171.67.108.26 - not accepting WUs? [True. CS offline]

[Joe_H]

If you Right-click on a folding slot in FAHControl, then you will get the options to Finish, Start or Pause the slot.

As for the Code tags, if you use the Full Editor option to post here in the forum, they can be selected from the options at the top of the posting window. Then you can paste your log in between them.

[Mancunian_Nick]

Hello Joe. Nice to talk to you. Thank you for the above. Your comments about the code tags are noted. I've not been using the full editor but if I have log entries to post in future, I'll make sure I do.

As for the right clicking on the folding slots, I'm sorry to say nothing happens except it brings up details in the right hand pane of FAHControl - selected work unit but no other options that I can see. It's frustrating. Forgive me for suggesting anything untoward and I don't mean this disrespecfully but I notice from your 'sig' that you're using a Mac. Is the software for the Mac different to Windows perchance? I'm only asking because maybe that's why? Although I have a feeling I might be insulting your intelligence here somewhat so apologies for that. Surely you would already have taken that into consideration before advising me.

I'll check back later. Will get this ruddy well sorted eventually if it kills me - just hope it doesn't do you lot in at the same time. Bruce and Panther must have been first in the queue when patience was handed out!

[P5-133XL]

Make sure you are right-clicking in the Folding Slot window. no where else will right-clicking do anything significant. The folding Slot Window is directly above the Work Queue Window; To the left of the selected Work Unit area; Right of the clients window; and below your identity. Just as a sanity check -- you are not right-clicking the folding slot tab from inside configure because that will not work either.

Yes, this feature works in Windows. At least on my Windows machines.

[Mancunian_Nick]

Hi P5. Thank you. Yes I see what you mean. I was clicking in the wrong place before, the Work Queue window as you said. However there are only 2 slots in the folding slot window - the gpu (paused: waiting for idle) and the 01 slot currently working on cpu: 1 normally. Below in the work queue window I now only have 2 slots - 02 running 8083 normally and correctly plus the faulty/expired send one in 01 I've been discussing with bruce and Panther. I think somehow that faulty send is going to have be deleted - either that or just left to remain in stasis and let the rest take care of itself. If I want to dispose of the gpu slot, presumably I just click on it and select delete? Or does it need to be removed from the configuration slots tab and remove it from there? Or possibly even both places? As I've said before, I'm not deleting anything or changing anything unless I'm advised/told to do so. Last thing I want to do is screw the thing up any further than it is already.

Thanks again anyway. Appreciate it.
Nick

[PantherX]

...I've just realised I'm using the FAH Control - Client Control for Expert Users application, not the web viewer. It's the one I downloaded. Did I download the wrong one?...

You downloaded the correct package since it contains both interfaces, WebControl and FAHControl (Advanced Control). You can access either interfaces by right-clicking the F@H icon in the notification bar and selecting the appropriate interface to use. In your current case, using FAHControl is advised since we are dealing with an individual slot, something that is considered as an advanced feature.

...I think somehow that faulty send is going to have be deleted - either that or just left to remain in stasis and let the rest take care of itself. ...

You log shows only the configuration so could you please post some more after that so we can see if there are any WUs that are queued up or are being processed?

...Is the software for the Mac different to Windows perchance?...

One of the main goal of V7 was to have a unified interface across all supported operating systems (Windows, Linux, OSX). So far, we are pretty successful in it. There are only some minor differences between them, the differences are not interface related, rather, they are functional ones. For example, GPU folding is only supported on Windows for now. Linux (Nvidia GPUs only) native folding is being beta tested now (AMD will be once the driver is mature enough) and there is no support for OSX due to lack of adequate OpenCL support.

...If I want to dispose of the gpu slot, presumably I just click on it and select delete? Or does it need to be removed from the configuration slots tab and remove it from there? Or possibly even both places?...

