Rage3D Discussion Area

Rage3D Discussion Area (http://www.rage3d.com/board/index.php)
-   Distributed Computing (Team Rage3D) (http://www.rage3d.com/board/forumdisplay.php?f=43)
-   -   Slowdowns? (http://www.rage3d.com/board/showthread.php?t=34050135)

jima13 Oct 16, 2019 03:23 PM

Slowdowns?
 
Noticed a drop in my production a few weeks back. Was checking some systems and noticed gpu's sitting on ready, and while rebooting got them working again it didn't stick so now I get to babysit them more than I'd like...at least the 2080ti is finally putting out what I bought it for. I'll be moving it to my new gaming system shortly :nuts:

absolutefunk Oct 30, 2019 12:33 PM

There are a couple of servers hosted at Temple which seem to have intermittent network problems. I was also having issues with my GPU slots hanging at WU download up until last week.

Lupine Mar 18, 2020 09:42 AM

[email protected]? :bleh:

Is this a consistent issue? Just rejoined the effort yesterday, so my data range is obviously very limited, but logged into the home Ryzen / Fury rig to see both CPU & GPU sitting idle.

Crawdaddy79 Mar 18, 2020 12:12 PM

Slow here as well. If I let it sit it will go hours without downloading a WU.

Lupine Mar 18, 2020 03:40 PM

Quite possibly unrelated, but when I closed down and restarted the client instead of pausing / unpausing it pulled new WUs.

edit: yeah, random. PC here is playing the waiting game, and no level of prompting makes a difference.

aRandomRobot Mar 18, 2020 05:36 PM

Yeah, I have to babysit the client to keep it from sitting idle which so far has been pretty easy since I'm working from for the foreseeable future. It seems like there are narrow windows where the servers are up and able to issue WUs and if the client misses the window it just sits idle all day with some absurd long timeout. According to the official [email protected] twitter things should get better in the coming days as more server capacity is brought online. The WUs are there you just have to catch the server at the right time.

jima13 Mar 18, 2020 06:26 PM

Quote:

Originally Posted by Lupine (Post 1338182234)
[email protected]? :bleh:

Is this a consistent issue? Just rejoined the effort yesterday, so my data range is obviously very limited, but logged into the home Ryzen / Fury rig to see both CPU & GPU sitting idle.

Yep, that's all happening. A post on the Fah forums is trying to explain it...the combination of new servers coming online and wu's being uploaded plus the number of new folders thanks to CV caused an overload? shortage?....I wish there was a way to limit the retrys to something like 30 min. max as I see next try in 5 hrs ...unless I reboot, and my hdd already sounds like a machine gun :D

Lupine Mar 18, 2020 08:36 PM

Quote:

Originally Posted by jima13 (Post 1338182351)
my hdd already sounds like a machine gun :D

HDD? Arcane wizardry afoot! :eek: :lol:

jima13 Mar 19, 2020 12:56 AM

Quote:

Originally Posted by Lupine (Post 1338182379)
HDD? Arcane wizardry afoot! :eek: :lol:

Yep, relegated to using the storage drive when the Samsung EVO died :bleh:

I have all the parts for another system, Ryzen 7 3700, x570, 2 Corsair 1tb MP600 M.2 NVMe...will move a 2080ti I have in a office system to the new gaming rig and I'll split the 2 1080ti in this rig...when all is done it will be 2080ti, 3 1080ti in seperate systems and my garage rig with 2 980ti...:nuts: I know

Lupine Mar 19, 2020 01:51 AM

Quote:

Originally Posted by jima13 (Post 1338182409)
Yep, relegated to using the storage drive when the Samsung EVO died :bleh:

I have all the parts for another system, Ryzen 7 3700, x570, 2 Corsair 1tb MP600 M.2 NVMe...will move a 2080ti I have in a office system to the new gaming rig and I'll split the 2 1080ti in this rig...when all is done it will be 2080ti, 3 1080ti in seperate systems and my garage rig with 2 980ti...:nuts: I know

You need help, and I shall help you by giving you my mailing address.

:D

Crawdaddy79 Mar 19, 2020 09:14 AM

Whoa - My CPU completed a WU, uploaded, then immediately started working on another WU. I think that's the first time this has happened this week.

Lupine Mar 19, 2020 10:21 AM

CPU idle. GPU working.

jima13 Mar 19, 2020 11:37 AM

Quote:

Originally Posted by Lupine (Post 1338182414)
You need help, and I shall help you by giving you my mailing address.

:D

I'll send ya my electric bill :D

gamefoo21 Mar 19, 2020 12:12 PM

Quote:

Originally Posted by Lupine (Post 1338182470)
CPU idle. GPU working.

That's how it is for me.

Then it switched from COVID-19 to something to do with muscles.

It was interesting because one machine was working on the virus, while the other was working on the 'vaccine'.

*sigh*

Lupine Mar 19, 2020 12:32 PM

Quote:

Originally Posted by jima13 (Post 1338182494)
I'll send ya my electric bill :D

Hrm, this isn't going where intended. :p

Android1 Mar 19, 2020 02:24 PM

Neither my CPU nor my GPU slots are getting any WUs at all. Is anyone else having the same problem?

jima13 Mar 19, 2020 02:41 PM

https://foldingforum.org/viewtopic.php?f=18&t=32424

This is what Bruce posted about the issue...

Android1 Mar 19, 2020 02:54 PM

Quote:

Originally Posted by jima13 (Post 1338182561)
https://foldingforum.org/viewtopic.php?f=18&t=32424

This is what Bruce posted about the issue...

Thanks. I will assume that the issue is with them for now.

