376) [rc=-1 errno=28] No space left on device)
|
Sounds like a resource problem on the box. If you Google the ProbeId (RM185002) and/or the error number you'll get a few potentially useful pointers. _________________ Honesty is the best policy.
Insanity is the best defence. |
|
Back to top |
|
 |
MQAltaf |
Posted: Wed Oct 11, 2006 1:11 am Post subject: |
|
|
 Centurion
Joined: 10 Feb 2005 Posts: 119
|
Hi,
Thanks for the swift response. The disk utilisation look ok to me....
$ df -k /var/mqm
Filesystem kbytes used avail capacity Mounted on
/dev/vx/dsk/rp7testdg/mqmvar
102400 27254 70492 28% /var/mqm |
|
Back to top |
|
 |
Vitor |
Posted: Wed Oct 11, 2006 1:18 am Post subject: |
|
|
 Grand High Poobah
Joined: 11 Nov 2005 Posts: 26093 Location: Texas, USA
|
Sorry, should have clarified a bit. When I said "resource problem" I meant the box is running out of memory (hence the inability to get a segment as indicated in the message).  _________________ Honesty is the best policy.
Insanity is the best defence. |
|
Back to top |
|
 |
Gaya3 |
Posted: Wed Oct 11, 2006 2:17 am Post subject: |
|
|
 Jedi
Joined: 12 Sep 2006 Posts: 2493 Location: Boston, US
|
Even i too suspect its resource issue...
Please let us know the contents of .profile for the mqm user.
I suggest you to search the Probe id in IBM website
Thanks and Regards
Gayathri _________________ Regards
Gayathri
-----------------------------------------------
Do Something Before you Die |
|
Back to top |
|
 |
mvic |
Posted: Wed Oct 11, 2006 7:12 am Post subject: |
|
|
 Jedi
Joined: 09 Mar 2004 Posts: 2080
|
Gaya3 wrote: |
Please let us know the contents of .profile for the mqm user. |
This gets run when logging in as mqm. The queue manager and channels don't log in as mqm. So I am confused why you would need to look at ~mqm/.profile |
|
Back to top |
|
 |
Gaya3 |
Posted: Thu Oct 12, 2006 1:27 am Post subject: |
|
|
 Jedi
Joined: 12 Sep 2006 Posts: 2493 Location: Boston, US
|
Hi
Please have a look in to this
http://www-1.ibm.com/support/docview.wss?uid=swg21195939
Place the following line in the .profile for the mqm user:
"ulimit -n 1024
Thanks and Regards
Gayathri _________________ Regards
Gayathri
-----------------------------------------------
Do Something Before you Die |
|
Back to top |
|
 |
MQAltaf |
Posted: Thu Oct 12, 2006 3:23 am Post subject: |
|
|
 Centurion
Joined: 10 Feb 2005 Posts: 119
|
Gayathri,
We are running version 6.01 of MQ on the Solaris server. Also, a second queue manager that has been created is running fine. |
|
Back to top |
|
 |
bbburson |
Posted: Thu Oct 12, 2006 6:08 am Post subject: |
|
|
Partisan
Joined: 06 Jan 2004 Posts: 378 Location: Nowhere near a queue manager
|
MQAltaf wrote: |
Gayathri,
We are running version 6.01 of MQ on the Solaris server. Also, a second queue manager that has been created is running fine. |
If you stop that queue manager, can you then start the one that is giving you problems? Kernel parameter settings may be too low for multiple queue managers; the settings shown in quick beginnings book are minimums and may have to be increased if more than one queue manager is defined on the server. |
|
Back to top |
|
 |
MQAltaf |
Posted: Fri Oct 13, 2006 6:58 am Post subject: |
|
|
 Centurion
Joined: 10 Feb 2005 Posts: 119
|
Here is what IBM suggested and it worked
I'm afraid that the required value for shmmni was omitted from the Quick Beginnings
Guide for WMQ v6. The recommended setting for this parameter is 1024. The
error you are getting is indicating that you need a higher value for this setting (errno=28
equates to "ENOSPC" which is described as followed in the man pages :
"A shared memory identifier is to be created but the system-imposed
limit on the maximum number of allowed shared memory identifiers
system-wide would be exceeded".)
Please see apar IY77743 for details of the doc apar related to this :
http://www-1.ibm.com/support/docview.wss?rs=171&context=SSFKSJ&context=SSWHKB&q1=iy77743&uid=swg1IY77743&loc=en_US&cs=utf-8&lang=en
Please try increasing shmmni and let me know if this resolves the problem nb it may
be necessary to increase this value to beyond 1024 as you have 2 queue managers
on the system, as well as Oracle. |
|
Back to top |
|
 |
|