mqweb does not stop / restart

[jwende]

I'm just running the latest container from dockerhub. It starts fine but if I stop and restart the mqweb component does not work any longer. Same behavior with and without volumes attached to the container.
9.2.0.0 works fine

Regards
Joerg

2021-03-22T15:33:06.951Z Error 1 starting web server:
Server mqweb is already running.
2021-03-22T15:33:06.951Z Error starting web server: exit status 1

[arthurbarr]

Can you clarify the steps you're taking to see the error, please? I've just tried restart the web server, and it seemed to work:

docker run -d -e LICENSE=accept ibmcom/mq
4b5c1547289d532109ec13e058c2bf2f73edd5a3942ee73451d743e5ef20c278
$ docker exec -ti 4b5 bash
bash-4.4$ ps -ef
UID        PID  PPID  C STIME TTY          TIME CMD
1001         1     0  2 15:59 ?        00:00:00 runmqserver -nologruntime -dev
1001        52     1  0 15:59 ?        00:00:00 /bin/sh /opt/mqm/bin/strmqweb
1001       152    52  0 15:59 ?        00:00:00 /bin/sh /opt/mqm/web/bin/server start mqweb
1001       175   152 99 15:59 ?        00:00:22 /opt/mqm/java/jre64/jre/bin/java -javaagent:/opt/mqm/web/bin/tools/ws-javaagent.jar -Djava.awt.headless=true -Djdk.attach.allowAttachSelf=true -XX:MaxPermSize=256m -Djdk.tls.ephemeralDHKey
1001       176   152  8 15:59 ?        00:00:00 /opt/mqm/java/jre64/jre/bin/java -javaagent:/opt/mqm/web/bin/tools/ws-javaagent.jar -Djava.awt.headless=true -Djdk.attach.allowAttachSelf=true -jar /opt/mqm/web/bin/tools/ws-server.jar mqw
1001       229     1  1 15:59 ?        00:00:00 /opt/mqm/bin/amqzxma0 -m 4b5c1547289d -x -u mqm
1001       245   229  0 15:59 ?        00:00:00 /opt/mqm/bin/amqzfuma -m 4b5c1547289d
1001       250   229  0 15:59 ?        00:00:00 /opt/mqm/bin/amqzmgr0 -m 4b5c1547289d
1001       253   229  1 15:59 ?        00:00:00 /opt/mqm/bin/amqzmuc0 -m 4b5c1547289d
1001       271   229  0 15:59 ?        00:00:00 /opt/mqm/bin/amqzmur0 -m 4b5c1547289d
1001       300   229  0 15:59 ?        00:00:00 /opt/mqm/bin/amqzmuf0 -m 4b5c1547289d
1001       316   229  0 15:59 ?        00:00:00 /opt/mqm/bin/amqrrmfa -m 4b5c1547289d -t2332800 -s2592000 -p2592000 -g5184000 -c3600
1001       339   250  0 15:59 ?        00:00:00 /opt/mqm/bin/runmqchi -m 4b5c1547289d -q SYSTEM.CHANNEL.INITQ -r
1001       343   250  0 15:59 ?        00:00:00 /opt/mqm/bin/amqpcsea 4b5c1547289d
1001       346   250  0 15:59 ?        00:00:00 /opt/mqm/bin/runmqlsr -r -m 4b5c1547289d -t TCP -p 1414
1001       353   229  0 15:59 ?        00:00:00 /opt/mqm/bin/amqzlaa0 -m4b5c1547289d -fip0
1001       358   300  0 15:59 ?        00:00:00 /opt/mqm/bin/amqfqpub -m4b5c1547289d
1001       391   358  0 15:59 ?        00:00:00 /opt/mqm/bin/amqfcxba -m 4b5c1547289d
1001       446     0  1 15:59 pts/0    00:00:00 bash
1001       454   446  1 15:59 pts/0    00:00:00 ps -ef
bash-4.4$ dspmqweb
MQWB1124I: Server 'mqweb' is running.
MQWB1121E: The status of the mqweb server applications cannot be determined.
A request was made to read the status of the deployed mqweb server applications, however the data was not available. This may indicate that there is already an mqweb server started on this system, probably related to another IBM MQ instance.
Check the startup logs for the mqweb server, looking in particular for conflicting usage of network ports. Ensure that if you have multiple mqweb servers on a system, they are configured to use distinct network ports. Restart the mqweb server and ensure it started correctly. If the problem persists, seek assistance from IBM support.
bash-4.4$ endmqweb

