Views |
||||||||||||||
Is there a way to check the consistency of the database?
You might want to pipe the output to a file as it can be quite large and quickly scrolls off the page. It also can take quite a while to run. For example, it took a 2 processor Sun450 with a 14GB database about 45 minutes to execute.
"Bad image header" "does not exist"
/usr/openv/netbackup/bin/admincmd/bpimage -cleanup -allclients This will produce output in that day's bpdbm log file. The output is much less verbose and clearly doesn't check as many things, but it does pickup the "Bad image header" errors. This is simply kicking off manually the same cleanup that bpsched initiates after being idle for (by default) 10 minutes (and which will force all later enqueued backup jobs to wait until it has completed absent a CLEAN_IN_BACKGROUND file in place). This is discussed in Veritas's technote 236274, and may no longer be valid under NetBackup 6.x (I don't have a test system handy). Also, note that one should take the "Bad image" assertion with a grain of salt and check the seconds-since-epoch date in the name. If a backup is running but hasn't finished creating a TOC for the archive it's about to make, then the image header file will be zero size and the file listing (same name, .f extension) won't exist yet. --Grammar 09:55, 15 October 2007 (PDT) |
||||||||||||||
| This page was last modified 17:07, 15 October 2007. | ||||||||||||||