Don Cunningham donc@bedford.progress.com writes:
This could be related to a similar situation we saw recently with a failed disk. We don't use SNMP but the filer email notification facility. Anyway, a disk failed and was subsequently removed by the OS and didn't generate an email. So essentially the system didn't know that the disk ever existed - similar to what you describe. I placed a call to NetApp and they said this shouldn't ever happen(the OS automatically removing a failed disk) and that it is a bug that will be fixed in a later release. Check your system log for a disk failure message followed immediately by a removal message for the failed disk. If it's there then you've probably got the bug.
Hmm, this seems to happen almost as often as not -- I am fairly aggressive on upgrading, due to obscure problems with SAN Backup, so I suspect that the bug has not yet been corrected. (I was told earlier this year that it was an unpublished bug)
When it comes to the script mentioned earlier, I would suggest that the script somehow stores the value from the prevous run, and then report an error if the number decreases - that is at least what I am planning to do Real Soon Now(tm).
>>>>>>.rune
-
Rune Bakken