Stopping server mqweb.
Server mqweb stopped.
bash-4.4$ strmqweb

Starting server mqweb.
Server mqweb started with process ID 759.
bash-4.4$ dspmqweb
MQWB1124I: Server 'mqweb' is running.
MQWB1123E: The status of the mqweb server applications cannot be determined.
A request was made to read the status of the deployed mqweb server applications, however the data appears corrupt. This may indicate that there is already an mqweb server started on this system, probably related to another IBM MQ instance.
Check the startup logs for the mqweb server, looking in particular for conflicting usage of network ports. Ensure that if you have multiple mqweb servers on a system, they are configured to use distinct network ports. Restart the mqweb server and ensure it started correctly. If the problem persists, seek assistance from IBM support.

[jwende]

docker run --env LICENSE=accept --env MQ_QMGR_NAME=QM1 --publish 1415:1414 --publish 9445:9443 --detach --volume qm1data:/mnt/mqm ibmcom/mq

docker ps
b9a74bfc8be6 ibmcom/mq "runmqdevserver" 51 seconds ago Up 49 seconds 9157/tcp, 0.0.0.0:1415->1414/tcp, 0.0.0.0:9445->9443/tcp unruffled_kare

docker stop b9a74bfc8be6

docker start b9a74bfc8be6

docker logs b9a74bfc8be6
....
2021-03-22T17:50:42.681Z Error 1 starting web server:
Server mqweb is already running.
2021-03-22T17:50:42.681Z Error starting web server: exit status 1

[jwende]

docker exec -ti b9a bash
bash-4.4$ dspmqweb
MQWB1124I: Server 'mqweb' is running.
URLS:
https://b9a74bfc8be6:9443/ibmmq/console/
https://b9a74bfc8be6:9443/ibmmq/rest/
bash-4.4$ endmqweb

Stopping server mqweb.
The command stop failed because of a communication error with the server.
Server mqweb is not running.
bash-4.4$ dspmqweb
MQWB1124I: Server 'mqweb' is running.
URLS:
https://b9a74bfc8be6:9443/ibmmq/console/
https://b9a74bfc8be6:9443/ibmmq/rest/
bash-4.4$

[adepusa]

I am also facing same issue with ibmcom/mq latest container

[matgenois]

Facing the same issue here!

$ docker exec -ti 4a7 bash
bash-4.4$ ps -ef
UID        PID  PPID  C STIME TTY          TIME CMD
1001         1     0  0 20:03 ?        00:00:01 runmqserver -nologruntime -dev
1001       102     1  0 20:03 ?        00:00:00 /opt/mqm/bin/amqzxma0 -m test -x -u mqm
1001       117   102  0 20:03 ?        00:00:00 /opt/mqm/bin/amqzfuma -m test
1001       122   102  0 20:03 ?        00:00:00 /opt/mqm/bin/amqzmgr0 -m test
1001       135   102  0 20:03 ?        00:00:00 /opt/mqm/bin/amqzmuc0 -m test
1001       169   102  0 20:03 ?        00:00:00 /opt/mqm/bin/amqzmur0 -m test
1001       183   102  0 20:03 ?        00:00:00 /opt/mqm/bin/amqzmuf0 -m test
1001       203   102  0 20:03 ?        00:00:00 /opt/mqm/bin/amqrrmfa -m test -t2332800 -s2592000 -p2592000 -g5184000 -c3600
1001       240   183  0 20:04 ?        00:00:00 /opt/mqm/bin/amqfqpub -mtest
1001       246   122  0 20:04 ?        00:00:00 /opt/mqm/bin/runmqchi -m test -q SYSTEM.CHANNEL.INITQ -r
1001       249   122  0 20:04 ?        00:00:00 /opt/mqm/bin/amqpcsea test
1001       250   122  0 20:04 ?        00:00:00 /opt/mqm/bin/runmqlsr -r -m test -t TCP -p 1414
1001       253   102  0 20:04 ?        00:00:00 /opt/mqm/bin/amqzlaa0 -mtest -fip0
1001       302   240  0 20:04 ?        00:00:00 /opt/mqm/bin/amqfcxba -m test
1001       362   250  0 20:08 ?        00:00:00 /opt/mqm/bin/amqrmppa -m test
1001       371     0  0 20:10 pts/0    00:00:00 bash
1001       377   371  0 20:10 pts/0    00:00:00 ps -ef
bash-4.4$ strmqweb

Server mqweb is already running.
bash-4.4$ endmqweb

Stopping server mqweb.
The command stop failed because of a communication error with the server.
Server mqweb is not running.
bash-4.4$

[matgenois]

I forced a refresh of my docker image (it was a 2 months old). It worked on the first execution but not on subsequent one...

