Wu's not turning in.

[Clutch]

I'm getting the following error message on my main computer. I have stopped and started the F@H services when I noticed it to see if it would fix it, but no fix.

[01:22:10] + Attempting to get work packet
[01:22:10] - Connecting to assignment server
[01:22:10] Couldn't send HTTP request to server (wininet)
[01:22:10] + Could not connect to Assignment Server
[01:22:11] Couldn't send HTTP request to server (wininet)
[01:22:11] + Could not connect to Assignment Server 2
[01:22:11] + Couldn't get work instructions.
[01:22:11] - Error: Attempt #90 to get work failed, and no other work to do.
Waiting before retry.

I sometimes get this on IC13 as well, but usually it will send in the WU after stopping and starting the service. My second computer seems to be working fine. I have 3 wu's in queue to be sent in on my main computer and they just will not send in.

[profdlp]

What server are they trying to connect to? :)

[Clutch]

[20:44:50] + Attempting to send results
[20:45:21] Couldn't send HTTP request to server (wininet)
[20:45:21] + Could not connect to Work Server (results)
[20:45:21] (171.65.103.160:8080)
[20:45:21] - Error: Could not transmit unit 06 (completed February 16) to work server.

Does that help?

[profdlp]

Stanford has a slew of servers down (http://fah-web.stanford.edu/serverstat.html) at the moment, but that doesn't appear to be one of them. It is listed as "Full" right now, meaning that you may have to wait a bit.

Hopefully they will go through soon. :) :fold:

[edcentric]

yes it is odd. One of my machines will backlog and the others not.
Keep folding, they will go sooner or later.
Sometimes I restart f@h to try and force the send, and sometimes it works.

[csimon]

If the server isn't down you can also try sending it manually. Make a shortcut to the client and add the "-send all" flag.

[SPIKE09]

Clutch the wininet message says the problem is your end, best guess you have been downgraded to IE7 and it has nadgered your settings, check and make sure it is set to use ie settings NO, use proxy NO.

[Clutch]

I'll check it when I get home. I have upgraded my main computer to IE7, but IC13 is still on 6, and it sometimes has the same problem.

[Clutch]

I have tried running the -config and -config only flag and setting the IE settings to NO but it does not save. I don't know what else to do. Even in the Client.cfg file I do not see where I can edit that out. I don't want to have to uninstall and reinstall the client and loose my three wu's that are just sitting there.

[profdlp]

List all of the steps you are taking when you try the -configonly tag. :)

[Clutch]

Okay this is weird. I get up this morning and do my morning EM3 check and sometime during the night my computer managed to get a new WU. But it shows that I still have 2 WU's in queue.

To tro the -configonly tag I went to a command prompt to the F@H dir. Then typed in FAH502-Console.exe -configonly I went through all the config screens again and change the IE settings to no each time. I'll see what it does after it complete this WU.

[EyesOnly]

I too have a core that hasn't returned it's wu but still it has got anotherone to work on. I'll do the changes as well and see what happens.

Edit
Ok the change has been made on both cores and i also tried to send a completed wu from core 2 but it didn't work. Interestly core 2 is on wu 8 even though it failed to send wu 6. 7 worked though. Hopefully it'll work when 8 is finished. I'll post back when i know more.

[csimon]

make sure to restart the client after you config

[EyesOnly]

I did shut down the service before using the flags. Anyway the wu has been sent now so i'm happy and also about 2500 points from 100k. :celebrate

[Clutch]

I just got mad enough to uninstall F@H, then reinstall it on my main machine and set IE settings to no from the start. Now on IC13, it is using IE6 and nothing has changed at all and I still get the same error. As soon as I start and stop the service it seems to send the WU just fine. I'm gonn see if anything changes on my main computer.

[QCH]

Weird problem, Clutch... :scratch:

[profdlp]

Have you checked your firewall settings?

