Re: WD Enterprise and TCQ

From: Henning P. Schmiedehausen (hps@tanstaafl.de)
Date: Tue Jan 18 2000 - 06:48:11 EST


belg@pheonix.org (Belgarath) writes:

>I'm curious to find out if the tagged command queuing problem with Western
>Digital Enterprise drives vs. the aic7xxx driver is soemthing fixable on
>the kernel end, or if it's something that's inherent in the WDE series of
>drives.

>I've got a server out in California that would really benefit from TCQ,
>but was built with the WDE Vendor: WDIGTL Model: WDE18310 ULTRA2 Rev:
>1.30 Drive.

>As would be expected, I'm getting errors when TCQ is enabled under a
>2.2.13 kernel, culminating in resets, and occasionally, total loss of
>service... requiring a trip to the colo.

>Is there any hope in sight for TCQ on the Enterprise drives?

>Errors it's reporting (with TCQ set to 96):

>(scsi0:0:0:-1) Unexpected busfree, LASTPHASE = 0x40, SEQADDR = 0x5f
>(scsi0:0:0:0) Parity error during Command phase.
>scsi : aborting command due to timeout : pid 2522656, scsi0, channel 0, id
>0, lun 0 0x0a 02 c0 87 02 00
>SCSI host 0 abort (pid 2522655) timed out - resetting
>SCSI bus is being reset for host 0 channel 0.

This looks exactly like the error I got from my ST39173LC drives when
I was using a bad SCA to Ultra2 converter. I switched to a real SCA
tray with backplane and now they work like a charm with the aic7xxx
driver under 2.2.1[2-4]. Rock solid and fast [1].

So this _could_ also be a bad cabling / termination problem.

        Regards
                Henning

[1] 65.86 MB/sec reading rate with bonnie. Five of these beasts
    striped as RAID0 on an single 7860 controller running sync @ 80
    MB/sec. Yes, this is Mega_byte_. Yes, this is from disk, not from
    buffer cache.

-- 
Dipl.-Inf. (Univ.) Henning P. Schmiedehausen --             hps@tanstaafl.de
TANSTAAFL! Consulting - Unix, Internet, Security      

Hutweide 15 Fon.: 09131 / 50654-0 "There ain't no such D-91054 Buckenhof Fax.: 09131 / 50654-20 thing as a free Linux"

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Sun Jan 23 2000 - 21:00:17 EST