MaxDB AutloLogs failing

[Former Member]

Hi

We have about 30 MaxDB instances, mostly v7.8.02.30, but also v7.9.08.06 running on SLES 11 SP2.

DB-Backups are written to an NFS share with automount.

A few weeks ago I realized that some of the databases had full Log Areas, although AutoLog Backup is generally enabled. Due to lack of time and unimportance of the related systems we solved the error by simply re-activating AutoLog which would then run fine for the next 4-7 days.

Now this starts to annoy since it occurs about twice a week and always affects 3 of our databases (both v7.8 and v7.9)

After digging around in the KnlMsg files I can see the following (more or less equal) lines in the affected DBs:

	AUTOSAVE   52024:  49000 pages -> "kup/SID/SID_AUTOLOG.271"
	AUTOSAVE   52024:  55800 pages -> "kup/SID/SID_AUTOLOG.271"
	AUTOSAVE   52002:  close error 1
	SAPDBErr   20003:  Any error causes autolog backup to terminate.,_FILE=vkb38+noPIC.cpp,_LINE=5756
	Backup	1:  Backupmedium #1 (/mnt/dbbackup/SID/SID_AUTOLOG.271)
	KernelComm	6:  Backup error occured, Error code 5750 "write_task_crash"

Now I'd like to know how to track down the error cause. Can you guys help me?

Michael

[yashwanthkumar_kethineed]

Hello Michael,

It seems that you are running out of disk space on your log volume.

When the Log Area is FULL, the backup activity is cancelled, and the AUTOSAVE LOG is deactivated. I asssume this is what is happening in your case.

You must always ensure that you explicitly reactivate the AUTOSAVE LOG when you have space in the file system again. If you do not activate AUTOSAVE LOG, the log area may become full again.

If this is not the case and you have enough space on your log volume and still face problems with backups, it would be helpful to check the KnlMsg file for further analysis. Please attach the KnlMsg to the thread.

Refer to SAP Note: 1377148 - FAQ: SAP MaxDB backup/recovery

Regards,

Yashwanth