"Andrew" == Andrew Hobson ahobson@eng.mindspring.net writes:
Andrew> On Mon, 17 May 1999 12:50:58 -0700, "Jim McCoy" Andrew> mccoy@yahoo-inc.com said: >> sirbruce@ix.netcom.com writes: >>> As to what happened, the most *likely* scenario for any >>> signficant downtime of the Netapp is double disk failure. >>> That is, one disk failed, and during reconstruction they lost >>> another.
>> I beg to differ.
Andrew> I just about to write this exact same email. We've had Andrew> the same experiences with filer failers and quality Andrew> control.
Andrew> We're having the same thoughts about the cost when you Andrew> want non-stop computing. Netapp is great for the 98% Andrew> uptime market. They aren't ready to play in the 99.99% Andrew> uptime market, at least in my view.
[Let's try this again. Hit send too soon.]
Maybe we're just not a big enough NetApp customer. We have an F540, an F630, and an F740. None of them have ever given us the least bit of trouble. We're about to upgrade to a pair of clutered F740s.
The F630 we've been pushing to the max for months. We were running over 10K NFSops every day for over a week. Typically it is running at 6-8K NFSops. This is in a web-server application, so we're mostly doing reads.
The only trouble we had was with a GNIC and that was a known bug in 4.x and it didn't show up till after the box had been up for almost a year. We upgraded to 5.2.2 and the problem went away.
We've had no hardware failures that have brought the boxes down. Never had to do a wack. The F540 stayed up solid for almost two years.
We haven't had any trouble with the F740 since we first powered it up 120 days ago.
j. -- Jay Soffian jay@cimedia.com UNIX Systems Administrator 404.572.1941 Cox Interactive Media