| Author |
Message
|
| ucbus1 |
 Posted: Thu Jan 22, 2004 6:57 am Post subject: Logon not possible: No connection to MQSeries Workflow serve |
 |
|
Knight
Joined: 30 Jan 2002
Posts: 560
|
I am getting the following errors
"Logon not possible: No connection to MQSeries Workflow server. "
I have checked the channels and every thing seems to be working on NT as well as AIX. We have a queue mgr on NT and another queue manager (FMCQM) on AIX which are in a cluster. I find the "receiver channel" on NT is in inactive tste. When I tried to start it it says the channel is Active state. Is it causing any problems?
Thanks |
|
| Back to top |
|
 |
| vennela |
| Posted: Thu Jan 22, 2004 8:21 am Post subject: |
|
|
Jedi Knight
Joined: 11 Aug 2002
Posts: 4055
Location: Hyderabad, India
|
From the box you are trying to logon do
fmczccd
You will be able to figure out the problem.
On both the boxes, a CLUSSDR and CLUSRCVR channels should be running. You cannot start a CLUSRCVR channel. From the other side, start the CLUSSDR and your CLUSRCVR should start automatically.
How are you trying to logon?
If from a web page, try logging in through fmcautil and see if it works. |
|
| Back to top |
|
|
| ucbus1 |
| Posted: Thu Jan 22, 2004 10:12 am Post subject: |
|
|
Knight
Joined: 30 Jan 2002
Posts: 560
|
Thanks Vennela,
On NT (client) I gave the following
fmczccd -y FMC1
Cannot find the channel table in the configuration profile.
No channel table found.
You are using the MQSeries Server API for the queue manager communication.
On AIX(server) I have tried to start CLUSSDR( SYSTEM.DEF.CLUSSDR) channel ists saying "retrying"
On NT same case.
fmcautil did not run on my client NT box. On AIX I could connect.
One more intersing aspect. I have different configuration on the same machines with "FMC". When I tried restarting channels clusssdr, I am seeing the same staus "retrying". But I do not have any problems with FMC either |
|
| Back to top |
|
|
| vennela |
| Posted: Thu Jan 22, 2004 1:26 pm Post subject: |
|
|
Jedi Knight
Joined: 11 Aug 2002
Posts: 4055
Location: Hyderabad, India
|
| Quote: |
| I have tried to start CLUSSDR( SYSTEM.DEF.CLUSSDR) |
That's not the one.
Usually if you have used config utility then the cluster channels would be named after their qmanagers in this way
TO.FMCQM.TCP
TO.FMCQM1.TCP
But these channels should start automatically. If your Win NT QMGR name is FMCQM1 then try starting TO.FMCQM1.TCP channel on the AIX box.
Also do
on both the boxes to see if the qmgrs are really in a cluster |
|
| Back to top |
|
|
| ucbus1 |
| Posted: Thu Jan 22, 2004 3:00 pm Post subject: |
|
|
Knight
Joined: 30 Jan 2002
Posts: 560
|
Thanks.
i checked dis clusqmgr(*) on both the boxes.
started the channels as suggested by you.
Still I am getting the "logon not possible.. Not sure what I am missing here |
|
| Back to top |
|
|
| ucbus1 |
| Posted: Thu Jan 22, 2004 3:04 pm Post subject: |
|
|
Knight
Joined: 30 Jan 2002
Posts: 560
|
I am seeing following error. on servlet.log
it says
doGet (logon): com.ibm.workflow.servlet.client.ClientException: FMC00013E Communication error - Failing Action: open, Reason Code: 2085, Failing Object: FMC1.FMCGRP1.EXE.": com.ibm.workflow.api.FmcException:
FMC38009E MQSeries Workflow API Error :
API Return Code : 13
Error Origin : e:\v330\src\fmcjcesv.cxx, line 2120
Error Message : FMC00013E Communication error - Failing Action: open, Reason Code: 2085, Failing Object: FMC1.FMCGRP1.EXE.
Nested Exception : NoneDoes that make any sense???