There isn't any right-click to delete the slot feature since it is not really needed (removal of slots is not very common amongst donors). You still have the GPU Slot (in the last log file posted) so once you set it to finish and the WU is successfully sent, you can easily remove it by:
FAHControl -> Configure -> Slots Tab -> Select GPU -> Click Remove Button -> Click Save Button

[Mancunian_Nick]

Hi Panther. Thank you again for your reply. Found it most interesting and enlightening too. I have just restarted the client after having it off for a while so I could watch some streaming video and converting some video files and needed the extra RAM so in a moment, I'll post what's come up since the restart so you can see. Once again this mention of "once you set it to finish" but I've been unable to find that feature anywhere. Please can you advise me where that is like you did above with FAHControl->Configure etc then I'll remove the GPU slot as under normal use, it's always (well for the most part always although at the moment the computer is on unused more than it normally is) idle.

Thanks again Panther. Now I'll go into full editor mode and post the rest of the log since my restart after the config part I've already posted and hope it tells you all you need to know. If not let me know and I'll send anything else you need but hopefully it will do.

[Mancunian_Nick]

Ok panther, here's the rest of the log you asked me for just now -

Code: Select all

14:38:02:Trying to access database...
14:38:02:Successfully acquired database lock
14:38:02:Enabled folding slot 00: PAUSED gpu:0:G94 [GeForce 9600 GT] (waiting for idle)
14:38:02:Enabled folding slot 01: READY cpu:1
14:38:02:WU01:FS00:Sending unit results: id:01 state:SEND error:DUMPED project:5772 run:1 clone:277 gen:3018 core:0x11 unit:0x48874a2e518cdb100bca01150001168c
14:38:04:WU01:FS00:Uploading 3.77KiB to 171.67.108.11
14:38:04:WU01:FS00:Connecting to 171.67.108.11:8080
14:38:04:WU02:FS01:Starting
14:38:04:WU02:FS01:Running FahCore: "C:\Program Files (x86)\FAHClient/FAHCoreWrapper.exe" C:/Users/Nick/AppData/Roaming/FAHClient/cores/www.stanford.edu/~pande/Win32/AMD64/Core_a4.fah/FahCore_a4.exe -dir 02 -suffix 01 -version 703 -lifeline 143572 -checkpoint 15
14:38:04:WU02:FS01:Started FahCore on PID 145928
14:38:05:WU02:FS01:Core PID:125332
14:38:05:WU02:FS01:FahCore 0xa4 started
14:38:05:WU02:FS01:0xa4:
14:38:05:WU02:FS01:0xa4:*------------------------------*
14:38:05:WU02:FS01:0xa4:Folding@Home Gromacs GB Core
14:38:05:WU02:FS01:0xa4:Version 2.27 (Dec. 15, 2010)
14:38:05:WU02:FS01:0xa4:
14:38:05:WU02:FS01:0xa4:Preparing to commence simulation
14:38:05:WU02:FS01:0xa4:- Looking at optimizations...
14:38:05:WU02:FS01:0xa4:- Files status OK
14:38:05:WU02:FS01:0xa4:- Expanded 1083737 -> 3055124 (decompressed 281.9 percent)
14:38:05:WU02:FS01:0xa4:Called DecompressByteArray: compressed_data_size=1083737 data_size=3055124, decompressed_data_size=3055124 diff=0
14:38:06:WU02:FS01:0xa4:- Digital signature verified
14:38:06:WU02:FS01:0xa4:
14:38:06:WU02:FS01:0xa4:Project: 8083 (Run 60, Clone 52, Gen 20)
14:38:06:WU02:FS01:0xa4:
14:38:06:WU02:FS01:0xa4:Assembly optimizations on if available.
14:38:06:WU02:FS01:0xa4:Entering M.D.
14:38:06:WARNING:WU01:FS00:Exception: Failed to send results to work server: Failed to read stream
14:38:06:WU01:FS00:Trying to send results to collection server
14:38:06:WU01:FS00:Uploading 3.77KiB to 171.67.108.25
14:38:06:WU01:FS00:Connecting to 171.67.108.25:8080
14:38:12:WU02:FS01:0xa4:Using Gromacs checkpoints
14:38:12:WU02:FS01:0xa4:Mapping NT from 1 to 1 
14:38:12:WU02:FS01:0xa4:Resuming from checkpoint
14:38:12:WU02:FS01:0xa4:Verified 02/wudata_01.log
14:38:12:WU02:FS01:0xa4:Verified 02/wudata_01.trr
14:38:12:WU02:FS01:0xa4:Verified 02/wudata_01.xtc
14:38:12:WU02:FS01:0xa4:Verified 02/wudata_01.edr
14:38:13:WU02:FS01:0xa4:Completed 120140 out of 500000 steps  (24%)
14:38:27:WARNING:WU01:FS00:WorkServer connection failed on port 8080 trying 80
14:38:27:WU01:FS00:Connecting to 171.67.108.25:80
14:38:48:ERROR:WU01:FS00:Exception: Failed to connect to 171.67.108.25:80: A connection attempt failed because the connected party did not properly respond after a period of time, or established connection failed because connected host has failed to respond.
