Subject: re: Ultra 10s premature death?
To: The bird <jonl@yubyub.net>
From: matthew green <mrg@eterna.com.au>
List: port-sparc
Date: 04/03/2002 11:47:40
> many people told that this means level-2 cache (2MB for utlra
> 10 440MHz CPU) is damaged.
>
> Is this problem of the ecache common for ultrasparc CPUs ??
> anyone experienced this problem ever?
About 2 years ago, we got a shipment of Ultra 10s which exhibited this
problem (or a very similar one). We had to get all 16 replaced. Sun
confirmed it at that time to be a CPU bug.
The newer suns we've been using (all Netra class) have been fine, at least
in that department. I haven't seen the problem since then.
i've had a u10/440 since feb 2000 and it's basically been working
flawlessly for me running solaris 2.6 and while ago 8.
"lucky" i guess.