Please let me know |
|
| Back to top |
|
|
| vennela |
| Posted: Thu Jan 22, 2004 3:33 pm Post subject: |
|
|
Jedi Knight
Joined: 11 Aug 2002
Posts: 4055
Location: Hyderabad, India
|
I have seen this error before so many times. This usually happens because the Web client QMGR and Workflow server QMGR are not added to the cluster in the way they need to be added.
Since your admin utility is also not working, I am pretty sure your web client QMGR is not talking to Workflow server QMGR. If it is, then it is only a one way communication. To tell you what the real problem is some more information is needed.
What is the output of
dis clusqmgr(*) on both the boxes
dis chs(*) on both the boxes (I am interested only in CLUSSDR and CLUSRCVR channels). |
|
| Back to top |
|
|
| ucbus1 |
| Posted: Thu Jan 22, 2004 3:49 pm Post subject: |
|
|
Knight
Joined: 30 Jan 2002
Posts: 560
|
on AIX
| Quote: |
5 : dis clusqmgr(*)
AMQ8441: Display Cluster Queue Manager details.
CLUSQMGR(FMCONQM1) CLUSTER(FMCGRP_DEV)
CHANNEL(TO.FMCONQM1.TCP)
AMQ8441: Display Cluster Queue Manager details.
CLUSQMGR(FMCQM1) CLUSTER(FMCGRP_DEV)
CHANNEL(TO.FMCQM1.TCP)
|
| Quote: |
AMQ8417: Display Channel Status details.
CHANNEL(SYSTEM.DEF.CLUSSDR) XMITQ(SYSTEM.CLUSTER.TRANSMIT.QUEUE)
CONNAME( ) CURRENT
CHLTYPE(CLUSSDR) STATUS(RETRYING)
AMQ8417: Display Channel Status details.
CHANNEL(TO.FMCONQM1.TCP) XMITQ(SYSTEM.CLUSTER.TRANSMIT.QUEUE)
CONNAME(NTnode(14010)) CURRENT
CHLTYPE(CLUSSDR) STATUS(RUNNING)
AMQ8417: Display Channel Status details.
CHANNEL(TO.FMCQM1.TCP) XMITQ( )
CONNAME(xx.xx.x.xx) CURRENT
CHLTYPE(CLUSRCVR) STATUS(RUNNING) |
NTNode: actual node name
xxx.xx.x.xx NT node IP address
on NT
| Quote: |
dis clusqmgr(*)
3 : dis clusqmgr(*)
AMQ8441: Display Cluster Queue Manager details.
CLUSQMGR(FMCONQM1) CLUSTER(FMCGRP_DEV)
CHANNEL(TO.FMCONQM1.TCP)
AMQ8441: Display Cluster Queue Manager details.
CLUSQMGR(FMCQM1) CLUSTER(FMCGRP_DEV)
CHANNEL(TO.FMCQM1.TCP)
dis chs(*)
AMQ8417: Display Channel Status details.
CHANNEL(SYSTEM.DEF.CLUSSDR) XMITQ(SYSTEM.CLUSTER.TRANSMIT.QUEUE)
CONNAME( ) CURRENT
CHLTYPE(CLUSSDR) STATUS(RETRYING)
AMQ8417: Display Channel Status details.
CHANNEL(TO.FMCONQM1.TCP) XMITQ( )
CONNAME(AIX IP) CURRENT
CHLTYPE(CLUSRCVR) STATUS(RUNNING)
AMQ8417: Display Channel Status details.
CHANNEL(TO.FMCQM1.TCP) XMITQ(SYSTEM.CLUSTER.TRANSMIT.QUEUE)
CONNAME(AIXnode(14010)) CURRENT
CHLTYPE(CLUSSDR) STATUS(RUNNING)
|
Thanks |
|
| Back to top |
|
|
| ucbus1 |
| Posted: Fri Jan 23, 2004 9:23 am Post subject: solved |
|
|
Knight
Joined: 30 Jan 2002
Posts: 560
|
I have rebooted the machien (NT) but it did not help. As final measure I did "refresh cluster " on NT and it did the tick.
But I am surprised when I rebooted it should have refreshed the clusters right?? But not in this case?
As per manual refreshing is a last resort. Any ideas what could have gone wrong with my set up. i confirmed that nobody fiddled with that and the set up was not used for a while like 10 days
Thanks |
|
| Back to top |
|
|
|
|
