LTT Folding Team's Emergency Response to Covid-19

[Guellan]

3 minutes ago, Macaw2000 said:

 

I'm not using the advanced flag. I toyed with in my devbox but it didn't seem to offer any point benefits. Grabbing the WU early seems to be the only benefit but at scale (e.g. 60 concurrent containers running, it doesn't seem to solve the issue. Would flip it on if I'm wrong here but yesterday's discussion seemed to confirm.

I have not had an empty slot since adding the advanced flag.

[ConceptOW]

I guess everyone got 0 points for 4/2 9pm & 4/3 6am & 4/3 9am.

[Guellan]

For me I was averaging 700k per day before adding advanced flag, and now am averaging 1.4M ppd since... ~50% idle time.

[42thgamer]

Yes! Getting all the WUs I need! Are they able to keep up? Or am I just lucky?

[Inkertus]

2 minutes ago, 42thgamer said:

Yes! Getting all the WUs I need! Are they able to keep up? Or am I just lucky?

Except for a handful of hiccups that require the odd tinker with I'm 6/6 all day.

[42thgamer]

1 minute ago, Inkertus said:

Except for a handful of hiccups that require the odd tinker with I'm 6/6 all day.

Nice! Get us those points!

[damnfinecoffee]

4 hours ago, Tomatodude95 said:

I just looked those up. Are you UK located? Those look like static IPs from Kcom ISP, might that be your ISP? Looks like just an FTP server running behind them, might that be your own IP, not the ones that get blocked?

Sorry, I posted the wrong IP's. These are the one's being blocked:

185.216.140.34
185.175.93.11

 

[LU0P10]

This has been the gap/pause between the WUs on my clients (so it´s about couple of seconds) 

[n0:name]

So, after fiddling (or rather, fighting) around with networking, Proxmox and getting the OpenCL drivers to work under Linux I finally have a rig up folding 24/7:

Spoiler

 

CPU working hard to push that info to the GPU as well:

Spoiler

 

 

Interestingly, the GPU pass-through worked flawlessly, thanks to this guide:

Spoiler

 

 

[walderston]

2 minutes ago, LU0P10 said:

This has been the gap/pause between the WUs on my clients (so it´s about couple of seconds) 

That would be the checkpoint save. 

[Darkjacky]

How do they expect the client to send back info with no collection server?

Spoiler

I can take a guess and they probably use the work server

Also this workload is extremely heavy. It takes 20 seconds to save the 5% snapshots.

3.7M points so far. Should be getting over 4M soon.

Yeah so my CPU looks like this in the period where my GPU is sitting idle.

[gmouz]

I'm still waiting to be credited the 20 WUs I have completed in the last 12 hours but EOC seems to be unable to update

 

[LU0P10]

1 minute ago, walderston said:

That would be the checkpoint save. 

You might be right... I try to take another screenshot when WU finishes.

[walderston]

Spoiler

1 minute ago, gmouz said:

I'm still waiting to be credited the 20 WUs I have completed in the last 12 hours but EOC seems to be unable to update

 

Its not EOC, its F@H.. I'm missing about 55 WUs but they are slowly being credited on https://apps.foldingathome.org/wu

[Inkertus]

3 minutes ago, LU0P10 said:

You might be right... I try to take another screenshot when WU finishes.

 

Oh I just noticed that's the middle of a job.  Yeah as they said that specifically is very likely a checkpoint.

7 minutes ago, LU0P10 said:

This has been the gap/pause between the WUs on my clients (so it´s about couple of seconds) 

 

Looks like about the spool up time between starting a job and it actually crunching numbers.  I've noticed a delay of up to 30 seconds tops when it starts a new one, so technically no downtime.

[LU0P10]

3 minutes ago, Inkertus said:

Looks like about the spool up time between starting a job and it actually crunching numbers.  I've noticed a delay of up to 30 seconds tops when it starts a new one, so technically no downtime.

Exactly what I was trying to say in the end... no downtime, at least on my machines... I dunno if one should say this out loud and risk rustling the feathers of all mighty WU Gods 

[Inkertus]

1 minute ago, LU0P10 said:

Exactly what I was trying to say in the end... no downtime, at least on my machines... I dunno if one should say this out loud and risk rustling the feathers of all mighty WU Gods 

Well apparently the secret now is to not be in Ohio since a couple people mentioned not getting anything over there. Otherwise, NU% 90, client type advanced, and a good internet connection have been doing me good.

[HaRdLy007]

I give up. Been trying to get either fahcontrol or hfm.net to track my linux box, but it refuses to work.

[walderston]

1 minute ago, HaRdLy007 said:

I give up. Been trying to get either fahcontrol or hfm.net to track my linux box, but it refuses to work.

make sure there isn't a firewall blocking the port.. thats caught me out 

[HaRdLy007]

Just now, walderston said:

make sure there isn't a firewall blocking the port.. thats caught me out 

Linux box has no firewall enabled as far as I know, and the Windows box can track 2 other Windows boxes just fine. Not sure what to do next.

