7.3.6 Not Following next-unit-setting

[P5-133XL]

There are several reports that the 7.3.6 is no longer paying attention to the next-unit-percentage setting.

See:  http://foldingforum.org/viewtopic.php?f=88&t=23754

[jcoffland]

Please provide more information about how it is failing and under what circumstances.

[bborden]

Case 1: next-unit-percentage set to 100 but download starts at 99.

06:29:14:******************************** Build ********************************
06:29:14: Version: 7.3.6
...
06:32:14: <slot id='1' type='CPU'>
06:32:14: <cpus v='2'/>
06:32:14: <next-unit-percentage v='100'/>
06:32:14: </slot>
...
11:14:55:WU01:FS01:0xa3:Completed 495000 out of 500000 steps (99%)
11:14:56:WU00:FS01:Connecting to assign3.stanford.edu:8080
11:14:57:WU00:FS01:News: Welcome to Folding@Home
11:14:57:WU00:FS01:Assigned to work server 171.67.108.60
11:15:00:WU00:FS01:Download complete
...
11:30:16:WU01:FS01:0xa3:Completed 500000 out of 500000 steps (100%)
11:30:27:WU01:FS01:0xa3:Finished Work Unit:

Note: 100 has worked correctly in earlier versions (and perhaps with a different FahCore?) starting the download immediately after 11:30:16.

[bborden]

Case 2: Download does start at 100%

20:40:16: Version: 7.3.2
...
22:41:33:WU01:FS00:0xa4:Completed 247500 out of 250000 steps (99%)
22:44:58:WU01:FS00:0xa4:Completed 250000 out of 250000 steps (100%)
22:44:58:WU00:FS00:Connecting to assign3.stanford.edu:8080
22:44:58:WU00:FS00:News: Welcome to Folding@Home
22:44:58:WU00:FS00:Requesting new work unit for slot 00: RUNNING cpu:4 from 171.67.108.60
22:44:58:WU00:FS00:Connecting to 171.67.108.60:8080
22:44:59:WU01:FS00:0xa4:DynamicWrapper: Finished Work Unit: sleep=10000
22:44:59:WU00:FS00:Downloading 1.46MiB
22:45:05:WU00:FS00:Download 60.06%
22:45:09:WU00:FS00:Download complete
22:45:09:WU01:FS00:0xa4:Finished Work Unit:

[PantherX]

BTW, with FahCore?_17, it doesn't report 100%. I am using V7.2.9, and this is what happens which is what I would expect:
17:10:02:WU02:FS00:0x17:Completed 2400000 out of 2500000 steps (96%)
17:22:11:WU02:FS00:0x17:Completed 2450000 out of 2500000 steps (98%)
17:34:26:WU02:FS00:0x17:Saving result file logfile_01.txt
17:34:26:WU02:FS00:0x17:Saving result file checkpointState.xml
17:34:26:WU02:FS00:0x17:Saving result file checkpt.crc
17:34:26:WU02:FS00:0x17:Saving result file log.txt
17:34:26:WU02:FS00:0x17:Saving result file positions.xtc
17:34:27:WU02:FS00:0x17:Folding@home Core Shutdown: FINISHED_UNIT
17:34:28:WU02:FS00:FahCore returned: FINISHED_UNIT (100 = 0x64)
17:34:28:WU02:FS00:Sending unit results: id:02 state:SEND error:NO_ERROR project:7662 run:27 clone:10 gen:60 core:0x17 unit:0x0000005aff3d4835513923d05137572c
17:34:28:WU02:FS00:Uploading 5.84MiB to 171.67.108.149
17:34:28:WU02:FS00:Connecting to 171.67.108.149:8080
17:34:28:WU01:FS00:Connecting to assign-GPU.stanford.edu:80
17:34:29:WU01:FS00:News: Welcome to Folding@Home
17:34:29:WU01:FS00:Assigned to work server 171.67.108.149
17:34:29:WU01:FS00:Requesting new work unit for slot 00: READY gpu:0:"GF114 GTX 675M?" from 171.67.108.149
17:34:29:WU01:FS00:Connecting to 171.67.108.149:8080
17:34:33:WU01:FS00:Downloading 1.63MiB

[bborden]

Some cores stop one step before reaching 100% and others do reach 100% and do report it. The missing message shouldn't matter. (You can ask Yutong to change what the core reports.)

In either case, the Client should recognize that the FahCore? has finished, is closing the work files and is preparing to zip the data for uploading. That has successfully triggered the download of a new WU, whether it's at 100% or at 99.99999% Nevertheless in your example, [and assuming that next-unit-percentage is set to 100] the download should be triggered at 17:34:26, and not before.

This did work correctly in previous versions.

In my first example, note the differences between times 11:14:56 and 11:30:16 and compare to the times 22:44:58 and 22:44:58.

[jcoffland]

Fixed in v7.3.11