[ahorribleidea]

I have the same issue. In a fresh container the web UI works fine, but if the container gets restarted it stops working.

The mqweb logs show some errors related to keystore bindings, I suspect the issue has to do with how mqweb is setting up a cert when it runs for the https connection to the UI, but I have no idea how to fix it.

[hsahota93]

Yup, I'm seeing the same behavior with 9.2.2.0-r1. When I run my docker run command, the UI comes up correctly. After running docker stop <container_name> followed by a docker start <container_name> the web UI fails to load. For now I've switched back to 9.2.1.0-r1 and the restart works as expected. lmk if there is anything I can provide to help debug.

Run command: docker run --env LICENSE=accept --env MQ_QMGR_NAME=QL20 --publish 1414:1414 --publish 9443:9443 -d ibmcom/mq

[fergardi]

For me, I had to downgrade to 9.2.0.0-r1 in order to restart correctly the web console.

[boubou191911]

Same behavior and same error message running 9.2.3.0-r1. Note that I am on docker for Windows but I assume this should have no influence.

[mtttcgcg]

I get the same behavior on latest but also 9.2.1.0-r1. Its log when trying to resume it with docker start :

AMQ8024I: IBM MQ channel initiator started. [CommentInsert1(SYSTEM.CHANNEL.INITQ)]
Error 1 starting web server:
Server mqweb is already running.

Error starting web server: exit status 1
AMQ8942I: Starting to process automatic MQSC configuration script.
AMQ8939I: Automatic MQSC configuration script has completed, and contained 22 command(s), of which 0 had errors. [ArithInsert1(22), CommentInsert1(0)]

Also, I noticed if I run the endmqweb and strmqweb while the container is running, even though the mqweb process is running, it no longer is accessible in the web browser.

[markdjay]

I am seeing the same issues. The web server starts fine when the container starts, but if stopped and restarted independently it fails with errors connecting to the keystore:

Initial state:
ec2-user installations]$ docker exec 45d9db556766 dspmqweb
MQWB1124I: Server 'mqweb' is running.
URLS:
https://45d9db556766:9443/ibmmq/console/
https://45d9db556766:9443/ibmmq/rest/

Stop:
[ec2-user@ installations]$ docker exec 45d9db556766 endmqweb

Stopping server mqweb.
Server mqweb stopped.

Check:
[ec2-user installations]$ docker exec 45d9db556766 dspmqweb
MQWB1125I: Server 'mqweb' is not running.

Restart:
[ec2-user installations]$ docker exec 45d9db556766 strmqweb

Starting server mqweb.
Server mqweb started with process ID 1251.

Check:
[ec2-user installations]$ docker exec 45d9db556766 dspmqweb
MQWB1124I: Server 'mqweb' is running.
MQWB1123E: The status of the mqweb server applications cannot be determined.
A request was made to read the status of the deployed mqweb server applications, however the data appears corrupt. This may indicate that there is already an mqweb server started on this system, probably related to another IBM MQ instance.
Check the startup logs for the mqweb server, looking in particular for conflicting usage of network ports. Ensure that if you have multiple mqweb servers on a system, they are configured to use distinct network ports. Restart the mqweb server and ensure it started correctly. If the problem persists, seek assistance from IBM support.

The web console is unusable at this point.

Checking the webserver logs gives me the following errors:

Console.log:
[WARNING ] CWWKG0033W: The value [${env.AMQ_WEBTRUSTSTOREREF}] specified for the reference attribute [trustStoreRef] was not found in the configuration.
[WARNING ] CWPKI0807W: The keyStore location /run/runmqserver/tls/${env.AMQ_WEBKEYSTORE} could not be found for element with id attribute MQWebKeyStore.
[ERROR ] CWPKI0033E: The keystore located at /run/runmqserver/tls/${env.AMQ_WEBKEYSTORE} did not load because of the following error: KeyStore "/run/runmqserver/tls/${env.AMQ_WEBKEYSTORE}" does not exist.
[WARNING ] CWPKI0809W: There is a failure loading the MQWebKeyStore keystore. If an SSL configuration references the MQWebKeyStore keystore, then the SSL configuration will fail to initialize.
[ERROR ] CWPKI0033E: The keystore located at /run/runmqserver/tls/trust.p12 did not load because of the following error: keystore password was incorrect
[WARNING ] CWPKI0809W: There is a failure loading the MQWebTrustStore keystore. If an SSL configuration references the MQWebTrustStore keystore, then the SSL configuration will fail to initialize.

