"Delayed write failed" while firewire drive is not doing a..

Ian

Distinguished
Apr 5, 2004
820
0
18,980
Archived from groups: microsoft.public.windowsxp.hardware (More info?)

I have a 4-bay external firewire enclosure, with Oxford Semiconductor
911 chipset. I have 4 drives connected (a couple of Maxtors, a Western
Digital and a Hitachi). Every so often (many times a day) I get
"Delayed write failed" error on one or more of the disks (usually the
Hitachi or one of the Maxtors), even though the disk is inactive.

I'm running Windows XP SP2 on a machine with an Athlon K7T266Pro
board. I have write caching DISABLED (the disks are "optimised for
quick removal"), so I can't see why there should be any delayed write
activity at all, let alone when I'm not actually using the disks.

I've seen alot of messages concerning the "delayed write" problem,
particularly with XP SP2, but would be keen to understand why I get
these errors when I'm not actually trying to write to the disks.

I've tried plugging the same enclosure into a different PC (an Acer
T120) and get exactly the same problem. This is also running XP SP2.

I've seen a recommendation to change the PCI latency in the BIOS, but
this option doesn't seem to be available in my BIOS settings.

I've also run a utility from the BUSTrace site to check whether write
packet sizes are an issue, but it didn't generate any failures.

Does anyone know why I would get these errors when the disks are not
being used??? Better still, can anyone suggest a workaround??

TIA for any feedback
Ian
 
G

Guest

Guest
Archived from groups: microsoft.public.windowsxp.hardware (More info?)

Ian Wrote:
> I have a 4-bay external firewire enclosure, with Oxford Semiconductor
> 911 chipset. I have 4 drives connected (a couple of Maxtors, a Western
> Digital and a Hitachi). Every so often (many times a day) I get
> "Delayed write failed" error on one or more of the disks (usually the
> Hitachi or one of the Maxtors), even though the disk is inactive.
>
> I'm running Windows XP SP2 on a machine with an Athlon K7T266Pro
> board. I have write caching DISABLED (the disks are "optimised for
> quick removal"), so I can't see why there should be any delayed write
> activity at all, let alone when I'm not actually using the disks.
>
> I've seen alot of messages concerning the "delayed write" problem,
> particularly with XP SP2, but would be keen to understand why I get
> these errors when I'm not actually trying to write to the disks.
>
> I've tried plugging the same enclosure into a different PC (an Acer
> T120) and get exactly the same problem. This is also running XP SP2.
>
> I've seen a recommendation to change the PCI latency in the BIOS, but
> this option doesn't seem to be available in my BIOS settings.
>
> I've also run a utility from the BUSTrace site to check whether write
> packet sizes are an issue, but it didn't generate any failures.
>
> Does anyone know why I would get these errors when the disks are not
> being used??? Better still, can anyone suggest a workaround??
>
> TIA for any feedback
> Ian


Hello Ian,

I too have the exact same problems with this error. I too have a 4 bay
FW case with Oxford911 chipset based bridgeboards.

Just like you, I can get this error simply by leaving the enclosure
turned on, and not be doing ANYTHING at all on the PC.

I have tried everything, including the PCI latency thing but to no
avail. I even bought and tried the uniBrian drivers from ubiCore but
while there were no delayed write errors withe these drivers, they can
only see one drive per board.

Writing to the author of the drivers has not produced much except that
he needs a multi-bay enclosure to be able to make changes to his
driver. He has never heard of the Oxford 911 being able to support two
drives per board.

Perhaps you've run across the BusTrace site? I've tried everything
there as well.

At this point, I'm thinking of going back to Windows 2000 and see if
that helps.


--
Hartosay
 

Ian

Distinguished
Apr 5, 2004
820
0
18,980
Archived from groups: microsoft.public.windowsxp.hardware (More info?)

On Sat, 3 Sep 2005 01:08:06 +0000, Hartosay
<Hartosay.1urppn@pcbanter.net> wrote:

