RE: Need help with SATA disk timing out in 8.1 Beta

2010-06-19 Thread Graeme Dargie


-Original Message-
From: Jerry Bell [mailto:je...@nrdx.com] 
Sent: 18 June 2010 06:11
To: freebsd-questions@freebsd.org
Subject: Need help with SATA disk timing out in 8.1 Beta

I am having all sorts of problems with drives in a new server.
I have a 450G sata drive that hold my root partition, works great, no 
issues.
I have a second, 1TB drive that has been all sorts of trouble.  When 
writing to this disk, I occasionally see errors like this:

Jun 17 07:40:36 www3 kernel: ad8: WARNING - WRITE_DMA48 UDMA ICRC error 
(retrying request) LBA=1564898207
Jun 17 07:40:36 www3 kernel: ad8: FAILURE - WRITE_DMA48 
status=51READY,DSC,ERROR error=10NID_NOT_FOUND LBA=1564898207
Jun 17 07:57:12 www3 kernel: ad8: WARNING - WRITE_DMA48 UDMA ICRC error 
(retrying request) LBA=1565052351
Jun 17 07:57:12 www3 kernel: ad8: FAILURE - WRITE_DMA48 
status=51READY,DSC,ERROR error=10NID_NOT_FOUND LBA=1565052351
Jun 17 09:45:12 www3 kernel: ad8: WARNING - WRITE_DMA48 UDMA ICRC error 
(retrying request) LBA=1565983775
Jun 17 09:45:12 www3 kernel: ad8: FAILURE - WRITE_DMA48 
status=51READY,DSC,ERROR error=10NID_NOT_FOUND LBA=1565983775
Jun 17 09:50:24 www3 kernel: ad8: WARNING - WRITE_DMA48 UDMA ICRC error 
(retrying request) LBA=1566082719
Jun 17 09:50:24 www3 kernel: ad8: FAILURE - WRITE_DMA48 
status=51READY,DSC,ERROR error=10NID_NOT_FOUND LBA=1566082719
Jun 17 10:01:25 www3 kernel: ad8: WARNING - WRITE_DMA48 UDMA ICRC error 
(retrying request) LBA=1566358623
Jun 17 10:01:25 www3 kernel: ad8: FAILURE - WRITE_DMA48 
status=51READY,DSC,ERROR error=10NID_NOT_FOUND LBA=1566358623
Jun 17 10:02:59 www3 kernel: ad8: WARNING - WRITE_DMA48 UDMA ICRC error 
(retrying request) LBA=1566387807
Jun 17 10:02:59 www3 kernel: ad8: FAILURE - WRITE_DMA48 
status=51READY,DSC,ERROR error=10NID_NOT_FOUND LBA=1566387807
Jun 17 10:18:59 www3 kernel: ad8: WARNING - WRITE_DMA UDMA ICRC error 
(retrying request) LBA=43231
Jun 17 10:18:59 www3 kernel: ad8: WARNING - WRITE_DMA UDMA ICRC error 
(retrying request) LBA=57567
Jun 17 10:18:59 www3 kernel: ad8: WARNING - WRITE_DMA UDMA ICRC error 
(retrying request) LBA=773471
Jun 17 10:18:59 www3 kernel: ad8: WARNING - WRITE_DMA UDMA ICRC error 
(retrying request) LBA=786271
Jun 17 10:18:59 www3 kernel: ad8: WARNING - WRITE_DMA UDMA ICRC error 
(retrying request) LBA=810079
Jun 17 10:19:00 www3 kernel: ad8: WARNING - WRITE_DMA UDMA ICRC error 
(retrying request) LBA=76767
Jun 17 10:19:00 www3 kernel: ad8: WARNING - WRITE_DMA UDMA ICRC error 
(retrying request) LBA=784479

Last week, I asked the datacenter to provide me with a new 1TB drive, 
and they did.  It formatted fine, no errors.  I copied files to it, ran 
bonnie, etc, and no signs of any DMA issues.
Until this morning when I started having the errors again.

