Folding Forum

Posted: **Mon Dec 02, 2013 10:47 am**

Same WU folds on two servers simultaneously! I did not think this was possible?

Posted: **Mon Dec 02, 2013 1:29 pm**

Rare, but very possible. There are several examples in the forum. See those for more detail.

Posted: **Mon Dec 02, 2013 2:06 pm**

This can also happen in v6 if a new installation was created by copying the configuration from another setup. Whenever that's done with v6 you must delete the machineindependent.dat file on the new setup, otherwise the servers will see both clients as the same and when one requests a new work unit the servers assume something went wrong and download the same work unit that is running on the copy.

viewtopic.php?f=44&t=11175

Posted: **Mon Dec 02, 2013 2:26 pm**

The G34 6272 server was set to fold back in early 2012, and the E5-4650 server only 3 months ago, so that is not the case I believe. 2 months ago I reinstalled the first one to fold from a ramdisk, where I use a script made by one of the gurus at hardforum. The only thing they have in common is Ubuntu 12.04, installed from a DVD made from http://www.ubuntu.com/download/desktop. I also remember having experienced this on several occasions in the past, especially back when I were doing GPU folding.

Posted: **Mon Dec 02, 2013 4:13 pm**

7im wrote:Rare, but very possible. There are several examples in the forum. See those for more detail.

As was explained in those other examples, it can also happen when somebody else worked on the the project and it was dumped or the processing was aborted. You have no control over that (and most likely will get credit if you finish both of them).

The first one has been completed:
Hi -alias- (team 37651),
Your WU (P8105 R0 C11 G193) was added to the stats database on 2013-12-02 03:14:07 for 443304 points of credit.

Posted: **Mon Dec 02, 2013 4:33 pm**

Thanks bruce!
Then I think the reason is that the project has been dumped or the processing was aborted for some reason before then, as you type!

Posted: **Mon Dec 02, 2013 7:46 pm**

Just for my own learning: would be the unique unit identifier in both case the same ? I would hope they are different; just to confirm. OP, could you please share the lines from the log with the "received unit" information ? Like

Code: Select all

15:21:53:WU00:FS00:Received Unit: id:00 state:DOWNLOAD error:NO_ERROR project:7516 run:0 clone:346 gen:29 core:0xa3 unit:0x0000013efbcb017q5050ab7d8dac1721

Thanks in advance

Posted: **Mon Dec 02, 2013 10:43 pm**

My logs is from V6.34, different from yours that are from V7 I guess!
The first one:

Code: Select all

[22:56:47] + Attempting to get work packet
[22:56:47] Passkey found
[22:56:47] - Will indicate memory of 64396 MB
[22:56:47] - Connecting to assignment server
[22:56:47] Connecting to http://assign.stanford.edu:8080/
[22:56:48] Posted data.
[22:56:48] Initial: 8F80; - Successful: assigned to (128.143.231.201).
[22:56:48] + News From Folding@Home: Welcome to Folding@Home
[22:56:48] Loaded queue successfully.
[22:56:48] Sent data
[22:56:48] Connecting to http://128.143.231.201:8080/
[22:56:54] Posted data.
[22:56:54] Initial: 0000; - Receiving payload (expected size: 30320117)
[22:58:05] - Downloaded at ~417 kB/s
[22:58:05] - Averaged speed for that direction ~410 kB/s
[22:58:05] + Received work.
[22:58:06] Trying to send all finished work units
[22:58:06] + No unsent completed units remaining.
[22:58:06] + Closed connections
[22:58:06] 
[22:58:06] + Processing work unit
[22:58:06] Core required: FahCore_a5.exe
[22:58:06] Core found.
[22:58:06] Working on queue slot 03 [December 1 22:58:06 UTC]
[22:58:06] + Working ...
[22:58:06] - Calling './FahCore_a5.exe -dir work/ -nice 19 -suffix 03 -np 64 -checkpoint 5 -verbose -lifeline 3423 -version 634'

[22:58:06] 
[22:58:06] *------------------------------*
[22:58:06] Folding@Home Gromacs SMP Core
[22:58:06] Version 2.27 (Thu Feb 10 09:46:40 PST 2011)
[22:58:06] 
[22:58:06] Preparing to commence simulation
[22:58:06] - Looking at optimizations...
[22:58:06] - Created dyn
[22:58:06] - Files status OK
[22:58:08] - Expanded 30319605 -> 33130012 (decompressed 109.2 percent)
[22:58:08] Called DecompressByteArray: compressed_data_size=30319605 data_size=33130012, decompressed_data_size=33130012 diff=0
[22:58:08] - Digital signature verified
[22:58:08] 
[22:58:08] Project: 8105 (Run 0, Clone 11, Gen 193)
[22:58:08] 
[22:58:08] Assembly optimizations on if available.
[22:58:08] Entering M.D.
[22:58:15] Mapping NT from 64 to 64 
[22:58:18] Completed 0 out of 250000 steps  (0%)
[23:05:36] Completed 2500 out of 250000 steps  (1%)
[23:12:36] Completed 5000 out of 250000 steps  (2%)

The second one:

