System hangs after USB-connect or -reconnect

Post Reply
User avatar
ProCompSys
Posts: 3
Joined: Jan 19th, 2010, 9:43 am
Location: Germany
Contact:

System hangs after USB-connect or -reconnect

Post by ProCompSys »

Dear all,

we're using external USB-HDDs for backup-purposes on a Windows Server 2003 R2 with SP2 (32 bit).
It's a Dell Poweredge 1800 (Dell Server PE1800) with an Intex Xeon Dual Core with 3 GHz and 4 GB of RAM.

The USB-Controller is an "Intel 82801EB".
Besides that there are the following:
Adaptec SCSI Card 39160/3960D - Ultra160 SCSI
DELL CERC SATA 1.5/6ch RAID Controller
LSI Logic 1020/1030 Ultra320 SCSI Adapter

When the USB-HDD is exchanged in the morning, the complete System hangs and can only be resetted.
Mouse still works, NUM-Key responds (keyboard-LED), but the system does not react on mouse-clicks, CTRL-ALT-DEL or anything else.

After reboot we found two times Event-ID 57 in System-Log:
Event Type: Warning
Event Source: Ftdisk
Event Category: Disk
Event ID: 57
Description:
The system failed to flush data to the transaction log. Corruption may
occur.
...just within the last 4 minutes (07:31:18 and 07:35:33) before
Event ID: 6008
Source: Event Log
Type: Error
Description:
The previous system shutdown at Time on Date was unexpected.
The USR-Log shows:
19.01.2010 07:35:33:578 UpdateListTimerTimer
19.01.2010 07:35:34:078 Message UPDATE_USB_LIST received
19.01.2010 07:35:34:578 UpdateListTimerTimer
19.01.2010 07:36:15:875 UpdateListTimerTimer
19.01.2010 08:47:02:921 ****************************** Start ******************************
19.01.2010 08:47:03:078 USBSafelyRemove Version 4.1.5.800 / Windows Server 2003 R2 Standard Edition (5.2) Service Pack 2 (Build 3790) [Win32 on X86]
Trying to circumvent the system-failure, we tried to implement the scheduled use of the commandline "usr.exe stop -d P:" to disconnect the drive early in the morning before exchanging disks.
Just testing the commandline "usr.exe return -d P:" lead to next system-failure as described above.
We restarted the system within a minute or two and did not find any eventlog-entries after the system came back up... it probably just didn't have the time to do so.
Which leads to the assumption, that the freeze in the morning was not a complete stall, but an "extremely extreme" slowdown, as it obviously eventually found the time to log ID 57. In the last freeze after the command "return", we probably did not wait long enough to give the system the chance to do a log-entry.

The USR-Log shows:
19.01.2010 10:29:18:656 Message UPDATE_USB_LIST received
19.01.2010 10:29:21:234 UpdateListTimerTimer
19.01.2010 10:30:32:468 UpdateListTimerTimer
19.01.2010 10:30:36:656 UpdateListTimerTimer
19.01.2010 11:07:27:703 ****************************** Start ******************************
19.01.2010 11:07:27:812 USBSafelyRemove Version 4.1.5.800 / Windows Server 2003 R2 Standard Edition (5.2) Service Pack 2 (Build 3790) [Win32 on X86]

On http://support.microsoft.com/kb/912593/en-us/ we found a dim hint on the issue related to "HP SecurePath", but neither do we have a HP-System (ours is a Dell as stated above), nor do we run "Windows Server 2003 SP 1", but "Windows Server 2003 R2 SP2"

It remains to be checked, if the problem also exists without "USB Safely Remove" being installed.
As most of the time we can only access the sytem remotely, there wouldn't be a chance to get the disconnected drive hooked back to the system... a feature that makes USR worth its weight in gold! :shock:
And as this is a productive system, we are somewhat retentive for the moment trying to de- and re-install this and that and try to be lucky.
Therefore we'd appreciate your, input if you have any experience with this issue.
Post Reply