If I run a tool like bonnie, I am very easily reproduce the errors.  
After some research, I find that these errors are often indicative of 
SATA cable problems.
The datacenter replaced the cable, and the problem continues.
The datacenter moved the sata cable to a new SATA port, and the problem 
continues
The datacenter adds a BRAND NEW 1TB drive (now the system has 3 drive), 
and I am unable to format the drive because of these errors:
ad10: WARNING - WRITE_DMA UDMA ICRC error (retrying request)
LBA=168172351
ad10: WARNING - WRITE_DMA48 UDMA ICRC error (retrying request)
LBA=602334847
ad10: FAILURE - WRITE_DMA48 status=51READY,DSC,ERROR 
error=10NID_NOT_FOUND LBA=602334847
ad10: WARNING - WRITE_DMA48 UDMA ICRC error (retrying request)
LBA=427014463
ad10: FAILURE - WRITE_DMA48 status=51READY,DSC,ERROR 
error=10NID_NOT_FOUND LBA=427014463
ad10: WARNING - WRITE_DMA UDMA ICRC error (retrying request)
LBA=15425407
ad10: WARNING - WRITE_DMA48 UDMA ICRC error (retrying request)
LBA=471408895
ad10: FAILURE - WRITE_DMA48 status=51READY,DSC,ERROR 
error=10NID_NOT_FOUND LBA=471408895
ad10: WARNING - WRITE_DMA UDMA ICRC error (retrying request)
LBA=91422655
ad10: WARNING - WRITE_DMA UDMA ICRC error (retrying request)
LBA=203161183
ad10: WARNING - WRITE_DMA48 UDMA ICRC error (retrying request) 
LBA=1211817727
ad10: FAILURE - WRITE_DMA48 status=51READY,DSC,ERROR 
error=10NID_NOT_FOUND LBA=1211817727
ad10: WARNING - WRITE_DMA UDMA ICRC error (retrying request)
LBA=37998847
ad10: WARNING - WRITE_DMA48 UDMA ICRC error (retrying request)
LBA=309632575
ad10: FAILURE - WRITE_DMA48 status=51READY,DSC,ERROR 
error=10NID_NOT_FOUND LBA=309632575
ad10: TIMEOUT - WRITE_DMA retrying (1 retry left) LBA=24831007
ad10: WARNING - WRITE_DMA UDMA ICRC error (retrying request)
LBA=59067391
ad10: WARNING - WRITE_DMA48 UDMA ICRC error (retrying request)
LBA=497744575
ad10: FAILURE - WRITE_DMA48 status=51READY,DSC,ERROR 
error=10NID_NOT_FOUND LBA=497744575
ad10: FAILURE - WRITE_MUL status=51READY,DSC,ERROR 
error=84ICRC,ABORTED LBA=1128895
ad10: WARNING - WRITE_DMA UDMA ICRC error (retrying request)
LBA=13920511
ad10: WARNING - WRITE_DMA48 UDMA ICRC error (retrying request)
LBA=547029919
ad10: FAILURE - WRITE_DMA48 status

Re: Need help with SATA disk timing out in 8.1 Beta

2010-06-19 Thread Adam Vande More
On Fri, Jun 18, 2010 at 12:11 AM, Jerry Bell je...@nrdx.com wrote:

 I am having all sorts of problems with drives in a new server.
 I have a 450G sata drive that hold my root partition, works great, no
 issues.
 I have a second, 1TB drive that has been all sorts of trouble.  When
 writing to this disk, I occasionally see errors like this:

 Jun 17 07:40:36 www3 kernel: ad8: WARNING - WRITE_DMA48 UDMA ICRC error
 (retrying request) LBA=1564898207
 snip
 I am at the end of my ability to troubleshoot this.  Could this be a
 problem with FreeBSD 8.1 beta and not the drives after all?
 I have seen a reference to a patch for previous versions that increase the
 DMA timeout time to 10 or 15 seconds, which fixes problems, but I am not
 certain that would fix my particular issue.