Code: Select all

[10:02:02] - Preparing to get new work unit...
[10:02:02] Cleaning up work directory
[10:02:03] + Attempting to get work packet
[10:02:03] Passkey found
[10:02:03] - Will indicate memory of 32233 MB
[10:02:03] - Connecting to assignment server
[10:02:03] Connecting to http://assign.stanford.edu:8080/
[10:02:04] Posted data.
[10:02:04] Initial: 8F80; - Successful: assigned to (128.143.231.201).
[10:02:04] + News From Folding@Home: Welcome to Folding@Home
[10:02:04] Loaded queue successfully.
[10:02:04] Sent data
[10:02:04] Connecting to http://128.143.231.201:8080/
[10:02:14] Posted data.
[10:02:14] Initial: 0000; - Receiving payload (expected size: 30320117)
[10:03:25] - Downloaded at ~417 kB/s
[10:03:25] - Averaged speed for that direction ~433 kB/s
[10:03:25] + Received work.
[10:03:25] Trying to send all finished work units
[10:03:25] + No unsent completed units remaining.
[10:03:25] + Closed connections
[10:03:25] 
[10:03:25] + Processing work unit
[10:03:25] Core required: FahCore_a5.exe
[10:03:25] Core found.
[10:03:25] Working on queue slot 03 [December 2 10:03:25 UTC]
[10:03:25] + Working ...
[10:03:25] - Calling './FahCore_a5.exe -dir work/ -nice 19 -suffix 03 -np 64 -checkpoint 3 -verbose -lifeline 2077 -version 634'

[10:03:25] 
[10:03:25] *------------------------------*
[10:03:25] Folding@Home Gromacs SMP Core
[10:03:25] Version 2.27 (Thu Feb 10 09:46:40 PST 2011)
[10:03:25] 
[10:03:25] Preparing to commence simulation
[10:03:25] - Looking at optimizations...
[10:03:25] - Created dyn
[10:03:25] - Files status OK
[10:03:29] - Expanded 30319605 -> 33130012 (decompressed 109.2 percent)
[10:03:29] Called DecompressByteArray: compressed_data_size=30319605 data_size=33130012, decompressed_data_size=33130012 diff=0
[10:03:30] - Digital signature verified
[10:03:30] 
[10:03:30] Project: 8105 (Run 0, Clone 11, Gen 193)
[10:03:30] 
[10:03:30] Assembly optimizations on if available.
[10:03:30] Entering M.D.
[10:03:37] Mapping NT from 64 to 64 
[10:03:41] Completed 0 out of 250000 steps  (0%)
[10:08:10] ng M.D.
[10:08:16] Using Gromacs checkpoints
[10:08:19] Mapping NT from 64 to 64 
[10:09:01] Resuming from checkpoint
[10:09:21] Verified work/wudata_03.log
[10:09:21] Verified work/wudata_03.trr
[10:09:21] Verified work/wudata_03.xtc
[10:09:21] Verified work/wudata_03.edr
[10:09:22] Completed 705 out of 250000 steps  (0%)
[10:16:44] Completed 2500 out of 250000 steps  (1%)
[10:27:04] Completed 5000 out of 250000 steps  (2%)

Posted: **Tue Dec 03, 2013 1:37 pm**

The logs indeed are different for v6 and V7. It seems that you got the credits for both WUs (first is what bruce posted above):

Hi -alias- (team 37651),
Your WU (P8105 R0 C11 G193) was added to the stats database on 2013-12-02 20:04:02 for 365810 points of credit.

Posted: **Tue Dec 03, 2013 2:13 pm**

Thanks!
Is that normal, I mean to get credit for 2 of a kind?

Posted: **Tue Dec 03, 2013 2:20 pm**

As stated above, it is rare. The reason is that the WU was first assigned to a donor but possibly, the Server got an error. To rule out hardware error, the Server will generate two additional copies of the WU and will be assigned to the donors. You were lucky to get two WUs on different systems. As long as the WU was assigned to your client and you successfully returned it and met the QRB requirements, you will be assigned points. The same rule applies even in this rare situation.

Posted: **Tue Dec 03, 2013 2:22 pm**

-alias- wrote:Thanks!
Is that normal, I mean to get credit for 2 of a kind?

I've received plenty of duplicate wus over the years and always got credit for them. Mine were never the result of cloning an install. Just keep folding and don't worry about them.

Posted: **Tue Dec 03, 2013 3:08 pm**

Both work units received and credited:

Hi -alias- (team 37651),
Your WU (P8105 R0 C11 G193) was added to the stats database on 2013-12-02 03:14:07 for 443304 points of credit.
Hi -alias- (team 37651),
Your WU (P8105 R0 C11 G193) was added to the stats database on 2013-12-02 20:04:02 for 365810 points of credit.

Posted: **Tue Dec 03, 2013 4:00 pm**

ChelseaOilman wrote:
-alias- wrote:Thanks!
Is that normal, I mean to get credit for 2 of a kind?
I've received plenty of duplicate wus over the years and always got credit for them. Mine were never the result of cloning an install. Just keep folding and don't worry about them.

Right. When a server decides a WU needs to be reissued to someone else (perhaps because the first one wasn't received in time) it creates a new WU with the same PRCG, not a duplicate. When you do something that duplicates a WU, it does not get double credit.

Folding Forum

Same WU folds on two servers simultaneously?

Same WU folds on two servers simultaneously?

Re: Same WU folds on two servers simultaneously?

Re: Same WU folds on two servers simultaneously?

Re: Same WU folds on two servers simultaneously?

Re: Same WU folds on two servers simultaneously?

Re: Same WU folds on two servers simultaneously?

Re: Same WU folds on two servers simultaneously?

Re: Same WU folds on two servers simultaneously?

Re: Same WU folds on two servers simultaneously?

Re: Same WU folds on two servers simultaneously?

Re: Same WU folds on two servers simultaneously?

Re: Same WU folds on two servers simultaneously?

Re: Same WU folds on two servers simultaneously?

Re: Same WU folds on two servers simultaneously?