Messages.log
[8/12/21 8:35:12:453 GMT] 00000023 com.ibm.ws.ssl.config.WSKeyStore I Successfully loaded default keystore: /mnt/mqm/data/web/installations/Installation1/servers/mqweb/resources/security/key.jks of type: JKS
[8/12/21 8:35:12:500 GMT] 00000023 com.ibm.ws.ssl.config.WSKeyStore W CWPKI0807W: The keyStore location /run/runmqserver/tls/${env.AMQ_WEBKEYSTORE} could not be found for element with id attribute MQWebKeyStore.
[8/12/21 8:35:12:500 GMT] 00000023 com.ibm.ws.ssl.config.WSKeyStore E CWPKI0033E: The keystore located at /run/runmqserver/tls/${env.AMQ_WEBKEYSTORE} did not load because of the following error: KeyStore "/run/runmqserver/tls/${env.AMQ_WEBKEYSTORE}" does not exist.
[8/12/21 8:35:12:501 GMT] 00000023 com.ibm.ws.ssl.config.WSKeyStore W CWPKI0809W: There is a failure loading the MQWebKeyStore keystore. If an SSL configuration references the MQWebKeyStore keystore, then the SSL configuration will fail to initialize.

[markdjay]

I am seeing the same issues. The web server starts fine when the container starts, but if stopped and restarted independently it fails with errors connecting to the keystore:

Initial state:
ec2-user installations]$ docker exec 45d9db556766 dspmqweb
MQWB1124I: Server 'mqweb' is running.
URLS:
https://45d9db556766:9443/ibmmq/console/
https://45d9db556766:9443/ibmmq/rest/

Stop:
[ec2-user@ installations]$ docker exec 45d9db556766 endmqweb

Stopping server mqweb.
Server mqweb stopped.

Check:
[ec2-user installations]$ docker exec 45d9db556766 dspmqweb
MQWB1125I: Server 'mqweb' is not running.

Restart:
[ec2-user installations]$ docker exec 45d9db556766 strmqweb

Starting server mqweb.
Server mqweb started with process ID 1251.

Check:
[ec2-user installations]$ docker exec 45d9db556766 dspmqweb
MQWB1124I: Server 'mqweb' is running.
MQWB1123E: The status of the mqweb server applications cannot be determined.
A request was made to read the status of the deployed mqweb server applications, however the data appears corrupt. This may indicate that there is already an mqweb server started on this system, probably related to another IBM MQ instance.
Check the startup logs for the mqweb server, looking in particular for conflicting usage of network ports. Ensure that if you have multiple mqweb servers on a system, they are configured to use distinct network ports. Restart the mqweb server and ensure it started correctly. If the problem persists, seek assistance from IBM support.

The web console is unusable at this point.

Checking the webserver logs gives me the following errors:

Console.log:
[WARNING ] CWWKG0033W: The value [${env.AMQ_WEBTRUSTSTOREREF}] specified for the reference attribute [trustStoreRef] was not found in the configuration.
[WARNING ] CWPKI0807W: The keyStore location /run/runmqserver/tls/${env.AMQ_WEBKEYSTORE} could not be found for element with id attribute MQWebKeyStore.
[ERROR ] CWPKI0033E: The keystore located at /run/runmqserver/tls/${env.AMQ_WEBKEYSTORE} did not load because of the following error: KeyStore "/run/runmqserver/tls/${env.AMQ_WEBKEYSTORE}" does not exist.
[WARNING ] CWPKI0809W: There is a failure loading the MQWebKeyStore keystore. If an SSL configuration references the MQWebKeyStore keystore, then the SSL configuration will fail to initialize.
[ERROR ] CWPKI0033E: The keystore located at /run/runmqserver/tls/trust.p12 did not load because of the following error: keystore password was incorrect
[WARNING ] CWPKI0809W: There is a failure loading the MQWebTrustStore keystore. If an SSL configuration references the MQWebTrustStore keystore, then the SSL configuration will fail to initialize.