You could use ahci which might workaround the issue and give you better
performance.

load it from /boot/loader.conf

beware it will change names of detected devices, you may want to consider
using glabel.

-- 
Adam Vande More
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org


Re: Need help with SATA disk timing out in 8.1 Beta

2010-06-18 Thread Matthias Gamsjager
Have you changed the cable?
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org


Re: Need help with SATA disk timing out in 8.1 Beta

2010-06-18 Thread Jerry Bell

Yes, twice.
On 6/18/2010 4:52 AM, Matthias Gamsjager wrote:

Have you changed the cable?
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
   


___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org


Need help with SATA disk timing out in 8.1 Beta

2010-06-17 Thread Jerry Bell

I am having all sorts of problems with drives in a new server.
I have a 450G sata drive that hold my root partition, works great, no 
issues.
I have a second, 1TB drive that has been all sorts of trouble.  When 
writing to this disk, I occasionally see errors like this:


Jun 17 07:40:36 www3 kernel: ad8: WARNING - WRITE_DMA48 UDMA ICRC error 
(retrying request) LBA=1564898207
Jun 17 07:40:36 www3 kernel: ad8: FAILURE - WRITE_DMA48 
status=51READY,DSC,ERROR error=10NID_NOT_FOUND LBA=1564898207
Jun 17 07:57:12 www3 kernel: ad8: WARNING - WRITE_DMA48 UDMA ICRC error 
(retrying request) LBA=1565052351
Jun 17 07:57:12 www3 kernel: ad8: FAILURE - WRITE_DMA48 
status=51READY,DSC,ERROR error=10NID_NOT_FOUND LBA=1565052351
Jun 17 09:45:12 www3 kernel: ad8: WARNING - WRITE_DMA48 UDMA ICRC error 
(retrying request) LBA=1565983775
Jun 17 09:45:12 www3 kernel: ad8: FAILURE - WRITE_DMA48 
status=51READY,DSC,ERROR error=10NID_NOT_FOUND LBA=1565983775
Jun 17 09:50:24 www3 kernel: ad8: WARNING - WRITE_DMA48 UDMA ICRC error 
(retrying request) LBA=1566082719
Jun 17 09:50:24 www3 kernel: ad8: FAILURE - WRITE_DMA48 
status=51READY,DSC,ERROR error=10NID_NOT_FOUND LBA=1566082719
Jun 17 10:01:25 www3 kernel: ad8: WARNING - WRITE_DMA48 UDMA ICRC error 
(retrying request) LBA=1566358623
Jun 17 10:01:25 www3 kernel: ad8: FAILURE - WRITE_DMA48 
status=51READY,DSC,ERROR error=10NID_NOT_FOUND LBA=1566358623
Jun 17 10:02:59 www3 kernel: ad8: WARNING - WRITE_DMA48 UDMA ICRC error 
(retrying request) LBA=1566387807
Jun 17 10:02:59 www3 kernel: ad8: FAILURE - WRITE_DMA48 
status=51READY,DSC,ERROR error=10NID_NOT_FOUND LBA=1566387807
Jun 17 10:18:59 www3 kernel: ad8: WARNING - WRITE_DMA UDMA ICRC error 
(retrying request) LBA=43231
Jun 17 10:18:59 www3 kernel: ad8: WARNING - WRITE_DMA UDMA ICRC error 
(retrying request) LBA=57567
Jun 17 10:18:59 www3 kernel: ad8: WARNING - WRITE_DMA UDMA ICRC error 
(retrying request) LBA=773471
Jun 17 10:18:59 www3 kernel: ad8: WARNING - WRITE_DMA UDMA ICRC error 
(retrying request) LBA=786271
Jun 17 10:18:59 www3 kernel: ad8: WARNING - WRITE_DMA UDMA ICRC error 
(retrying request) LBA=810079
Jun 17 10:19:00 www3 kernel: ad8: WARNING - WRITE_DMA UDMA ICRC error 
(retrying request) LBA=76767
Jun 17 10:19:00 www3 kernel: ad8: WARNING - WRITE_DMA UDMA ICRC error 
(retrying request) LBA=784479