By the way, who is Bruce?

jima13 Mar 19, 2020 03:20 PM

Quote:

Originally Posted by Android1 (Post 1338182565)
Thanks. I will assume that the issue is with them for now.

By the way, who is Bruce?

Bruce is one of the forum admins...

Android1 Mar 19, 2020 04:55 PM

Quote:

Originally Posted by jima13 (Post 1338182570)
Bruce is one of the forum admins...

Gotcha. Thanks. :up:

aRandomRobot Mar 20, 2020 01:58 AM

I think maybe itís fixed? My GPU has been crunching on a steady supply of WUs this evening for the first time this week. Hopefully this continues so we can actually see a ppd boost from all the new and returning folders :p

DarkFoss Mar 20, 2020 01:18 PM

More Covid-19 cpu wu's are on the way atleast for those running the advance flag according to this beta thread today will hit the new assignment servers too.
HTML Code:

https://foldingforum.org/viewtopic.php?f=66&t=32888

Crawdaddy79 Mar 21, 2020 07:27 PM

Just now I have a record 42 attempts over the course of 8 hours before a WU got downloaded. I'd been pausing/unpausing periodically but it wasn't working.

Android1 Mar 21, 2020 07:57 PM

Quote:

Originally Posted by Crawdaddy79 (Post 1338183186)
Just now I have a record 42 attempts over the course of 8 hours before a WU got downloaded. I'd been pausing/unpausing periodically but it wasn't working.

Have you tried killing the client in Task Manager and restarting it? I had a bit of trouble getting my last WU, but nothing as serious as what you described.

Android1 Mar 23, 2020 11:06 AM

I am having problems with the servers right now like everyone else, but I am experiencing a different issue. When my client is able to download a GPU work unit, the client says that the WU is is bad. This occurs right after download and has been happening for the past twelve hours. Has anyone else had this happen to them?

Crawdaddy79 Mar 23, 2020 11:23 AM

I've seen that message about three times in the last few days. I don't think it's for us to worry about. Whoever is producing the WU isn't packaging them correctly, according to some post on the Folding forum.

DarkFoss Mar 23, 2020 12:19 PM

Quote:

Originally Posted by Android1 (Post 1338183531)
I am having problems with the servers right now like everyone else, but I am experiencing a different issue. When my client is able to download a GPU work unit, the client says that the WU is is bad. This occurs right after download and has been happening for the past twelve hours. Has anyone else had this happen to them?

AMD card? What error in particular I see this one crop up the wu never starts folding just fails the check and gets sent off.
Code:

ERROR:exception: Error invoking kernel sortShortList: clEnqueueNDRangeKernel (-5)
Also there is even talk about banning AMD cards from certain wu's with very large atom counts per this thread.
HTML Code:

https://foldingforum.org/viewtopic.php?f=81&t=32771
From Joe H (site admin)
Quote:

Someone else looked at this kind of error, it appears from their research that there is a limit set in the AMD OpenCL support that may be causing problems if the size of the WU exceeds some large number of atoms. This will need to be checked into, it is not clear whether this just applies to some version of the drivers, is actually the problem, and so on. The projects running into this error may get restricted from assignments to AMD GPU's at some point, a report was passed on to the [email protected] group. Or the fix might be something else, too soon to be more definite.

Android1 Mar 23, 2020 02:58 PM

Quote:

Originally Posted by Crawdaddy79 (Post 1338183533)
I've seen that message about three times in the last few days. I don't think it's for us to worry about. Whoever is producing the WU isn't packaging them correctly, according to some post on the Folding forum.

Are you referring to this thread? That would explain my problem perfectly.

Quote:

Originally Posted by DarkFoss (Post 1338183550)
AMD card? What error in particular I see this one crop up the wu never starts folding just fails the check and gets sent off.
Code:

ERROR:exception: Error invoking kernel sortShortList: clEnqueueNDRangeKernel (-5)
Also there is even talk about banning AMD cards from certain wu's with very large atom counts per this thread.
HTML Code:

https://foldingforum.org/viewtopic.php?f=81&t=32771
From Joe H (site admin)

I have an AMD Radeon RX 570, and that is the exact same error that I have been getting. Check out the latest post in the thread that I linked to in my reply to Crawdaddy79. It looks as though the issue might be with the WUs.

Turbo_Rage3d_ Mar 23, 2020 03:38 PM

I've been having the same problem getting WUs for my GPUs. I decided to start folding on my 3900x and it's pulling WUs consistantly.

Hope they get it sorted out soon. I miss the ppd I was getting with my 5700xt , gtx 1070, & 5700 (flashed to an xt).

Thankfully I'm not seeing any errors on any of the cards, just not get WUs consistantly at all, even killing [email protected] in task manger & restarting. It will get better soon I'm sure.

Turbo_Rage3d_ Mar 23, 2020 03:40 PM

Quote:

Originally Posted by Turbo_Rage3d_ (Post 1338183599)
I've been having the same problem getting WUs for my GPUs. I decided to start folding on my 3900x and it's pulling WUs consistantly.

Hope they get it sorted out soon. I miss the ppd I was getting with my 5700xt , gtx 1070, & 5700 (flashed to an xt).

Thankfully I'm not seeing any errors on any of the cards, just not get WUs consistantly at all, even killing [email protected] in task manger & restarting. It will get better soon I'm sure.

After I posted this I decided to log into my main rig at home and finally it got a WU!! Woooohoooooo!! LOL


All times are GMT -5. The time now is 02:48 AM.

Powered by vBulletin® Version 3.6.5
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.
All trademarks used are properties of their respective owners. Copyright ©1998-2011 Rage3D.com