A week or so ago, I sent the attached query to the list and got
zero response. OK, maybe there's nobody technical here. Or maybe
everybody's busy. If I get no response to this second try, I'll
go find somewhere else to ask.
Basically, we've noticed that continuous disk I/O on an Alpha running
stock RedHat 6.2 locks the system up. No mouse movement, no keyboard
response, no network activity except for answering pings. When the
long-running I/O completes, the system returns to normal.
I need to know if this is normal behavior and if there is anything
I can do to improve the situation. I am a long-time Unix admin and
have been running Linux on Intel hardware since about the 0.95 days,
but am new to Linux on Alpha hardware. Any suggestions or pointers
to which FM to R would be appreciated. I am on the list, so you
can reply to me or to the list as you see fit.
Thanks!
Paul Allen
Here's my previous message:
> I've recently come into possession of a Symbolics-branded XP1000
> box. It's apparently a 500MHz Alpha with 512MB and a nice fast
> SCSI disk. It had some sort of lisp software loaded on top of
> Tru64 Unix. Being unfamiliar with the hardware, it was a bit of a
> struggle, but the machine now runs RedHat 6.2. At this point, it's
> still pretty much in its out-of-the-box condition. In particular,
> I haven't upgraded the 2.2.14 kernel.
>
> One of my co-workers with an identical system was testing files
> >2Gb and noticed that the system appeared to lock up during large
> file copies. I tried "cp /dev/zero /local/junk" on my system and
> noticed the same thing.
>
> The window system was responsive for about 20 seconds after starting
> the copy and then the mouse stopped moving. Telnetting in from
> outside got the "connected" message but no login prompt. Rsh'ing
> in from outside got zero response. It took a few minutes to fill
> the 3.2Gb partition, and then everything was fine. The second time
> I tried this experiment, I nice'd the cp and got the same lockup.
>
> This doesn't seem like correct behavior. Is it taking 20 seconds
> to fill the page cache and then freezing out all other processes?
> Is there a scheduler problem? We're going to be setting up a
> computing cluster using Linux-based dual Alpha nodes. Management
> has observed our experiment and is now having second thoughts about
> the hardware and OS choice. But worst of all, the local NT admin
> is smirking at me. :-(
>
> Suggestions (including "Take it to another list, please!") will be
> appreciated. If you reply, please cc: me. I've sent my subscribe
> request in, but the list is apparently not automated.
>
> Thanks!
>
> Paul Allen
-- Paul L. Allen | voice: (425) 865-3297 fax: (425) 865-2964 Unix Technical Support | paul.l.allen@boeing.com Boeing Phantom Works Math & Computing Technology Site Operations, POB 3707 M/S 7L-68, Seattle, WA 98124-2207-- To unsubscribe: send e-mail to axp-list-request@redhat.com with 'unsubscribe' as the subject. Do not send it to axp-list@redhat.com
This archive was generated by hypermail version 2a22 on Tue Aug 1 04:53:11 2000 PDT
Send any problems or questions about this archive to webmaster@alphalinux.org.