Last week, I asked the datacenter to provide me with a new 1TB drive, 
and they did.  It formatted fine, no errors.  I copied files to it, ran 
bonnie, etc, and no signs of any DMA issues.

Until this morning when I started having the errors again.

If I run a tool like bonnie, I am very easily reproduce the errors.  
After some research, I find that these errors are often indicative of 
SATA cable problems.

The datacenter replaced the cable, and the problem continues.
The datacenter moved the sata cable to a new SATA port, and the problem 
continues
The datacenter adds a BRAND NEW 1TB drive (now the system has 3 drive), 
and I am unable to format the drive because of these errors:

ad10: WARNING - WRITE_DMA UDMA ICRC error (retrying request) LBA=168172351
ad10: WARNING - WRITE_DMA48 UDMA ICRC error (retrying request) LBA=602334847
ad10: FAILURE - WRITE_DMA48 status=51READY,DSC,ERROR 
error=10NID_NOT_FOUND LBA=602334847

ad10: WARNING - WRITE_DMA48 UDMA ICRC error (retrying request) LBA=427014463
ad10: FAILURE - WRITE_DMA48 status=51READY,DSC,ERROR 
error=10NID_NOT_FOUND LBA=427014463

ad10: WARNING - WRITE_DMA UDMA ICRC error (retrying request) LBA=15425407
ad10: WARNING - WRITE_DMA48 UDMA ICRC error (retrying request) LBA=471408895
ad10: FAILURE - WRITE_DMA48 status=51READY,DSC,ERROR 
error=10NID_NOT_FOUND LBA=471408895

ad10: WARNING - WRITE_DMA UDMA ICRC error (retrying request) LBA=91422655
ad10: WARNING - WRITE_DMA UDMA ICRC error (retrying request) LBA=203161183
ad10: WARNING - WRITE_DMA48 UDMA ICRC error (retrying request) 
LBA=1211817727
ad10: FAILURE - WRITE_DMA48 status=51READY,DSC,ERROR 
error=10NID_NOT_FOUND LBA=1211817727

ad10: WARNING - WRITE_DMA UDMA ICRC error (retrying request) LBA=37998847
ad10: WARNING - WRITE_DMA48 UDMA ICRC error (retrying request) LBA=309632575
ad10: FAILURE - WRITE_DMA48 status=51READY,DSC,ERROR 
error=10NID_NOT_FOUND LBA=309632575

ad10: TIMEOUT - WRITE_DMA retrying (1 retry left) LBA=24831007
ad10: WARNING - WRITE_DMA UDMA ICRC error (retrying request) LBA=59067391
ad10: WARNING - WRITE_DMA48 UDMA ICRC error (retrying request) LBA=497744575
ad10: FAILURE - WRITE_DMA48 status=51READY,DSC,ERROR 
error=10NID_NOT_FOUND LBA=497744575
ad10: FAILURE - WRITE_MUL status=51READY,DSC,ERROR 
error=84ICRC,ABORTED LBA=1128895

ad10: WARNING - WRITE_DMA UDMA ICRC error (retrying request) LBA=13920511
ad10: WARNING - WRITE_DMA48 UDMA ICRC error (retrying request) LBA=547029919
ad10: FAILURE - WRITE_DMA48 status=51READY,DSC,ERROR 
error=10NID_NOT_FOUND LBA=547029919


So, the problem has occurred on 3 different drives.
SATA ports and cables do not appear to impact the problem.
The