headers
Torsten Harenberg | 1 Sep 2006 09:06
Picon
Favicon

SCSI debugging

Dear alpha fellows,

my good ol' PWS 433au (running as our main web and backup server) started to 
fail occasionally about half a year ago.

Console says:

isp0: command timed out for 0.1.0
... wpdu2 syslogd: /var/log/messages: imput/output error
sd0:(isp0:0:0:0): adapter resource shortage

So I assume it's a hardware problem.

Does anybody has a good idea how to debug if it's the hard drive or the 
controller (or both)?

Any hint is appreciated,

   Torsten

--

-- 
<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>
<>                                                              <>
<> Torsten Harenberg         harenberg <at> physik.uni-wuppertal.de  <>
<> Bergische Universitaet                                       <>
<> FB C - Physik             Tel.: +49 (0)202 439-3521          <>
<> Gaussstr. 20              Fax : +49 (0)202 439-2811          <>
<> 42097 Wuppertal                                              <>
<>                                                              <>
<><><><><><><>< Of course it runs NetBSD http://www.netbsd.org ><>
(Continue reading)

Permalink | Reply | 1 comment |
headers
Manuel Bouyer | 3 Sep 2006 19:47

Re: SCSI debugging

On Fri, Sep 01, 2006 at 09:06:09AM +0200, Torsten Harenberg wrote:
> Dear alpha fellows,
> 
> my good ol' PWS 433au (running as our main web and backup server) started 
> to fail occasionally about half a year ago.
> 
> Console says:
> 
> isp0: command timed out for 0.1.0
> ... wpdu2 syslogd: /var/log/messages: imput/output error
> sd0:(isp0:0:0:0): adapter resource shortage
> 
> So I assume it's a hardware problem.
> 
> Does anybody has a good idea how to debug if it's the hard drive or the 
> controller (or both)?
> 
> Any hint is appreciated,

I've also seen this happen with connectors issues. Sometimes
unplugging/replugging the cables fixes it

--

-- 
Manuel Bouyer <bouyer <at> antioche.eu.org>
     NetBSD: 26 ans d'experience feront toujours la difference
--
Permalink | Reply | 0 comments |
headers
Ray Phillips | 8 Sep 2006 10:17

reliability of SMP on alpha

Can anyone comment on how well NetBSD runs on alphas with more than 
one CPU now, please?  I'm wondering since a quad 667 MHz ES40 has 
appeared on ebay.  There were some messages in late July last year 
saying Michael and Jason had fixed the stability problems it had, so 
I suppose it's rock solid by now?  Are the fixes in NetBSD 3.0.1?

Ray
Permalink | Reply | 0 comments |
headers
Brian L. Stuart | 8 Sep 2006 16:18

Re: reliability of SMP on alpha

> Can anyone comment on how well NetBSD runs on alphas with more than 
> one CPU now, please?  I'm wondering since a quad 667 MHz ES40 has 
> appeared on ebay.  There were some messages in late July last year 
> saying Michael and Jason had fixed the stability problems it had, so 
> I suppose it's rock solid by now?  Are the fixes in NetBSD 3.0.1?

I can't really comment on NetBSD 3 because I've been swamped and
haven't done much updating for a while.  But I did apply the patches
(and July last year sounds about right) and it's been rock solid for me
on my DS5300 (white box version of the AS1200) with 2 processors.
To be honest though, I haven't really stressed it that much so I can't
promise there are no gotchas under heavy load though.

BLS
Permalink | Reply | 0 comments |
headers
Iain Hibbert | 10 Sep 2006 18:01

bluetooth on alpha

Hi,
   This week I cross-compiled alpha/GENERIC and alpha/ALPHA kernels with
bluetooth enabled. I can't test it because I have no hardware, but it does
compile at least - is it ok to commit the below change to the alpha port?

iain

Index: sys/arch/alpha/conf/files.alpha
===================================================================
RCS file: /cvsroot/src/sys/arch/alpha/conf/files.alpha,v
retrieving revision 1.169
diff -u -r1.169 files.alpha
--- sys/arch/alpha/conf/files.alpha	29 Jul 2006 19:10:56 -0000	1.169
+++ sys/arch/alpha/conf/files.alpha	10 Sep 2006 13:13:10 -0000
 <at>  <at>  -524,6 +524,10  <at>  <at> 
 #
 include "dev/usb/files.usb"
 #
+# Machine-independent Bluetooth support
+#
+include "dev/bluetooth/files.bluetooth"
+#
 # Machine-independent IEEE 1394 device support
 #
 include "dev/ieee1394/files.ieee1394"
Index: sys/arch/alpha/conf/GENERIC
===================================================================
RCS file: /cvsroot/src/sys/arch/alpha/conf/GENERIC,v
retrieving revision 1.296
diff -u -r1.296 GENERIC
(Continue reading)

Permalink | Reply | 1 comment |
headers
Jason Thorpe | 13 Sep 2006 22:18

Re: bluetooth on alpha


On Sep 10, 2006, at 9:01 AM, Iain Hibbert wrote:

> Hi,
>    This week I cross-compiled alpha/GENERIC and alpha/ALPHA kernels  
> with
> bluetooth enabled. I can't test it because I have no hardware, but  
> it does
> compile at least - is it ok to commit the below change to the alpha  
> port?

Go for it.