14:39:02:WU01:FS00:Sending unit results: id:01 state:SEND error:DUMPED project:5772 run:1 clone:277 gen:3018 core:0x11 unit:0x48874a2e518cdb100bca01150001168c
14:39:02:WU01:FS00:Uploading 3.77KiB to 171.67.108.11
14:39:02:WU01:FS00:Connecting to 171.67.108.11:8080
14:39:03:WARNING:WU01:FS00:Exception: Failed to send results to work server: Failed to read stream
14:39:03:WU01:FS00:Trying to send results to collection server
14:39:03:WU01:FS00:Uploading 3.77KiB to 171.67.108.25
14:39:03:WU01:FS00:Connecting to 171.67.108.25:8080
14:39:24:WARNING:WU01:FS00:WorkServer connection failed on port 8080 trying 80
14:39:24:WU01:FS00:Connecting to 171.67.108.25:80
14:39:46:ERROR:WU01:FS00:Exception: Failed to connect to 171.67.108.25:80: A connection attempt failed because the connected party did not properly respond after a period of time, or established connection failed because connected host has failed to respond.
14:40:02:WU01:FS00:Sending unit results: id:01 state:SEND error:DUMPED project:5772 run:1 clone:277 gen:3018 core:0x11 unit:0x48874a2e518cdb100bca01150001168c
14:40:02:WU01:FS00:Uploading 3.77KiB to 171.67.108.11
14:40:02:WU01:FS00:Connecting to 171.67.108.11:8080
14:40:03:WARNING:WU01:FS00:Exception: Failed to send results to work server: Failed to read stream
14:40:03:WU01:FS00:Trying to send results to collection server
14:40:03:WU01:FS00:Uploading 3.77KiB to 171.67.108.25
14:40:03:WU01:FS00:Connecting to 171.67.108.25:8080
14:40:24:WARNING:WU01:FS00:WorkServer connection failed on port 8080 trying 80
14:40:24:WU01:FS00:Connecting to 171.67.108.25:80
14:40:45:ERROR:WU01:FS00:Exception: Failed to connect to 171.67.108.25:80: A connection attempt failed because the connected party did not properly respond after a period of time, or established connection failed because connected host has failed to respond.
14:41:39:WU01:FS00:Sending unit results: id:01 state:SEND error:DUMPED project:5772 run:1 clone:277 gen:3018 core:0x11 unit:0x48874a2e518cdb100bca01150001168c
14:41:40:WU01:FS00:Uploading 3.77KiB to 171.67.108.11
14:41:40:WU01:FS00:Connecting to 171.67.108.11:8080
14:41:40:WARNING:WU01:FS00:Exception: Failed to send results to work server: Failed to read stream
14:41:40:WU01:FS00:Trying to send results to collection server
14:41:40:WU01:FS00:Uploading 3.77KiB to 171.67.108.25
14:41:40:WU01:FS00:Connecting to 171.67.108.25:8080
14:42:01:WARNING:WU01:FS00:WorkServer connection failed on port 8080 trying 80
14:42:01:WU01:FS00:Connecting to 171.67.108.25:80
14:42:22:ERROR:WU01:FS00:Exception: Failed to connect to 171.67.108.25:80: A connection attempt failed because the connected party did not properly respond after a period of time, or established connection failed because connected host has failed to respond.
14:44:17:WU01:FS00:Sending unit results: id:01 state:SEND error:DUMPED project:5772 run:1 clone:277 gen:3018 core:0x11 unit:0x48874a2e518cdb100bca01150001168c
14:44:17:WU01:FS00:Uploading 3.77KiB to 171.67.108.11
14:44:17:WU01:FS00:Connecting to 171.67.108.11:8080
14:44:17:WARNING:WU01:FS00:Exception: Failed to send results to work server: Failed to read stream
14:44:17:WU01:FS00:Trying to send results to collection server
14:44:17:WU01:FS00:Uploading 3.77KiB to 171.67.108.25
14:44:17:WU01:FS00:Connecting to 171.67.108.25:8080
14:44:38:WARNING:WU01:FS00:WorkServer connection failed on port 8080 trying 80
14:44:38:WU01:FS00:Connecting to 171.67.108.25:80
14:44:59:ERROR:WU01:FS00:Exception: Failed to connect to 171.67.108.25:80: A connection attempt failed because the connected party did not properly respond after a period of time, or established connection failed because connected host has failed to respond.
14:48:31:WU01:FS00:Sending unit results: id:01 state:SEND error:DUMPED project:5772 run:1 clone:277 gen:3018 core:0x11 unit:0x48874a2e518cdb100bca01150001168c
14:48:31:WU01:FS00:Uploading 3.77KiB to 171.67.108.11
14:48:31:WU01:FS00:Connecting to 171.67.108.11:8080
14:48:32:WARNING:WU01:FS00:Exception: Failed to send results to work server: Failed to read stream
14:48:32:WU01:FS00:Trying to send results to collection server
14:48:32:WU01:FS00:Uploading 3.77KiB to 171.67.108.25
14:48:32:WU01:FS00:Connecting to 171.67.108.25:8080
14:48:53:WARNING:WU01:FS00:WorkServer connection failed on port 8080 trying 80
14:48:53:WU01:FS00:Connecting to 171.67.108.25:80