>
>Ian Wrote:
>> I have a 4-bay external firewire enclosure, with Oxford Semiconductor
>> 911 chipset. I have 4 drives connected (a couple of Maxtors, a Western
>> Digital and a Hitachi). Every so often (many times a day) I get
>> "Delayed write failed" error on one or more of the disks (usually the
>> Hitachi or one of the Maxtors), even though the disk is inactive.
>>
>> I'm running Windows XP SP2 on a machine with an Athlon K7T266Pro
>> board. I have write caching DISABLED (the disks are "optimised for
>> quick removal"), so I can't see why there should be any delayed write
>> activity at all, let alone when I'm not actually using the disks.
>>
>> I've seen alot of messages concerning the "delayed write" problem,
>> particularly with XP SP2, but would be keen to understand why I get
>> these errors when I'm not actually trying to write to the disks.
>>
>> I've tried plugging the same enclosure into a different PC (an Acer
>> T120) and get exactly the same problem. This is also running XP SP2.
>>
>> I've seen a recommendation to change the PCI latency in the BIOS, but
>> this option doesn't seem to be available in my BIOS settings.
>>
>> I've also run a utility from the BUSTrace site to check whether write
>> packet sizes are an issue, but it didn't generate any failures.
>>
>> Does anyone know why I would get these errors when the disks are not
>> being used??? Better still, can anyone suggest a workaround??
>>
>> TIA for any feedback
>> Ian
>
>
>Hello Ian,
>
>I too have the exact same problems with this error. I too have a 4 bay
>FW case with Oxford911 chipset based bridgeboards.
>
>Just like you, I can get this error simply by leaving the enclosure
>turned on, and not be doing ANYTHING at all on the PC.
>
>I have tried everything, including the PCI latency thing but to no
>avail. I even bought and tried the uniBrian drivers from ubiCore but
>while there were no delayed write errors withe these drivers, they can
>only see one drive per board.
>
>Writing to the author of the drivers has not produced much except that
>he needs a multi-bay enclosure to be able to make changes to his
>driver. He has never heard of the Oxford 911 being able to support two
>drives per board.
>
>Perhaps you've run across the BusTrace site? I've tried everything
>there as well.
>
>At this point, I'm thinking of going back to Windows 2000 and see if
>that helps.

Hi,

Thanks for your reply. Since I posted this I seem to have found a
workaround. Using the "Power Options" in Control panel I've set the
hard drives to never power down. So far it's been a few days and I
haven't seen a recurrence of the problem.

It's not an ideal solution, but it seems to get around the immediate
problem.

Please let me know if you discover a better solution...

Ian
 
G

Guest

Guest
Archived from groups: microsoft.public.windowsxp.hardware (More info?)

Ian,

I read somewhere about Cacheman 5.50 (the free version) not
CachemanXP.

Using it and your suggestion to disable power management for the hard
drives has worked so far, 6 hours and counting without a delayed write
error (knock on wood)...


--
Hartosay
 
G

Guest

Guest
Archived from groups: microsoft.public.windowsxp.hardware (More info?)

Hartosay Wrote:
> Ian,
>
> I read somewhere about Cacheman 5.50 (the free version) not
> CachemanXP.
>
> Using it and your suggestion to disable power management for the hard
> drives has worked so far, 6 hours and counting without a delayed write
> error (knock on wood)...


Ian,

take a look at this page:

http://tinyurl.com/9n9b6

scroll down to response #9.

I have been running now for more than 12 hours with no delayed write
errors.

whether its due to setting the "Turn off Hard Disks" in the power
management applet to "Never" and/or to the cacheman utility is
something for me to find out.

Right now I'm just happy that I haven't seen one of the dreaded
"Delayed Write Error" popups in half a day.

Cheers.


--
Hartosay
 

Ian

Distinguished
Apr 5, 2004
820
0
18,980
Archived from groups: microsoft.public.windowsxp.hardware (More info?)

On Mon, 5 Sep 2005 22:29:46 +0000, Hartosay
<Hartosay.1uwt1p@pcbanter.net> wrote:

>
>Hartosay Wrote:
>> Ian,
>>
>> I read somewhere about Cacheman 5.50 (the free version) not
>> CachemanXP.
>>
>> Using it and your suggestion to disable power management for the hard
>> drives has worked so far, 6 hours and counting without a delayed write
>> error (knock on wood)...
>
>
>Ian,
>
>take a look at this page:
>
>http://tinyurl.com/9n9b6
>
>scroll down to response #9.
>
>I have been running now for more than 12 hours with no delayed write
>errors.
>
>whether its due to setting the "Turn off Hard Disks" in the power
>management applet to "Never" and/or to the cacheman utility is
>something for me to find out.
>
>Right now I'm just happy that I haven't seen one of the dreaded
>"Delayed Write Error" popups in half a day.
>
>Cheers.

Thanks for the pointer. I've tried Cacheman as an alternative to the
"turn off hard disks" setting, but found that it didn't work for me.
Have you made any further discoveries?

Ian