{process_name}_xlate wouldn’t clear

[Andrew Deters]

I had a strange situation.  process crashed with what looked like a 2Gig log file.  Couldn’t view the log file because it was too large.  Tried clearing the errors and restarting but it wouldn’t come up.  Error was dbuserid in use.  So i tried clearing that error by issueing the hciclmclear -p {process_name} -v which showed it cleared all dbuserid’s associated with the process.  Didn’t work same dbuserid error.  Tried clearing just that userid with lmclear and the dbuserid but still wouldnt’ clear.   What is the desired fix for clearning a dbuserid for a process that won’t clear?

Topic

[Max Drown (Infor)]

The command “hcidbdump -ACf” will reinitialize the databases for that site including the recovery and error databases. All data in the database will be lost and you will be left with a brand new, clean database. This is a last resort if you can’t recovery from the database errors. If you have to go this route to solve your problem, then you can resend data from the SMAT files as needed.

-- Max Drown (Infor)

[Andrew Deters]

I was hoping for a less destructive answer.  That is what i had to resort to but determining what messages where in the recovery database is an issue i haven’t had to tackle.  What would the easiest way be to resend messages that were in the recovery databse at the time of initialization.

[Henry Bauer]

Is that the only log file that large? Are there other log files of that size or larger that might be overwhelming the system. Do you have Log file cycling set up, max number of files and max size of files? Is it just this process crashing? Can you deleete the log file or save it off to your pc and analyze it. Just some thoughts.

[Max Drown (Infor)]

There isn’t an easy way that I know of.   😕

-- Max Drown (Infor)

[Author]

Replies