[bruce]

The GeForce 9600 GT is about 5 years old now, which in terms of GPUs, means is in late-middle-age. While it still has some life left, it's pretty slow compared to current generation GPUs. You'll have to decide when it can be on-line long enouigh to finish an assignment by the timeout and when it can't. Returning a WU before the timeout should always be the goal. It can be returned between the timeout (Preferred Deadline) and the Final Deadline but it's much less useful because it probably will be duplicated by someone else.

You can use either FAHControl (expert) or WebControl (much simpler) depending on what you need to do with it.

Right-click on the GPU slot and choose FINISH. Let the GPU continue until is says it's Paused-finished (Yellow) not Pause-waiting... . Then in Configure Slots gpu Remove. If you save, a new slot will be created and you'll need to manage the completion dates. If you choose to discontinue GPU folding, go to Configure Expert Extra_client_options and Add gpus =false and a new slot will not be added automatically.

[PantherX]

Have you updated to the latest Nvidia Driver? The reason is that I didn't experience any screen lag with GTX 260 and 9500M GS when folding the FahCore_11 WUs. Thus, if you are not experiencing any screen lag, you could fold on the GPU even when using your system. If you notice screen lag in a particular application like internet browsers, you can reduce/eliminate it by disabling the option to use the GPU.

[bruce]

RIGHT:

If the screen-lag problem has been solved by drivers or it's small enough that you don't care, then move the slider to "FULL" and let the GPU fold while you're using the computer.

[Mancunian_Nick]

Thanks bruce, I've done the finish thing on the gpu slot so I'll wait for it to say finished and then do as you say, although it's still idle at the moment despite the fact I've had it running most of the evening, certainly since 4pm without me being on the computer and it's now almost 2am the next day. And you're right about the 9600 GT being old. It was ahead of it's time probably when I bought it new but it was a long time ago and yes it needs updating but can't afford to at the moment. Probably birthday (November) or Christmas, we'll see.