[walderston]

2 hours ago, J1mjam said:

Okay, so there's something REALLY weird going on right now.

One of the WU I havn't been awarded points for is Project: 14543 (Run 0, Clone 284, Gen 10).

 

According to my logs, I finished this, this morning and uploaded successfully. 

 

07:29:29:WU01:FS01:Sending unit results: id:01 state:SEND error:NO_ERROR project:14543 run:0 clone:284 gen:10 core:0x22 unit:0x0000000d80fccb045e7fc1c120db0e08
07:29:55:WU01:FS01:Upload complete
07:29:55:WU01:FS01:Server responded WORK_ACK (400)
07:29:55:WU01:FS01:Final credit estimate, 171338.00 points
07:29:55:WU01:FS01:Cleaning up

But, according to the fah website, that WU was done by someone else, which errored, and was credited before it was returned. 

https://apps.foldingathome.org/wu#project=14543&run=0&clone=284&gen=10 

 

However, I was assigned that WU according to my logs..

04:18:59:WU01:FS01:Connecting to 65.254.110.245:8080
04:18:59:WU01:FS01:Assigned to work server 128.252.203.4
04:18:59:WU01:FS01:Requesting new work unit for slot 01: RUNNING gpu:0:GP102 [GeForce GTX 1080 Ti] 11380 from 128.252.203.4
04:18:59:WU01:FS01:Connecting to 128.252.203.4:8080
04:19:00:WU01:FS01:Downloading 38.42MiB
04:19:06:WU01:FS01:Download 36.12%
04:19:12:WU01:FS01:Download 81.35%
04:19:14:WU01:FS01:Download complete
04:19:14:WU01:FS01:Received Unit: id:01 state:DOWNLOAD error:NO_ERROR project:14543 run:0 clone:284 gen:10 core:0x22 unit:0x0000000d80fccb045e7fc1c120db0e08

 

I am so confused...
 

If you check now, its been credited to you. 

[cbigfoot]

 

It updated

 

 

[enterprise-psi]

Me: Hey PC hows it going?

PC:

15:43:05:WU00:FS01:0x22:Bad State detected... attempting to resume from last good checkpoint. Is your system overclocked?
15:43:05:WU00:FS01:0x22:Following exception occured: Particle coordinate is nan

 

Me: why bruh

PC:

16:47:04:WU00:FS01:0x22:Bad State detected... attempting to resume from last good checkpoint. Is your system overclocked?
16:47:04:WU00:FS01:0x22:Following exception occured: Particle coordinate is nan

Me: You suck bruh , i have even down clocked you!

 

PC:

17:05:10:WU00:FS01:0x22:Bad State detected... attempting to resume from last good checkpoint. Is your system overclocked?
17:05:10:WU00:FS01:0x22:Following exception occured: Particle coordinate is nan
17:05:10:WU00:FS01:0x22:ERROR:114: Max Retries Reached
17:05:10:WU00:FS01:0x22:Saving result file ..\logfile_01.txt
17:05:10:WU00:FS01:0x22:Saving result file badstate-0.xml
17:05:10:WU00:FS01:0x22:Saving result file badstate-1.xml
17:05:10:WU00:FS01:0x22:Saving result file badstate-2.xml
17:05:10:WU00:FS01:0x22:Saving result file checkpointState.xml
17:05:10:WU00:FS01:0x22:Saving result file checkpt.crc
17:05:10:WU00:FS01:0x22:Saving result file positions.xtc
17:05:10:WU00:FS01:0x22:Saving result file science.log
17:05:10:WU00:FS01:0x22:Folding@home Core Shutdown: BAD_WORK_UNIT
17:05:11:WARNING:WU00:FS01:FahCore returned: BAD_WORK_UNIT (114 = 0x72)
17:05:11:WU00:FS01:Sending unit results: id:00 state:SEND error:FAULTY project:14541 run:0 clone:439 gen:21 core:0x22 unit:0x0000001a0d5262775e7d2523aa0cb1d4

 

 

Me: Bruhhhhhhhhhhhhhhhh

 

Classic AMD , issues with folding, issues with drivers, issues with mining , any new omega drivers ?

[cbigfoot]

6 minutes ago, enterprise-psi said:

Me: Hey PC hows it going?

PC:

Spoiler

 

 

Me: why bruh

PC:

Spoiler

 

Me: You suck bruh , i have even down clocked you!

 

PC:

Spoiler

 

 

 

Me: Bruhhhhhhhhhhhhhhhh

 

Classic AMD , issues with folding, issues with drivers, issues with mining , any new omega drivers ?

He watch it i have many amd rigs only 1 giving me issues and i think its my decade old psu New one enroute GRRRRRRRRRRRRRRRRR Looking at you Bender

bender Replies...

[Shigeru]

Looks like some servers are a bit wonky, first time wwith problems to send a finished WU