Messages.log
[8/12/21 8:35:12:453 GMT] 00000023 com.ibm.ws.ssl.config.WSKeyStore I Successfully loaded default keystore: /mnt/mqm/data/web/installations/Installation1/servers/mqweb/resources/security/key.jks of type: JKS
[8/12/21 8:35:12:500 GMT] 00000023 com.ibm.ws.ssl.config.WSKeyStore W CWPKI0807W: The keyStore location /run/runmqserver/tls/${env.AMQ_WEBKEYSTORE} could not be found for element with id attribute MQWebKeyStore.
[8/12/21 8:35:12:500 GMT] 00000023 com.ibm.ws.ssl.config.WSKeyStore E CWPKI0033E: The keystore located at /run/runmqserver/tls/${env.AMQ_WEBKEYSTORE} did not load because of the following error: KeyStore "/run/runmqserver/tls/${env.AMQ_WEBKEYSTORE}" does not exist.
[8/12/21 8:35:12:501 GMT] 00000023 com.ibm.ws.ssl.config.WSKeyStore W CWPKI0809W: There is a failure loading the MQWebKeyStore keystore. If an SSL configuration references the MQWebKeyStore keystore, then the SSL configuration will fail to initialize.

UPDATE: Reported this directly to IBM and they have recreated the issue. Working with the platform team to resolve.

[formalist]

+1

ver 9.2.3.0-r1

[StefanoVuerich]

Anyone of IBM messaging team has some kind of update about this issue?

[markdjay]

The update from the IBM platform team to the ticket I raised above is that this is working as designed now. Control of the web server has been moved to the container level.

[Amplus]

Hello I'm using latest image: 9.2.4.0-r1 and I'm having the same problem. If I get latest and run it, it works the first time but if I stop the container and start again I get the above error.

I'm following this tutorial and using the commands provided there:
https://developer.ibm.com/tutorials/mq-connect-app-queue-manager-containers/

Workaround is to not use volumes and add the --rm flag to remove the container when it stops:

docker run --rm --env LICENSE=accept --env MQ_QMGR_NAME=QM1 --publish 1414:1414 --publish 9443:9443 --detach --env MQ_APP_PASSWORD=passw0rd ibmcom/mq:latest

[estreetco]

I'd also like to mention that I experienced the very same problem using the latest (9.2.4.0-r1) as well as 9.2.1.0-r1.

As per fergardi, mqweb is working fine (post restart) after downgrading to 9.2.0.0-r1.

Hopefully IBM addresses this issue soon.

Thanks all for the inputs, this thread was very helpful!

[shin-san]

+1. I have tried icr.io/ibm-messaging/mq:9.2.5.0-r1 and the problem still persists if you run the endmqweb and follow it with strmqweb command:

MQWB1124I: Server 'mqweb' is running.
URLS:
  https://853db90b8f32:9443/ibmmq/console/
  https://853db90b8f32:9443/ibmmq/rest/
bash-4.4$ endmqweb

Stopping server mqweb.


Server mqweb stopped.
bash-4.4$ 
bash-4.4$
bash-4.4$ strmqweb

Starting server mqweb.
Server mqweb started with process ID 1085.
bash-4.4$ dspmqweb
MQWB1124I: Server 'mqweb' is running.
MQWB1123E: The status of the mqweb server applications cannot be determined.
A request was made to read the status of the deployed mqweb server applications, however the data appears corrupt. This may indicate that there is already an mqweb server started on this system, probably related to another IBM MQ instance.
Check the startup logs for the mqweb server, looking in particular for conflicting usage of network ports. Ensure that if you have multiple mqweb servers on a system, they are configured to use distinct network ports. Restart the mqweb server and ensure it started correctly. If the problem persists, seek assistance from IBM support.

Another thing, it seems like IBM have deprecated the containers in DockerHub and the new versions of ibm-mq are available in IBM Container Registry.

[Blem]

Still an issue in v9.3.0.0

[MarkJamesHoward]

Still issue for me here with v9.3.0.0 too

[KTJ-Demo]

Me too. Spun up a docker container with the following: docker run --env LICENSE=accept --env MQ_QMGR_NAME=QM1 --publish 1414:1414 --publish 9443:9443 --detach --volume qm1data:/mnt/mqm icr.io/ibm-messaging/mq:latest

IBM MQ queue manager 'QM1' started using V9.3.0.0

[rubenramosDev]

Any updates on this?

