GUPPI Rescheduler for Linux and Windows - Move GUPPI work to CPU and non-GUPPI to GPU

Message boards : Number crunching : GUPPI Rescheduler for Linux and Windows - Move GUPPI work to CPU and non-GUPPI to GPU
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 30 · 31 · 32 · 33 · 34 · 35 · 36 . . . 37 · Next

AuthorMessage
Profile Brent Norman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester

Send message
Joined: 1 Dec 99
Posts: 2786
Credit: 685,657,289
RAC: 835
Canada
Message 1871486 - Posted: 7 Jun 2017, 0:45:30 UTC - in response to Message 1871480.  

You can NOT reschedule if you are ACTIVLY running both S@H and S@H Beta.

It makes sense the Beta server sent you CPU tasks and you try to send them back to S@H server ... the servers go .... WTF?????
ID: 1871486 · Report as offensive
Stephen "Heretic" Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 20 Sep 12
Posts: 5557
Credit: 192,787,363
RAC: 628
Australia
Message 1871502 - Posted: 7 Jun 2017, 2:03:25 UTC - in response to Message 1871480.  

Ok, I'm confused. I got this message because I am running SetiBeta.

Reading configuration files...
Warning: This machine appears to be attached to SETI@Home Beta.
If there are active Beta work units, this program may reassign them improperly,
as they look like SETI@Home work. OK to proceed (Y to continue)?Y


So what am I supposed to do, I don't want the SetiBeta files to get scrambled into the production queue?

Thanks,
Tom


. . The only thing I can suggest is that you reduce your Beta-testing cache to only one or two jobs then set NNT and wait for them to complete and upload before you run the rescheduler. After it is complete you can remove the NNT setting and crunch for Beta-testing again.

. . Have fun!

Stephen

:)
ID: 1871502 · Report as offensive
rob smith Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer moderator
Volunteer tester

Send message
Joined: 7 Mar 03
Posts: 22158
Credit: 416,307,556
RAC: 380
United Kingdom
Message 1871518 - Posted: 7 Jun 2017, 5:00:16 UTC - in response to Message 1871480.  

If you don't want to run Beta then set Beta to "No Tasks", run that part of the cache down, then "remove" the project.
As Beta is about testing NEW applications with a known sample of work that has been tried and tested against the baseline applications rescheduling any Beta task is a VERY BAD THING.
Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?
ID: 1871518 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13720
Credit: 208,696,464
RAC: 304
Australia
Message 1871521 - Posted: 7 Jun 2017, 6:09:20 UTC - in response to Message 1871480.  

So what am I supposed to do, I don't want the SetiBeta files to get scrambled into the production queue?

Or just don't reschedule?
Some people like to micro-manage, others get by ok without it.
Grant
Darwin NT
ID: 1871521 · Report as offensive
Stephen "Heretic" Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 20 Sep 12
Posts: 5557
Credit: 192,787,363
RAC: 628
Australia
Message 1871522 - Posted: 7 Jun 2017, 6:35:52 UTC - in response to Message 1871521.  

So what am I supposed to do, I don't want the SetiBeta files to get scrambled into the production queue?

Or just don't reschedule?
Some people like to micro-manage, others get by ok without it.


. . Hi Grant,

. . You puzzle me, you are against what you call "micromanagement" but you read this forum. Why does it fascinate you ? :)

Stephen

??
ID: 1871522 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13720
Credit: 208,696,464
RAC: 304
Australia
Message 1871523 - Posted: 7 Jun 2017, 6:57:24 UTC - in response to Message 1871522.  

. . You puzzle me, you are against what you call "micromanagement"

Not just me, don't forget all the dictionary definitions as well.

Why does it fascinate you ? :)

What would make you think i'm fascinated by it?
I've checked out all the other threads, this was one was recently active, so I checked it out too.
Grant
Darwin NT
ID: 1871523 · Report as offensive
Profile Jimbocous Project Donor
Volunteer tester
Avatar

Send message
Joined: 1 Apr 13
Posts: 1849
Credit: 268,616,081
RAC: 1,349
United States
Message 1871526 - Posted: 7 Jun 2017, 7:31:29 UTC - in response to Message 1871523.  
Last modified: 7 Jun 2017, 7:49:31 UTC

. . You puzzle me, you are against what you call "micromanagement"

Not just me, don't forget all the dictionary definitions as well.

Why does it fascinate you ? :)

What would make you think i'm fascinated by it?
I've checked out all the other threads, this was one was recently active, so I checked it out too.

