<2017 March>

On this page...



Member of...

ASP Insiders

MVP Visual Developer ASP/ASP.NET

Enter CodeZone

Blog Categories



Deutsche Resourcen


Sign In

#  Tuesday, 03 January 2006

This sordid story starts with a good thing to do: I wanted to perform a backup onto my external USB hard disk. So I took the drive that was attached to my old workstation and did my routine: delete an old backup to make space for new ones. But Windows decided that it might be a rather good time to annoy me:

Huh, what are you trying to tell me? At the same time, a tray notification popped up:

USB is usually not a network connection, and I most decidedly did not abuse the cord for bungee jumping at that time. This error also yielded tons of event log entries like this one here:

Being already a little ticked off, I decided to take a peek into Computer Management, section Disk Management:

Gotta be kidding me! Being annoyed already, I dug deeper into the property sheets for this drive:

You remember the error messages about delayed write failures as well as paging? Write caching is disabled?

For good measure, I attached my harddisk to another (non-x64) machine. Working fine there. Next, I switched USB cables. Same result. Used a different external USB hard drive. Again, same result. So that pretty much means that external hard disks don't work at my new workstation.

I am wondering if that is another problem of x64, or the chipset driver in this case. Cables and physical disks are already ruled out - ideas?

Update My drive enclosures have Firewire support too. So I connected the drives via Firewire. That works around the DWF problem, however, now I stumbled into a different issue (event log record):

The device, \Device\Sbp2\ASSMANN AB-PENR35 Combo Devi, did not respond within the timeout period.

The KB article SBP-2 drive stops responding when you try to write data in Windows XP nicely fits this error. Back to square one, research on the USB issue. At least I am not the only one that experiences those kinds of DWF problems, as an Internet search proves.

Update Tried again with USB after updating the nForce chipset driver to v6.82. It got farther this time, but it still crashes with DWF. Judging from a search for "delayed write failed", I am most decidedly not alone.

Solution At least sort-of... I poached ye olde Adaptec DuoConnect card from my old workstation and plugged it into the shiny new one (which, thanks to the A8N board, already has 10 USB 2.0 ports of its own). Booted the machine, drivers were installed automatically, plugged in the harddisk, did the same operations as before - and it worked flawlessly.

Conclusion: There must be an issue in the combination chipset / NVIDIA nForce (x64 only maybe?) chipset driver.

Seems like my computer is living up to its name.

Update I decided that I had to have another go at it. So I bought a brand-new external 3.5" enclosure, a Map-H31S (according to the documentation it should use a Genesyslogic GL811E). I disassembled the old enclosure and put the hard disk in the new enclosure. Connected it to the mainboard's USB connectors - et voila, it works! Seems that this enclosure's IDE to USB chip can deal with my motherboard.

Categories: this | x64
Tuesday, 03 January 2006 14:40:39 (W. Europe Standard Time, UTC+01:00)  #    Comments [2]


Saturday, 20 May 2006 19:46:02 (W. Europe Daylight Time, UTC+02:00)
I bought a new Hitachi 80GB External HDD and connected to my DELL Latitude D600. It consistently prompted the "Delayed Write Failed" error. After research on the Web, I have changed the setting under:
Control Panel -> System -> Hardware -> Device Manager -> Universal Serial Bus controllers -> USB Root Hub -> Properties -> Power Management

such that the option "Allow the computer to turn off this device to save power" is UNCHECKED. My external HDD is working fine now.
Wednesday, 10 January 2007 15:40:02 (W. Europe Standard Time, UTC+01:00)
I suspect that the problem is a packet size being sent is too large for the chipset on the drive enclosure to handle. I've seen this problem on Windows 98, 2000 and XP on at least three different machines (2 Dell laptops and 1 ASUS bases home built). As a result of this particular problem, I've gone off USB drives altogether. This conclusion is based upon a blog from a guy who said that he'd put a packet snooper on to monitor communications between box and drive. He said that when a too large packet was sent, the drive would drop the packet and that would put an end to communications between box and drive. My own experience confirms this. I only encountered the error when moving large amounts of data, as in a backup. It has been long enough that I've forgotten the problem packet size.
Comments are closed.

© Copyright 2017 Christoph Wille

newtelligence dasBlog 2.3.9074.18820
Subscribe to this weblog's RSS feed with SharpReader, Radio Userland, NewsGator or any other aggregator listening on port 5335 by clicking this button.   RSS 2.0|Atom 1.0  Send mail to the author(s)

Don't contact us via this (fleischfalle@alphasierrapapa.com) email address.