problems with sata disks (taskqueue timeout)

Marc UBM Bocklet ubm at u-boot-man.de
Fri May 8 20:00:06 UTC 2009


On Sun, 29 Mar 2009 11:01:53 +0200
Marc "UBM" Bocklet <ubm at u-boot-man.de> wrote:

> On Tue, 20 Jan 2009 08:08:29 +0100
> Marc "UBM" Bocklet <ubm at u-boot-man.de> wrote:
> 
> > On Tue, 20 Jan 2009 09:39:51 +1100
> > Andrew Snow <andrew at modulus.org> wrote:
> > 
> > > 
> > > I think that if you use eSATA you probably need dedicated eSATA 
> > > controller ports.  eSATA standard specifies a higher voltage for
> > > the longer cable distances.
> > > 
> > > Judging from the sporadic problem reports, Promise TX4 is probably
> > > not the best at signal purity to begin with so using it for eSATA
> > > pushes it over the edge.
> > > 
> > > 
> > > Hope that helps,
> > 
> > Thanks for the fast answer! :-)
> > 
> > Although my version of the TX4 has two dedicated e-sata ports, the
> > other posts seem to indicate that it got something to do with the
> > controller (maybe signal purity, like you said). I'll try upgrading
> > next and will report back after that.
> 
> A very late followup here:
> 
> I upgraded to the latest stable, but things did not improve:
> 
> Mar 29 10:57:29 hamstor kernel: ad10: WARNING - WRITE_DMA48 UDMA ICRC
> error (retrying request) LBA=1087300992 Mar 29 10:57:34 hamstor
> kernel: ad10: FAILURE - SET_MULTI status=51<READY,DSC,ERROR>
> error=4<ABORTED>
> 
> Mar 29 10:57:34 hamstor kernel: ad10: TIMEOUT - WRITE_DMA48 retrying
> (0 retries left) LBA=1087300992
> 
> Mar 29 10:57:34 hamstor kernel: ad10: FAILURE - WRITE_DMA48
> status=ff<BUSY,READY,DMA_READY,DSC,DRQ,CORRECTABLE,INDEX,ERROR>
> error=ff<ICRC,UNCORRECTABLE,MEDIA_CHANGED,NID_NOT_FOUND,MEDIA_CHANGE_REQEST,ABORTED,NO_MEDIA,ILLEGAL_LENGTH>
> LBA=1087300992 
> 
> Mar 29 10:57:34 hamstor root: ZFS: vdev I/O failure,
> zpool=gedaerm path=/dev/ad10 offset=556698042368 size=131072 error=5
> 
> Mar 29 10:57:43 hamstor kernel: ad10: WARNING - SETFEATURES SET
> TRANSFER MODE taskqueue timeout - completing request directly 
> 
> Mar 29 10:57:47 hamstor kernel: ad10: WARNING - SETFEATURES SET
> TRANSFER MODE taskqueue timeout - completing request directly 
> 
> Mar 29 10:57:51 hamstor kernel: ad10: WARNING - SETFEATURES ENABLE
> WCACHE taskqueue timeout - completing request directly 
> 
> Mar 29 10:57:55 hamstor kernel: ad10: WARNING - SET_MULTI taskqueue
> timeout - completing request directly 
> 
> Mar 29 10:57:55 hamstor kernel: ad10: TIMEOUT - WRITE_DMA48 retrying
> (1 retry left) LBA=1087301248 
> 
> Mar 29 10:57:55 hamstor kernel: ad10: WARNING - WRITE_DMA48 UDMA ICRC
> error (retrying request) LBA=1087301248 
> 
> Mar 29 10:58:00 hamstor kernel: ad10: FAILURE - SET_MULTI
> status=51<READY,DSC,ERROR> error=4<ABORTED> 
> 
> Mar 29 10:58:00 hamstor kernel: ad10: FAILURE - WRITE_DMA48 timed out
> LBA=1087301248
> 
> Mar 29 10:58:00 hamstor root: ZFS: vdev I/O failure, zpool=gedaerm
> path=/dev/ad10 offset=556698173440 size=131072 error=5
> 
> 
> 
> Any further ideas anybody? :-)

Another update, upgrading to -current dating from April 25th 2009 seems
to have "fixed" the problem, I've encountered no errors as of yet and
I've copied about 250GB in large chunks, something that was sure to
provoke the errors with -stable.

FreeBSD xxx 8.0-CURRENT FreeBSD 8.0-CURRENT #1: Sat Apr 25 13:33:18
CEST 2009 xxx:/usr/obj/usr/src/sys/xxx  amd64

Bye
Marc

-- 
"And what rough beast, its hour come round at last,
Slouches towards Bethlehem to be born?"

W.B. Yeats, The Second Coming


More information about the freebsd-stable mailing list