[rubenramosDev]

Any updates on this?

I was facing the same problem, I was following this tutorial..

I solved the problem deleting the docker volumen where Ibm stores data with the next command:

docker volume rm yourIBMVolume

I created the volumen once again and the container too and problem was fixed !!

[Ronmck]

I have the same issue with 9.3.1.0 from docker pull icr.io/ibm-messaging/mq:latest
Only works the first time, restarts do not work
console.log
messages.log
docker-container.log

[SamehSaeed]

Any update on this? The same issue with the latest version!!

[arthurbarr]

There seem to be two sets of people doing different things on this issue:

1. Stopping and restarting just the web component within a running container. The code was designed for a container to be an atomic entity, so you restart the whole container, and not just an individual component.
2. Stopping and restarting the whole container, and the web component still not working.

I think the original issue was for option (1), which would be for a new feature of the code, which we're not working on, and have no plans for. Therefore I think this issue should be closed.

However, if someone can confirm that option (2) is happening, I think that might be worth a new issue. I cannot reproduce that behaviour with Podman, so the answer may be Docker specific, but we can handle that if someone raising a new issue.

[IniasP]

There seem to be two sets of people doing different things on this issue:

1. Stopping and restarting just the web component within a running container. The code was designed for a container to be an atomic entity, so you restart the whole container, and not just an individual component.
2. Stopping and restarting the whole container, and the web component still not working.

I think the original issue was for option (1), which would be for a new feature of the code, which we're not working on, and have no plans for. Therefore I think this issue should be closed.

However, if someone can confirm that option (2) is happening, I think that might be worth a new issue. I cannot reproduce that behaviour with Podman, so the answer may be Docker specific, but we can handle that if someone raising a new issue.

This issue is about restarting the entire container. Last time I tried it, and as others mentioned above, only deleting the volume solved the issue.

[arthurbarr]

Multiple comments in this issue are specific to just restarting the web component within a running container. The issue has become confused, so I have closed it. If someone wants to raise a new issue for the full container restart, providing full details of how to reproduce, including their environment details, we can try reproduce it and fix the problem. As it is, I cannot reproduce it.

[Ronmck]

I was able to reproduce the problem, the web console worked the first time when the container was started, I restarted the container and web console stopped working.
Error starting web server: exit status 1

I am running Docker Desktop 4.17.0 for windows with WSL2 enabled in docker.
See attached docker log:
docker-container-log.txt

[arthurbarr]

I previously closed this as I thought it was confusing, and hoped to clear up that confusion with new issues. Seeing the negative reactions, I realise this was the wrong approach. My apologies. Re-opening.

[chamila-lfs]

I also faced same issue and could not find a solution from IBM for this. After some digging based on previous comments on deleting the volume, I figured out the web configuration is in the folder /mnt/mqm/data/web.
I was able to start up the web service just by deleting only the folder /mnt/mqm/data/web rather than deleting the whole volume and loosing the config/data/messages inside it. The container startup process creates this folder again.

To make life easy I have added a script to delete this folder at the start up. Every time you restart the container, the web folder is deleted and recreated.

This is not the most elegant solution but at least it works for me until IBM comes up with a solution.

Dockerfile

FROM icr.io/ibm-messaging/mq:9.3.1.1-r1

USER root
COPY init_script.sh /usr/local/bin/init_script.sh
RUN sed -i 's/\r//' /usr/local/bin/init_script.sh
&& chmod +x /usr/local/bin/init_script.sh
USER 1001
ENTRYPOINT ["/usr/local/bin/init_script.sh"]

init_script.sh

#!/bin/bash
rm -rf /mnt/mqm/data/web
runmqserver

---

Hope this helps.

