From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ish Rattan To: <9fans@cse.psu.edu> Subject: Re: [9fans] Errors.. In-Reply-To: <0a5d2c527ddc8e5cfe3ab79ada8872b4@plan9.bell-labs.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Date: Tue, 21 May 2002 07:09:09 -0400 Topicbox-Message-UUID: 98ac00a6-eaca-11e9-9e20-41e7f4b1d025 I replaced the disk and in single cpu operation the io/error has not been reported yet. I have activated the other cpu too! BTW, I tried the system on a dual Celeron ABIT-BP6 motherboard. As soon as second cpu is activated, the following error is printed repeatedly: cpu1: apic error 0x0000008 Looks like the mp kernel has problems? -ishwar On Mon, 20 May 2002 jmk@plan9.bell-labs.com wrote: > On Sun May 19 09:35:20 EDT 2002, ishwar@pali.cps.cmich.edu wrote: > > > > I am runing a standalone auth/cpu server on dual PIII box. > > > > The console is full of errors of type > > > > ..... > > command 30 > > data 80020cb0 limit 80020eb0 dlen 8192 status 0 error 0 > > lba 1253968 -> 123968, count 16 => 16 (16) > > wrenwrite failed: i/o error > > > > ..... > > > > The system is usable though. I understand that error pertains to > > hard disk, could it be sign of a bad disk (I do not have a replacement > > disk at present)? > > > > -ishwar > > In the above I'm assuming you mistyped or the console mangled > lba 1253968 -> 123968, count 16 => 16 (16) > The 2nd lba number should be the same as the first if the count (16) > is the same. > > Anyway, it's complaining that a write of 16 blocks at lba 1253968 > didn't complete after 30 seconds so the driver aborted it. The command > is retried by the driver with DMA, RWM turned off, and only returns > the error if that 2nd attempt fails. > > Do you have the same problems if you run with only one processor > (*nomp=1 in plan9.ini)? >