>
> iain
>
> Index: sys/arch/alpha/conf/files.alpha
> ===================================================================
> RCS file: /cvsroot/src/sys/arch/alpha/conf/files.alpha,v
> retrieving revision 1.169
> diff -u -r1.169 files.alpha
> --- sys/arch/alpha/conf/files.alpha	29 Jul 2006 19:10:56 -0000	1.169
> +++ sys/arch/alpha/conf/files.alpha	10 Sep 2006 13:13:10 -0000
>  <at>  <at>  -524,6 +524,10  <at>  <at> 
>  #
>  include "dev/usb/files.usb"
>  #
> +# Machine-independent Bluetooth support
> +#
> +include "dev/bluetooth/files.bluetooth"
(Continue reading)

Permalink | Reply | 0 comments |
headers
Aaron J. Grier | 16 Sep 2006 03:16

Re: TLZ10 ??

On Mon, Aug 28, 2006 at 09:01:35AM +0200, Niels S. Eliasen wrote:
> I would like to be able to use the TLZ10 for "dump/restore" and the
> dump works ok, now that I have re-created the device files..but
> whenever I try to do a "restore" I get a "Device busy".... same goes
> for "tar" .....
> 
> 
> 
> Any ideas anyone ?

what does 

mt -f /dev/nrst0 status

give?  (assuming the tape is st0)

--

-- 
  Aaron J. Grier | "Not your ordinary poofy goof." | agrier <at> poofygoof.com
              "silly brewer, saaz are for pils!"  --  virt
Permalink | Reply | 0 comments |
headers
Garrett D'Amore | 16 Sep 2006 10:42

Need help with timecounters/todr

I could use some help with todr and timecounters on alpha.  Any intrepid
souls want to help tackle it?  SMP expertise would be helpful.

--

-- 
Garrett D'Amore, Principal Software Engineer
Tadpole Computer / Computing Technologies Division,
General Dynamics C4 Systems
http://www.tadpolecomputer.com/
Phone: 951 325-2134  Fax: 951 325-2191
Permalink | Reply | 10 comments |
headers
Tobias Nygren | 16 Sep 2006 14:06
Picon

Re: Need help with timecounters/todr

Garrett D'Amore wrote:
> I could use some help with todr and timecounters on alpha.  Any intrepid
> souls want to help tackle it?  SMP expertise would be helpful.
>
>   
I actually did a semi-working implementation a couple of weeks
ago, before the generic todr changes.
While the old patch works in uniprocessor mode the algorithm
wasn't very well thought-through. I've made some major
improvements since then. I will clean up the new patch and
post it if someone wants to try it.

I think for some alpha machines that provide an i8254 timer it
would be nice to be able to use it. But for this the i8254 code
needs to be made MI. Maybe this is already work-in-progress?

SMP box running my local pcc timecounter patches:

ntpq> rv
assID=0 status=06f4 leap_none, sync_ntp, 15 events, event_peer/strat_chg,
version="ntpd 4.2.2p2-o Fri Aug 25 15:11:34 GMT 2006 (1)",
processor="alpha", system="NetBSD/4.99.1", leap=00, stratum=2,
precision=-19, rootdelay=6.355, rootdispersion=47.683, peer=63146,
refid=192.36.144.23,
reftime=c8b66191.defe12f1  Sat, Sep 16 2006 13:42:09.871, poll=10,
clock=c8b664e5.160989b9  Sat, Sep 16 2006 13:56:21.086, state=4,
offset=11.155, frequency=11.421, jitter=4.069, noise=3.572,
stability=0.610, tai=0

-Tobias
(Continue reading)

Permalink | Reply | 0 comments |
headers
Greg A. Woods | 18 Sep 2006 21:11
X-Face
Favicon

Re: reliability of SMP on alpha

At Fri, 8 Sep 2006 18:17:11 +1000,
Ray Phillips wrote:
> 
> Can anyone comment on how well NetBSD runs on alphas with more than
> one CPU now, please?  I'm wondering since a quad 667 MHz ES40 has
> appeared on ebay.  There were some messages in late July last year
> saying Michael and Jason had fixed the stability problems it had, so
> I suppose it's rock solid by now?  Are the fixes in NetBSD 3.0.1?

1.6.x, with the various stability patches, is very solid, though still
not 100% perfect, on the various multi-processor boxes I'm running (4000
(off now, but had 2 cpus), 4100 with 3 cpus, ES40 with 4 cpus).  The
ES40 has run for as long as 140 days under constant moderate load with
Cyrus IMAP, Smail SMTP, and Apache HTTP being the primary applications.

I'm still working on integrating my changes with the netbsd-3 tree and
so I'm not sure yet if all the stability patches are in there already or
not (or had missed the netbsd-3 branch and haven't been pulled up), but
I'm assuming they are there since most (though perhaps not all) of the
ones I applied to my local netbsd-1-6 tree were taken from CVS and that
would have happened sometime before the netbsd-3 branch.  However if any
of the related PRs are still open then the latest changes suggested in
those PRs may not yet be committed.  Michael and Jason are probably the
two people who'll know the most about any as-yet uncommitted changes.

--

-- 
						Greg A. Woods

H:+1 416 218-0098 W:+1 416 489-5852 x122 VE3TCP RoboHack <woods <at> robohack.ca>
Planix, Inc. <woods <at> planix.com>       Secrets of the Weird <woods <at> weird.com>
(Continue reading)

Permalink | Reply | 2 comments |

Gmane