Roscoe62
09-11-2006, 02:26 PM
I'm asking this on the offchance someone here knows a bit about diagnosing S3 resumption issues.
I am setting up my Media Server to awake from an S3 suspend state to serve up media files. The m/board I'm using and the BIOS options I've selected can be found in this thread...
http://www.avsforum.com/avs-vb/showthread.php?t=714738
The graphics card I have onboard is a nVidia FX5200. Nothing special as it's only a server so I don't need anything flash.
I'm using a Promise SATA300 TX4 as a SATA drive controller but currently I've just got one media drive attached. I am still testing it but, at the moment, 9 times out of 10 the server will fire up and serve files correctly. However, yesterday I got a weird response. The way I've got it working at the moment is to send it a magic packet, wait 10 seconds for it to fire up before asking it for my chosen media. On the one time it didn't work the server powered up, but the monitor I have attached to it did not fire up (it usually does) and any attempt to contact the server using VNC from another machine results in the request to logon to timeout. When this occurred, no amount of mouse or keyboard activity could bring the machine to life. In the end I had to reboot. It was odd because I had performed the same test successfully only an hour earlier. Previous tests have been successfully performed with 2 days between the server entering S3 and it firing up again.
Now, 9 out of 10 isn't really reliable enough for a media server. The problem is that I have NO idea where to start looking into what might be causing the problem - and the fact that it's only an intermittent fault makes it that much harder to diagnose!
Also, last night it faulted again. This time the monitor fired up but it just displayed a blank screen. About 20-30 seconds later the system fired up properly but, by that time, the file request had already timed out and then, as a final touch it refused to drop into S3 again. (I currently have it set up to drop into S3 after 10 minutes of inactivity - just during this testing phase).
Any takers? This one has me stumped!
I am setting up my Media Server to awake from an S3 suspend state to serve up media files. The m/board I'm using and the BIOS options I've selected can be found in this thread...
http://www.avsforum.com/avs-vb/showthread.php?t=714738
The graphics card I have onboard is a nVidia FX5200. Nothing special as it's only a server so I don't need anything flash.
I'm using a Promise SATA300 TX4 as a SATA drive controller but currently I've just got one media drive attached. I am still testing it but, at the moment, 9 times out of 10 the server will fire up and serve files correctly. However, yesterday I got a weird response. The way I've got it working at the moment is to send it a magic packet, wait 10 seconds for it to fire up before asking it for my chosen media. On the one time it didn't work the server powered up, but the monitor I have attached to it did not fire up (it usually does) and any attempt to contact the server using VNC from another machine results in the request to logon to timeout. When this occurred, no amount of mouse or keyboard activity could bring the machine to life. In the end I had to reboot. It was odd because I had performed the same test successfully only an hour earlier. Previous tests have been successfully performed with 2 days between the server entering S3 and it firing up again.
Now, 9 out of 10 isn't really reliable enough for a media server. The problem is that I have NO idea where to start looking into what might be causing the problem - and the fact that it's only an intermittent fault makes it that much harder to diagnose!
Also, last night it faulted again. This time the monitor fired up but it just displayed a blank screen. About 20-30 seconds later the system fired up properly but, by that time, the file request had already timed out and then, as a final touch it refused to drop into S3 again. (I currently have it set up to drop into S3 after 10 minutes of inactivity - just during this testing phase).
Any takers? This one has me stumped!