[monnetchr]

Alternate workaround based on the solution by @chamila-lfs : before stopping the container, execute endmqweb and rm -rf /mnt/mqm/data/web

[Nicholas2014]

Any updates? I faced the same issue with the latest image

[elawady55]

the same issue here:
2023-05-09 12:56:15 2023-05-09T10:56:15.731Z Created directory structure under /var/mqm
2023-05-09 12:56:15 2023-05-09T10:56:15.731Z Image created: 2023-04-11T13:03:48+00:00
2023-05-09 12:56:15 2023-05-09T10:56:15.731Z Image tag: ibm-mqadvanced-server-dev:9.3.2.1-r1.20230411130007.96135e8-amd64
2023-05-09 12:56:15 2023-05-09T10:56:15.882Z MQ version: 9.3.2.1
2023-05-09 12:56:15 2023-05-09T10:56:15.882Z MQ level: p932-001-230405
2023-05-09 12:56:15 2023-05-09T10:56:15.882Z MQ license: Developer
2023-05-09 12:56:51 2023-05-09T10:56:51.613Z FIPS cryptography is not enabled.
2023-05-09 12:56:51 2023-05-09T10:56:51.613Z Creating queue manager QM1
2023-05-09 12:56:51 2023-05-09T10:56:51.613Z Starting web server
2023-05-09 12:56:51 2023-05-09T10:56:51.715Z Detected existing queue manager QM1
2023-05-09 12:56:51 2023-05-09T10:56:51.807Z Removing existing ServiceComponent configuration
2023-05-09 12:56:51 2023-05-09T10:56:51.812Z Starting queue manager
2023-05-09 12:56:52 2023-05-09T10:56:52.317Z Error 1 starting web server:
2023-05-09 12:56:52 Server mqweb is already running.
2023-05-09 12:56:52
2023-05-09 12:56:52 2023-05-09T10:56:52.317Z Error starting web server: exit status 1
2023-05-09 12:56:55 2023-05-09T10:56:55.279Z Started queue manager

[peters-axon]

Restart is still not working for "latest".
9.2.0.0 seems to work fine.

[vgavinash]

Hi all, can anyone here try the same using podman for linux or docker for linux and confirm if this issue is encountered or not?

[afahmy11]

i am having the exact same issue with my setup "AWS fargate + EFS Storage" . if the container got restarted for any reason, the mqweb reports that it is already running.
the solution for us is to shutdown the container, wait for about 5 min then start again
i think it has to do with a file being locked in the storage and prevents the web service in the container from starting again as it looks like another service is using the file

[KoakiMiku]

+1, same issue here, and delete /mnt/mqm/data/web can fix it.

log

2024-02-06 14:23:05 2024-02-06T06:23:05.893Z Error 1 starting web server:
2024-02-06 14:23:05 Server mqweb is already running.
2024-02-06 14:23:05
2024-02-06 14:23:05 2024-02-06T06:23:05.893Z Error starting web server: exit status 1

image

icr.io/ibm-messaging/mq:latest

dspmqver

Name: IBM MQ
Version: 9.3.4.1
Level: p934-001-231123-cp4i
BuildType: IKAP - (Production)
Platform: IBM MQ for Linux (x86-64 platform)
Mode: 64-bit
O/S: Linux 5.15.133.1-microsoft-standard-WSL2
O/S Details: Red Hat Enterprise Linux 8.9 (Ootpa)
InstName: Installation1
InstDesc: IBM MQ V9.3.4.1 (Unzipped)
Primary: N/A
InstPath: /opt/mqm
DataPath: /mnt/mqm/data
MaxCmdLevel: 934
LicenseType: Developer
ReleaseType: Continuous Delivery (CD)

[jcreforme]

https://community.ibm.com/community/user/integration/discussion/mq-in-a-container-err-connection-closed-problem

It works for me:

cp /opt/mqm/web/mq/samp/configuration/basic_registry.xml /var/mqm/web/installations/Installation1/servers/mqweb/mqwebuser.xml

setmqweb properties -k httpHost -v '*'
setmqweb properties -k mqRestMftCommandQmgr -v MQ1.QM

setmqweb properties -k mqRestMftEnabled -v true

[espaciomore]

https://community.ibm.com/community/user/integration/discussion/mq-in-a-container-err-connection-closed-problem

It works for me:

cp /opt/mqm/web/mq/samp/configuration/basic_registry.xml /var/mqm/web/installations/Installation1/servers/mqweb/mqwebuser.xml

setmqweb properties -k httpHost -v '*' setmqweb properties -k mqRestMftCommandQmgr -v MQ1.QM

setmqweb properties -k mqRestMftEnabled -v true

Yes, this worked for me too.

[vgavinash]

Hi all, please note that latest cd release IBM MQ v9.3.5 has the fix for the issue wrt WSL & liberty web server identified in the past in this long chain of posts . Please have a go with the latest version and let us know. Thanks!

