[danvvc]

im having problems sending some units off.
i've been through the list of websites and can see them all.
(the ones where it shows you the 'it works' screen)
the clients can download new units,

11:47:55] - Ask before connecting: No
[11:47:55] - Proxy: 10.124.4.30:80
[11:47:55] - User name: danvvc (Team 32)
[11:47:55] - User ID = 2D440792A9A60D3
[11:47:55] - Machine ID: 1
[11:47:55]
[11:47:55] Loaded queue successfully.
[11:47:55] + Benchmarking ...
[11:47:57] The benchmark result is 3596
[11:47:57]
[11:47:57] - Autosending finished units...
[11:47:57] Trying to send all finished work units


[11:47:57] + Attempting to send results
[11:47:57] - Reading file work/wuresults_01.dat from core
[11:47:57] (Read 2097152 bytes from disk)
[11:47:57] + Processing work unit
[11:47:57] Core required: FahCore_79.exe
[11:47:57] Core found.
[11:47:57] Working on Unit 04 [November 16 11:47:57]
[11:47:57] Connecting to http://171.65.103.160:8080/
[11:47:57] + Working ...
[11:47:57] - Calling 'FahCore_79.exe -dir work/ -suffix 04 -service -forceasm -lifeline 2700
'

[11:48:03]
[11:48:03] *------------------------------*
[11:48:03] Folding@Home Double Gromacs Core
[11:48:03] Version 1.91 (April 11, 2006)
[11:48:03]
[11:48:03] Preparing to commence simulation
[11:48:03] - Assembly optimizations manually forced on.
[11:48:03] - Not checking prior termination.
[11:48:05] Initial: 0000 + Could not connect to Work Server (results)
[11:48:05] - Error: Could not transmit unit 01 (completed November 11). Keeping unit in queue.
[11:48:05] + Sent 0 completed units to the server
[11:48:05] - Autosend completed

any ideas anybody.

[Midnight Dream]

Got a firewall?

[danvvc]

yeah all the machines are behind Birmingham City Councils Firewall (which i dont administer), they were all up/downloading earlier in the week, but yesterday and today the clients just seem to be queing up wu's..

[ChasR]

My guess is a proxy port issue. The proxy is open on port 80 but the attempt to transmit to the work server was on port 8080. If you can open the proxy on port 8080, TCP and UDP, it should work. After a while, no more than 6 hours, the client will attempt to send the unit to the collection server which should accept the WU. What happens when you enter http://171.65.103.160:8080/ in the address bar of your browser and click go? If you get an OK, I'm wrong.
Edit:
Looks like you won't be changing the proxy settings. How long have they been queued?

[danvvc]

yeah i can see the 'ok' screen.
since tuesday according to the logs?

Arguments: -local -service -verbosity 9 -forceasm

[13:03:57] - Ask before connecting: No
[13:03:57] - Proxy: 10.124.4.30:80
[13:03:57] - User name: danvvc (Team 32)
[13:03:57] - User ID = 7F981D146C31D4AD
[13:03:57] - Machine ID: 1
[13:03:57]
[13:03:57] Loaded queue successfully.
[13:03:57] + Benchmarking ...
[13:04:01] The benchmark result is 3492
[13:04:01]
[13:04:01] - Autosending finished units...
[13:04:01] Trying to send all finished work units


[13:04:01] + Attempting to send results
[13:04:01] - Reading file work/wuresults_05.dat from core
[13:04:01] + Processing work unit
[13:04:01] Core required: FahCore_78.exe
[13:04:01] Core found.
[13:04:02] Working on Unit 09 [November 15 13:04:02]
[13:04:02] + Working ...
[13:04:02] - Calling 'FahCore_78.exe -dir work/ -suffix 09 -service -forceasm -lifeline 1600
'

[13:04:02] (Read 2097152 bytes from disk)
[13:04:02] Connecting to http://171.65.103.160:8080/
[13:04:07]
[13:04:07] *------------------------------*
[13:04:07] Folding@Home Gromacs Core
[13:04:07] Version 1.90 (March 8, 2006)
[13:04:07]
[13:04:07] Preparing to commence simulation
[13:04:07] - Assembly optimizations manually forced on.
[13:04:07] - Not checking prior termination.
[13:04:08] - Expanded 78888 -> 420497 (decompressed 533.0 percent)
[13:04:11]
[13:04:11] Project: 2125 (Run 301, Clone 66, Gen 1)
[13:04:11]
[13:04:11] Assembly optimizations on if available.
[13:04:11] Entering M.D.
[13:04:13] Initial: 0000 + Could not connect to Work Server (results)
[13:04:13] - Error: Could not transmit unit 05 (completed November 12). Keeping unit in queue.
[13:04:13] + Sent 0 completed units to the server[13:04:13] - Autosend completed

should i leave it a bit longer, the internet does seem to be playing up this morning, i've had four people in my office since writing this..
"it doesnt work, blah, blah,"..
i think i'll keep monitoring them for a while and see what happens..

if i stop the service and restart it, will it autosend straight away? or when its writing the local files?
also, when you stop the service, should it post a msg in the log or not? i forget which i read?

[ChasR]

Restarting the client will lead to an immediate attempt to send in the WU. If they've been queued since Tuesday, the client should have made several attempts to reach the collection server, http://171.65.103.100. I assume you get an OK contacting it with your browser.
Firewalls, privacy guards, or proxys could disallow the upload of binary content, but allow harmless GET requests. That would explain connectivity with the browser but the inability to transmit the WU. You could try configuring to use IE settings and see if that works or being sure the proxy port is set to 80 instead of 8080. Perhaps someone has tightened firewall security?

When windows is shutdown, SIGTERM is written in the log for service installs. When you close with Net stop, sc stop, or services.msc nothing is written in the log.