How to do Cold Restart of Queue Manager ? (or) How to replace the Corrupted Queue Manager logs ?

The following steps are required to cold start MQ on Linux: 

  

Replacing Corrupted logs of a QMGR:

1. Delete all current queue manager logs:                                                                                              

mqm@mqm:~> cd /                                                                                                          

mqm@mqm:~> cd /var/mqm/log/OLDQMGR                                                                   

mqm@mqm:~> rm -r *                                                                                                       

                   

2. Create a DUMMY queue manager:  

(Note:  LOG PROPERTIES MUST MATCH OLDQMGR!!!)

If the corrupted QMGR logs looks like below:

   LogPrimaryFiles=55           -lp

   LogSecondaryFiles=2         -ls

   LogFilePages=16384          -lf

   LogType=LINEAR             -ll

 

Command to duplicate it:

mqm@mqm:~> crtmqm –lp 55 –ls 2 –lf 16384 -ll DUMMY

This might take little time.                                                            

Note: DON'T START the dummy qmgr once it is done.

                                                                                                                                    

3. Move file (amqhlctl.lfh) and directory (active) under:

/var/mqm/log/DUMMY into /var/mqm/log/OLDQMGR                                                                                                                                                                                                                  

mqm@mqm:~> mv /var/mqm/log/DUMMY/* 

        

/var/mqm/log/OLDQMGR                                                   

                                                                                                                                     

4. Restart your QMGR:

Note: make sure all qmgr related processes are down:

ps –ef | grep  <qmgrname >

If you see some process still running Kill it with process id just like below:

kill –9 pid                                                                                                      

mqm@mqm:~> strmqm OLDQMGR

                                                                                                                                     

5. Delete the dummy QMGR:

                                                                                                                                    

       mqm@mqm:~> dltmqm DUMMY

Since the logs been replaced, We lost all information to commit or backout messages but all persistent messages are still in the queues (you don't lose persistent messages).

Thanks,

Guru