Can we not revive this debate in this thread? Not only is the horse dead, it smells, badly!
And I believe there was another thread for it, in any event, so that this thread could stick to its stated purpose of supporting those that do reschedule ...
ID: 1871526 · Report as offensive
Profile Jimbocous Project Donor
Volunteer tester
Avatar

Send message
Joined: 1 Apr 13
Posts: 1849
Credit: 268,616,081
RAC: 1,349
United States
Message 1871528 - Posted: 7 Jun 2017, 7:47:22 UTC - in response to Message 1871480.  

Ok, I'm confused. I got this message because I am running SetiBeta.

Reading configuration files...
Warning: This machine appears to be attached to SETI@Home Beta.
If there are active Beta work units, this program may reassign them improperly,
as they look like SETI@Home work. OK to proceed (Y to continue)?Y


So what am I supposed to do, I don't want the SetiBeta files to get scrambled into the production queue?

Thanks,
Tom

Pretty straight forward, actually.
No need to remove the Beta project, just complete or abort any existing tasks, make sure you're set for "no New Tasks" in Project management, and reschedule to your hearts content.
If at some point there's actually some Beta development that needs someone running tasks that you wish to participate in, stop Rescheduling.
I have Beta on most of my crunchers, haven't seen any issues with the above approach.
ID: 1871528 · Report as offensive
Stephen "Heretic" Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 20 Sep 12
Posts: 5557
Credit: 192,787,363
RAC: 628
Australia
Message 1871543 - Posted: 7 Jun 2017, 11:52:25 UTC - in response to Message 1871523.  


Why does it fascinate you ? :)

What would make you think i'm fascinated by it?
I've checked out all the other threads, this was one was recently active, so I checked it out too.


. . Fair cop, that would explain it ...

Stephen

:)
ID: 1871543 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1871599 - Posted: 7 Jun 2017, 17:17:17 UTC

Just a quick note that I had very good success in using the rescheduler to bank enough work for Tuesday's outage to make it through mostly. I undershot by one rescheduler session on one FX system that left me without enough GPU work for the last hour of the outage. Its just the nature of the beast I will always run out of CPU tasks on the Ryzen system as I will always work through the maximum allowed of 100 tasks before the servers come back. Still fine tuning my procedure, but overall a complete success with the rescheduler this past Tuesday.

Thank you Mr. Kevvy. Good job!
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1871599 · Report as offensive
Stephen "Heretic" Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 20 Sep 12
Posts: 5557
Credit: 192,787,363
RAC: 628
Australia
Message 1871642 - Posted: 7 Jun 2017, 23:21:27 UTC - in response to Message 1871599.  

Still fine tuning my procedure, but overall a complete success with the rescheduler this past Tuesday.
Thank you Mr. Kevvy. Good job!


. . Is that his original version or has he released an update I am not aware of? It is a very useful thing either way.

Stephen

?
ID: 1871642 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1872548 - Posted: 12 Jun 2017, 7:01:07 UTC

Anyone else having issues using the rescheduler today? I have't been able to get it to work on any of my machines today. I wonder if it is not working because of the large amount of AP work that is on every machine. I don't remember this being an issue in the past but that was when there was only a task or two of AP, not the dozens I have onboard today.

The rescheduler gets "stuck" at "reading configuration file" It doesn't go any further than that. You can let the command window sit for 15 minutes and it never progresses further. You just exit the command window and restart BOINC and things start back up from where you left. All the client_state files have the same timestamp and file size so the backup copy ran. But no work was swapped between CPU and GPU. I had 40+ tasks to swap on Numbskull and nothing changed. Any others having problems today?
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1872548 · Report as offensive
Profile Jimbocous Project Donor
Volunteer tester
Avatar

Send message
Joined: 1 Apr 13
Posts: 1849
Credit: 268,616,081
RAC: 1,349
United States
Message 1872549 - Posted: 12 Jun 2017, 7:02:11 UTC - in response to Message 1872548.  

Anyone else having issues using the rescheduler today? I have't been able to get it to work on any of my machines today. I wonder if it is not working because of the large amount of AP work that is on every machine. I don't remember this being an issue in the past but that was when there was only a task or two of AP, not the dozens I have onboard today.

The rescheduler gets "stuck" at "reading configuration file" It doesn't go any further than that. You can let the command window sit for 15 minutes and it never progresses further. You just exit the command window and restart BOINC and things start back up from where you left. All the client_state files have the same timestamp and file size so the backup copy ran. But no work was swapped between CPU and GPU. I had 40+ tasks to swap on Numbskull and nothing changed. Any others having problems today?

