Your queue manager fails to start with the following error :
AMQ7017 Log not available
The qm.ini files had been altered and the LogPath entries were removed.
After making the change to qm.ini LogPath, the queue manager failed to start because S0000001.log was not available. The log file was present and the permissions appeared to be correct, but it seemed to be corrupted.
Edit the qm.ini file and specify the correct LogPath.
If the queue manager fails to start and you suspect a corrupted log file, then do the following:
Here are the steps to follow to recover from the damaged log file(s) in /var/mqm/log/qmgrname/active. The queue manager is referred to as TESTQMGR.
mv TESTQMGR TEST.OLD
NOTE:This will rename the TESTQMGR directory to be TEST.OLD, effectively backing up all your queues and objects.
Note: This command may not completely run and it may produce error messages. This is OK. At the end of this command there are 3 places to check to make sure everything is cleaned up from TESTQMGR and if it did not then you will need to manually remove
what is left.
There should not be a directory under /var/mqm/qmgrs by the name of TESTQMGR as it should have been renamed to TEST.OLD.
In /var/mqm/log there should not be a directory name TESTQMGR. If there is a directory left there, from /var/mqm/log issue the command rm -rf TESTQMGR to remove the old logs.
In /var/mqm there is a file called mqs.ini. Within file mqs.ini there will be QueueManager: stanzas for each queue manager defined on this box. There should no longer be a stanza referring to TESTQMGR. If there is a stanza for TESTQMGR, remove it; the whole
stanza (should be 4 lines.)
cp TESTQMGR/amqalchk.fil TEST.OLD
(This is to copy the new check point file over to the backup directory where you are saving our old queue manager files.)
rm -rf TESTQMGR
(This will remove the new qmgr files you just created so you can replace them with the files from the old queue manager that you had saved.)
mv TEST.OLD TESTQMGR
(This is where you just restored the old definitions. At this point you should have a new queue manager created with new logs but you should have our old queues and channels still there.)
(Start up the queue manager. Everything should be back to normal except there will be no attempt to recover any in-flight UOWs because you have nothing in the logs.)