[Sledgehammer70]

Hey guys I seem to be having an issue, and I can't figure it out as of yet :( I have over 30+ CPU's folding but they seem to be folding very slow, I have had them online for over a few weeks and have only had 13 report :( i check all the names and and CPU settings, no sure what else to do :(

[SPIKE09]

Hey guys I seem to be having an issue, and I can't figure it out as of yet :( I have over 30+ CPU's folding but they seem to be folding very slow, I have had them online for over a few weeks and have only had 13 report :( i check all the names and and CPU settings, no sure what else to do :( HMM are you able to check the fahlog and ensure SSE boost is being used, how is FAH configured start as service etc? do you have -forceasm flag in place in case someone kills the FAH console or service. It will use standard loops if this happens and that is really slow.

[Sledgehammer70]

I am the only one touching these PC's and they are set to run as a service. I have EMIII running showing all cores running, but they just haven't registered yet :( punks....

[SPIKE09]

Is the SSE boost showing though sledgehammer?

[profdlp]

Why not attach a couple of the logs from the slackers? :)

[Clutch]

Have you checked your firewall settings?

I don't use a Firewall, and I know IC13 doesn't have a firewall on it. My sisters machine turns WU's in just fine. And so did mine and IC13 before about maybe two weeks ago. And IC13 is running Windows 2000 and I'm running XP, so I don't think it is an OS problem.

[profdlp]

Attach a copy of the .cfg file from each computer.

Have you tried monkeying around with the port settings? (80 and 8080)

[danball1976]

My brother's computer is having the same problem too. I need to get a screenshot or post a copy of the cfg file here.

Also, I am attaching a screenshot from EMIII of the config of my machine

[danball1976]

It seems I have found the problem. On my computer, "Use Internet Explorer Settings" was set to no. On his, it was set to yes. I set it to no on his computer, and now it uploads the work units.

[QCH]

:thumbsup:

[edcentric]

ie = spawn of satan

[jmoney3457]

I am also having similar issue, I went in and noticed it was set to use IE=yes so I changed it to no still same results..I've c/p the portion of my log to show my error..[21:39:09] Working on Unit 02 [March 6 21:39:09]
[21:39:09] + Working ...
[21:39:09]
[21:39:09] *------------------------------*
[21:39:09] Folding@Home Gromacs Core
[21:39:09] Version 1.90 (March 8, 2006)
[21:39:09]
[21:39:09] Preparing to commence simulation
[21:39:09] - Looking at optimizations...
[21:39:09] - Files status OK
[21:39:09] - Expanded 291288 -> 1461493 (decompressed 501.7 percent)
[21:39:09]
[21:39:09] Project: 3039 (Run 7, Clone 403, Gen 12)
[21:39:09]
[21:39:09] Assembly optimizations on if available.
[21:39:09] Entering M.D.
[21:39:11] - Couldn't send HTTP request to server
[21:39:11] + Could not connect to Work Server (results)
[21:39:11] (171.65.103.106:8080)
[21:39:11] - Error: Could not transmit unit 06 (completed February 26) to work server.


[21:39:11] + Attempting to send results
[21:39:29] (Starting from checkpoint)
[21:39:29] Protein: p3039_supervillin-03
[21:39:29]
[21:39:29] Writing local files
[21:39:29] Completed 3820941 out of 5000000 steps (76)
[21:39:29] Extra SSE boost OK.
[21:39:45] - Couldn't send HTTP request to server
[21:39:45] + Could not connect to Work Server (results)
[21:39:45] (171.65.103.100:8080)
[21:39:45] Could not transmit unit 06 to Collection server; keeping in queue.
are the servers down or is it on my side? but I see others are having problems lately also..

[Kentigern]

According to the pande forum this particular (171.65.103.100:8080) keeps going to port 80, they are trying to find out what is causing the problem

[nonstop301]

I receive the exact message as jmoney3457 so hopefully they will fix the problem with that Work Server soon.

[jmoney3457]

I receive the exact message as jmoney3457 so hopefully they will fix the problem with that Work Server soon.
hopefully;) glad i'm not only 1 with this problem..at 1st I thought it was on my end..guess not:D

[jmoney3457]

any updates on this? as I still can't finish (send) this WU in for a new 1...

[csimon]

any updates on this? as I still can't finish (send) this WU in for a new 1...Have you tried alternating flags? Like if you have no flags then set -advmethods and if you have that then no flags.

