Re: 6 dead drives in 6 weeks.. after doing oracle exports

"Hub" <hub_at_houston.rr.NOSPAM.com> wrote in message news:<NZNr9.174632$121.4540398_at_twister.austin.rr.com>...
> Ok 2 months ago we converted our databases over to linux servers,
> specifically boxes running Red Hat Advanced Server 7.2. and we're running
> oracle 9i standard edition. My life has been a total nightmare since then.
> We've been getting data block corruptions and all shorts of shit. just about
> one drive a week has been melting down.
>
> Whats funny tho is it seems like lots of disk I/O, for example a full oracle
> export, usually preceeds the drive meltdowns. These are seagate drives in
> JMR enclosuers all RAID1 and RAID0+1.. but the amount of corruption is
> unbelievable and very frustrating.
>
> Anybody have any clues what might be causing this problem? SCSI cables that
> are too long? bad RAID config? anything?
>
> thx... Hub

Strange. I have no cause, nor a solution for this but you might want to test the drives generating lots of io by copying the drives to /dev/null using dd. If the error occurs you can exclude the oracle software from the suspects list.

Ronald.