Panther if not THE latest, pretty much so, yes. You say I could fold on the gpu even when I'm using the computer but as I've said before, that never happens because it's always paused: waiting for idle. As you can see, bruce has explained a way (by using FULL) to 'force' it to fold even when the computer is in use so I'll try it and see how it affects performance, although the cpu seems to run without any problems at all, even when the computer's in use. I'm happy to try it before I disable it completely. I'll try it tomorrow as I'll be going to bed and hibernating the system in about an hour but I'm in all day tomorrow (well today technically speaking ) so I'll see how it goes.

[Mancunian_Nick]

Hi guys. Quick update - after trying using FULL and seemingly working ok, I then started getting a couple of other 'problems' in the work queue despite the fact the closing date of the WU was 19th May but from what you've told me, this can happen quite normally if others have got theirs in quicker/earlier and that particuar protein/WU has been completed. Consequently I did as was suggested to me and set the GPU to finish and once it said finished, I've deleted that and am now only running on CPU and so far no further problem in that direction, if you know what I mean. However I now have a number of failed WUs in my work queue namely slot 00 5771 (3, 141, 3507); 01 5772 (1, 277, 3018) and 03 (0, 279, 4400).

They don't appear to be doing any harm or causing any problem from what I can see other than continually (not in the literal sense of course) trying to connect and send WUs that are never going to be accepted. Shall/can I delete them? If so how can I do that safely without affecting the WU currently being folded on my cpu slot which is running without problem please?

Thank you as always for any help/advice that can be offered. I have taken a look round the other topics to see if I could find an answer before posting further here but unfortunately wasn't able to find anything of significance.

Cheers, Nick

[PantherX]

I am assuming that you will now not be folding on your GPU, right? In that case, you have to options to remove the GPU WUs:

1) Make note of the ID under Work Queue in FAHControl (AKA Advanced Control) for the GPU WUs that you want to delete
2) Exit the F@H Client
3) Navigate to the data folder, C:\Users\Nick\AppData\Roaming\FAHClient\work
4) Delete the appropriate number folder
5) Restart F@H Client and you will notice that those WUs are no longer there.

Alternatively, what you could do is to set your CPU Slot to finish and once done. do a full uninstall (select the option to remove the data) and then do a fresh installation and configure the F@H Client to your liking.

Normally, whenever I remove a slot, any WU associated with it also get automatically dumped. It seems that that isn't the case for completed WUs.

[bruce]

- after trying using FULL and seemingly working ok, I then started getting a couple of other 'problems' in the work queue despite the fact the closing date of the WU was 19th May but from what you've told me, this can happen quite normally if others have got theirs in quicker/earlier and that particular protein/WU has been completed.

That's not what I said and it's not what I understand to be true.

1) WU is issued to person A.
2) Preferred deadline passes and WU is (most likely) lost, so the server issues the same WU to person B.
3) Person B completes his assignment before his preferred deadline expires and he gets a bonus.
4) Person A completes his assignment before his final deadline and he gets no bonus but he does get baseline points.

The work only needed to be done once, so the duplicate effort was wasted, but both people get credit.

An alternate scenario might be
1) WU is issued to person A.
2) Preferred deadline passes and WU is (most likely) lost, so the server issues the same WU to person B.
3) Person A completes his assignment before his final deadline and he gets no bonus but he does get baseline points.
4) Person B completes his assignment before his preferred OR AFTER the Preferred deadline expires and before his Final Deadline and gets a bonus or baseline points

The result was completed earlier (by A than by B) so there's a scientific benefit of the duplicated work because the schedule was improved.

[Mancunian_Nick]

Panther, thank you very much. Followed your instructions and all's good. And yes you assumed rightly, I'm not now folding on GPU.

Bruce thank YOU too. I can understand what you're saying now and apologies for not having done so before. My mistake.

Ok all's good now. I'll leave you guys in peace - at least for the moment or if/until something else crops up.