[jmoney3457]

Have you tried alternating flags? Like if you have no flags then set -advmethods and if you have that then no flags.
no clue what you mean or how to do that..I thought others were getting same error I was..WU not turning in? like is that work server down or something?

[Kentigern]

Have posted on the Pande folding forum asking what's happening, as the server stats page is currently showing that they are accepting WU's. Previous posts for yesterday are showing that they are still working on the problem.
Will let you know when I get an answer, from them.

[nonstop301]

Have you tried alternating flags? Like if you have no flags then set -advmethods and if you have that then no flags.

I tried changing the advmethods flag csimon but it didn't make a difference

I have the same error as jmoney3457 and the work servers that can't be reached are 171.64.65.65:8080 and 171.65.103.100:8080 for the 2 work units that have been completed

The results have been queued and I have received new WUs so it isn't a big problem.

[csimon]

I tried changing the advmethods flag csimon but it didn't make a difference.
Sorry that was my misinterpretation ...that only helps when you have trouble getting more work usually.
Having you tried making a shortcut and adding the "-send all" flag?
When that doesn't work I just let it be til it goes ...for whatever reason they do seem to get stuck from time to time.
Did the wu finish on time?

[nonstop301]

Hi again csimon,

The WUs I receive are processed properly without delays.

I made the -send all shortcut as well but I still receive the message that it can't transmit the unit to the work server. It looks like it's a matter for server administrators to sort out, unless most people are able to send their results successfully to these particular servers when they complete a WU.

[jmoney3457]

I tried changing the advmethods flag csimon but it didn't make a difference

I have the same error as jmoney3457 and the work servers that can't be reached are 171.64.65.65:8080 and 171.65.103.100:8080 for the 2 work units that have been completed

The results have been queued and I have received new WUs so it isn't a big problem. so how do I get a new WU? with this problem (same as nonstop has) seems I can't receive new wu..

[SPIKE09]

You already have a new wu 76% finished, the client will now try to send the completed Wu every 6 hours or so, or if you restart the console it should retry immediately.


21:39:09] Entering M.D.
[21:39:11] - Couldn't send HTTP request to server
[21:39:11] + Could not connect to Work Server (results)
[21:39:11] (171.65.103.106:8080)
[21:39:11] - Error: Could not transmit unit 06 (completed February 26) to work server.


[21:39:11] + Attempting to send results
[21:39:29] (Starting from checkpoint)
[21:39:29] Protein: p3039_supervillin-03
[21:39:29]
[21:39:29] Writing local files
[21:39:29] Completed 3820941 out of 5000000 steps (76)
[21:39:29] Extra SSE boost OK.
[21:39:45] - Couldn't send HTTP request to server
[21:39:45] + Could not connect to Work Server (results)

[jmoney3457]

thx 4 the info spike..any update on that particular WU not turning in though with the pande forum? (server down)?

[SPIKE09]

Nothing more than Kentigern has reported, she has the last post in the thread at the FCF. Xuhuang the main sponsor using this server did state it had a high net load due to the number of large WU's sending to it though, why they have not updated the collection server with a list of Wus in the wild is beyond me.

[jmoney3457]

what exactly is the pande forums? lol:p

[SPIKE09]

not really the pande forums they are an independent folding support forum associated with the pande group.

http://forum.folding-community.org/forums.html

[Kentigern]

Ok folks, this just in from the Pande Forum.

Let's start with verifying connectivity to those servers.

What message do you (they) see when typed into the address bar of IE?

http://171.65.103.106:8080
http://171.65.103.100:8080

You should see the word OK

[nonstop301]

Hi Kentigern,

I see OK when I check those addresses with my browser but the results haven't been sent yet.

For both servers 171.65.103.100:8080 and 171.64.65.65:8080 the FAH console tells me it can't connect to the Work Server

[Kentigern]

Update from Pande folks:)

Anyone else's logs we can check for a similar problem, other than money's

Please post the relevant parts of your fah logs, so they can do some checking :)

[nonstop301]

I have attached a piece of the FAH log below :)

22762

[jmoney3457]

hi kenti, any update/fix on that particular work server?