Cc: @arthurbarr @RamSubbarao

[vgavinash]

Hi all, please note that latest cd release IBM MQ v9.3.5 has the fix for the issue wrt WSL & liberty web server identified in the past in this long chain of posts . Please have a go with the latest version and let us know. Thanks!

Cc: @arthurbarr @RamSubbarao

@jwende @espaciomore @jcreforme @KoakiMiku @afahmy11 @peters-axon @elawady55 @rubenramosDev ^^^

[estreetco]

Hi all, please note that latest cd release IBM MQ v9.3.5 has the fix for the issue wrt WSL & liberty web server identified in the past in this long chain of posts . Please have a go with the latest version and let us know. Thanks!
Cc: @arthurbarr @RamSubbarao

@jwende @espaciomore @jcreforme @KoakiMiku @afahmy11 @peters-axon @elawady55 @rubenramosDev ^^^

Thanks, I downloaded this latest MQ 9.3.5 image, and the issue with the web console appears to be fixed in that it successfully launches each time the container is executed. However, has something changed with the login id/password mechanism? I used the --env MQ_APP_PASSWORD=passw0rd option in my docker run, but I can't log in (userid admin). I tried a bunch of basic credential combinations and none of them are working.

I'm also following the mqweb setup instructions here:

https://developer.ibm.com/tutorials/mq-setting-up-using-ibm-mq-console/

[arthurbarr]

The MQ_APP_PASSWORD environment variable controls the password for the app user. The MQ_ADMIN_PASSWORD environment variable is the one you want for the admin user.

There was a related change in MQ 9.3.4, where you now have to set a password if you want to have either user: there's no default password.

[estreetco]

The MQ_APP_PASSWORD environment variable controls the password for the app user. The MQ_ADMIN_PASSWORD environment variable is the one you want for the admin user.

There was a related change in MQ 9.3.4, where you now have to set a password if you want to have either user: there's no default password.

Thanks a lot, this resolved my issue. One small note for the IBM folks who compiled this latest 9.3.5 image -- vi doesn't appear to be available (it is in 9.2.0.0). So I'm not sure how to edit the mqwebuser.xml file.

bash: /usr/bin/vi: No such file or directory

[arthurbarr]

Glad that solved the problem.

IBM MQ uses the ubi-minimal base image, and I think Red Hat have now removed vi from that image. For security reasons, there is always a desire to have as little non-essential software in the container as possible. The intention is that you should edit the file outside the container, and mount or copy it in.

[vgavinash]

@jwende as the issue is now resolved in the latest version of mq-container, can this git issue be closed?

[dariodsa]

No, in version 9.4.0.0-r3 we are still having this issue. So basically, if OOM terminates the web console process, strmqweb is unable to normally start the web console.

[dariodsa]

The reason why is quite simply, runmqserver is creating a wildcard.p12 and it is generating a keystore with some password which is saved in the environment variable of runmqserver. If I start strmqweb from the bash, mqweb can't "reach" the env variables set in the runmqserver (pid 1).
Normally mqweb is being started by the runmqserver (pid 1) and in its ENV variables has password of the keystore, otherwise it will not bind to port and it will print this error.

There is a failure loading the MQWebKeyStore keystore. If an SSL configuration references the MQWebKeyStore keystore, then the SSL configuration will fail to initialize.

If I set those variables in the terminal and then start strmqweb, MQ web console starts normally because strmqweb has those variables from the bash process.
Example:

kill -9 $(pgrep java)
strmqweb # won't start MQ console

Example:

kill -9 $(pgrep java)
export AMQ_ADDITIONAL_JSON_LOG=1
export AMQ_WEBTRUSTSTOREREF=MQWebTrustStore
export AMQ_WEBKEYSTOREPW=************
export AMQ_DIAGNOSTIC_MSG_SEVERITY=1
export AMQ_WEBKEYSTORE=wildcard.p12
strmqweb # will start MQ console

[vgavinash]

@dariodsa the initial issue reported in this git issue is something like this - the user first starts mq-container. Web server will be running fine. Later stops the container and restarts it, the container console logs throw the error, Error starting web server: exit status 1 . This is already fixed as per our previous update.

Your scenario seems a little different unlike the one mentioned above. Suggest you to raise a new git issue for the same, so that it can be discussed in a new thread.

Cc: @arthurbarr

[dariodsa]

Ok, I opened a new issue #583.