Haven't seen any issues here ...
ID: 1872549 · Report as offensive
Stephen "Heretic" Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 20 Sep 12
Posts: 5557
Credit: 192,787,363
RAC: 628
Australia
Message 1872578 - Posted: 12 Jun 2017, 15:37:52 UTC - in response to Message 1872548.  

Anyone else having issues using the rescheduler today? I have't been able to get it to work on any of my machines today. I wonder if it is not working because of the large amount of AP work that is on every machine. I don't remember this being an issue in the past but that was when there was only a task or two of AP, not the dozens I have onboard today.

The rescheduler gets "stuck" at "reading configuration file" It doesn't go any further than that. You can let the command window sit for 15 minutes and it never progresses further. You just exit the command window and restart BOINC and things start back up from where you left. All the client_state files have the same timestamp and file size so the backup copy ran. But no work was swapped between CPU and GPU. I had 40+ tasks to swap on Numbskull and nothing changed. Any others having problems today?


. . Wasn't it an issue about NOT using the rescheduler when you have AP tasks? Just like when you have Beta tasks??

Stephen

??
ID: 1872578 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1872581 - Posted: 12 Jun 2017, 16:00:07 UTC

In the 0.51 readme Changelog, it says that it was fixed to handle AP tasks without issues.
0.51	Fixed error with wrong state file being checked to avoid AP version

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1872581 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1872591 - Posted: 12 Jun 2017, 16:51:10 UTC - in response to Message 1872549.  


Haven't seen any issues here ...

Do you in fact have any AP tasks on board, Jim?
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1872591 · Report as offensive
Stephen "Heretic" Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 20 Sep 12
Posts: 5557
Credit: 192,787,363
RAC: 628
Australia
Message 1872592 - Posted: 12 Jun 2017, 16:58:58 UTC - in response to Message 1872581.  

In the 0.51 readme Changelog, it says that it was fixed to handle AP tasks without issues.
0.51	Fixed error with wrong state file being checked to avoid AP version



. . OK, good to know.

Stephen

.
ID: 1872592 · Report as offensive
Profile petri33
Volunteer tester

Send message
Joined: 6 Jun 02
Posts: 1668
Credit: 623,086,772
RAC: 156
Finland
Message 1872606 - Posted: 12 Jun 2017, 18:51:11 UTC

I'm not rescheduling. I'm cheating. I fake I have a GTX 1080 Tu. That is a Tuesday edition. I get 400 tasks per GP.U.
Since the the day I started that my daily score has levelled out to a flat line. My GPUs were task-limited during the week. The TUesday will still hurt me. I'm not planning to do more harm. With shorties I'll run out of cache, with Guppi I can manage -- the Tuesdays.

I'm sorry if I have offeded anyone with my settings.

I feel (And I think that many Linux users are going to feel) that something must be done to the 100 WU limit.

I will not release the one/two line fix to this problem. It is the Administrations work to do.
--
p33

p.s. I recommend a cache that is somehow bound to the daily output/RAC.
To overcome Heisenbergs:
"You can't always get what you want / but if you try sometimes you just might find / you get what you need." -- Rolling Stones
ID: 1872606 · Report as offensive
Profile Brent Norman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester

Send message
Joined: 1 Dec 99
Posts: 2786
Credit: 685,657,289
RAC: 835
Canada
Message 1872608 - Posted: 12 Jun 2017, 18:54:43 UTC - in response to Message 1872606.  
Last modified: 12 Jun 2017, 18:55:55 UTC

p.s. I recommend a cache that is somehow bound to the daily output/RAC.

I agree, even if that is a <EDIT> 15 12 hour cache, many of us would be happy!
ID: 1872608 · Report as offensive
Profile petri33
Volunteer tester

Send message
Joined: 6 Jun 02
Posts: 1668
Credit: 623,086,772
RAC: 156
Finland
Message 1872609 - Posted: 12 Jun 2017, 18:57:40 UTC - in response to Message 1872608.  

p.s. I recommend a cache that is somehow bound to the daily output/RAC.

I agree, even if that is a <EDIT> 15 12 hour cache, many of us would be happy!

+1
To overcome Heisenbergs:
"You can't always get what you want / but if you try sometimes you just might find / you get what you need." -- Rolling Stones
ID: 1872609 · Report as offensive
Previous · 1 . . . 30 · 31 · 32 · 33 · 34 · 35 · 36 . . . 37 · Next

Message boards : Number crunching : GUPPI Rescheduler for Linux and Windows - Move GUPPI work to CPU and non-GUPPI to GPU


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.