[Kentigern]

:wave: Hi Folks

here is a link to the folding forum thread http://forum.folding-community.org/viewtopic.php?p=170971#170971

As we are on different time zones you might see any updates before I can see them and post them on to you all. :)

[nonstop301]

Hi Kentigern,

Thanks for that link to the related thread on the pande forum.

I still have 2 WUs queued to be sent to the problematic servers but I completed a third WU last night and its results were sent successfully since it was associated with a different server.

[Kentigern]

Hi nonstop
We probably won't get any further news until around 11pm GMT due to time zones, apparently though the various servers do attempt to connect every so often I think about 6 hourly, to get any queued WU's.
Let's hope yours get picked up soon :)

[jmoney3457]

thanks for the updates kenti;)

[nonstop301]

I don't know whether the results that are queued will ever be sent :)

I completed another WU last night and it's results were sent off without problems but it seems the queued results are ignored.

Is there something else I have to do indicate to the server that I have results for it ? It seems to me that the queue function isn't working properly and the server isn't picking up data that is queued up behind the current assignment

[Kentigern]

Hi Nonstop,
Can you post these parts of the details from your FAHlog, for each of the work units in the queue. Thanks :)

Please keep each work unit details seperately - see below

This is unit one

User name: Kentigern (Team 93)
User ID: 31A213BD175D90CA
Machine ID: 1


Preparing to get new work unit...
[07:13:13] + Attempting to get work packet
[07:13:13] - Connecting to assignment server
[07:13:13] - Successful: assigned to (171.64.65.58).
[07:13:13] + News From Folding@Home: Welcome to Folding@Home
[07:13:14] Loaded queue successfully.
[07:13:21] + Closed connections
[07:13:21]
[07:13:21] + Processing work unit
[07:13:21] Core required: FahCore_78.exe
[07:13:21] Core found.
[07:13:21] Working on Unit 06 [March 12 07:13:21]
[07:13:21] + Working ...
[07:13:21]
[07:13:21] *------------------------------*
[07:13:21] Folding@Home Gromacs Core
[07:13:21] Version 1.90 (March 8, 2006)
[07:13:21]
[07:13:21] Preparing to commence simulation
[07:13:21] - Looking at optimizations...
[07:13:21] - Created dyn
[07:13:21] - Files status OK
[07:13:22] - Expanded 292209 -> 1461493 (decompressed 500.1 percent)
[07:13:22] - Starting from initial work packet
[07:13:22]
[07:13:22] Project: 3041 (Run 9, Clone 390, Gen 7)
[07:13:22]
[07:13:22] Assembly optimizations on if available.
[07:13:22] Entering M.D.
[07:13:28] Protein: p3041_supervillin-03


[07:12:27] + Attempting to send results
[07:13:09] + Results successfully sent
[07:13:09] Thank you for your contribution to Folding@Home.
[07:13:09] + Number of Units Completed: 42

this is unit 2 etc etc etc

[nonstop301]

I'm attaching the logs for each item in the queue

The first set results are waiting to be received by the 171.64.65.65:8080 server

The second set of resutls are waiting to be received by the 171.65.103.100:8080 server

Seperating out the queued items in this way made me notice that both items in the queue have been given the same unit number (unit 08) with one of them also having the completion date added to it ie. unit 08 (completed March 4) whereas the second one is just called unit 08

Furthermore, the unit 08 (completed March 4) is attempting a transmission to the work server whereas the second unit 08 is attempting a transmission to the Collection server.

Maybe the fact that they are both called unit 08 is causing the problem


22812

22813

[Kentigern]

:wave: Thanks - will let the guys over at pande know hopefully thay can sort this out for you. :)

[nonstop301]

I think I'll just have to give up on those queued results since it's been more than a week that they've been queued but they aren't been accepted.

I posted on the pande forum about it as well but there hasn't been any feedpack so it seems that nothing can be done about it.

Fortunately the queued items aren't interfering with the sending of results from current WUs that are completed so it isn't much of a setback at the end of the day.