Problems and Technical Issues with Rosetta@home

Message boards : Number crunching : Problems and Technical Issues with Rosetta@home

To post messages, you must log in.

Previous · 1 . . . 214 · 215 · 216 · 217 · 218 · 219 · 220 . . . 257 · Next

AuthorMessage
kotenok2000
Avatar

Send message
Joined: 22 Feb 11
Posts: 183
Credit: 103,927
RAC: 5
Message 106324 - Posted: 29 May 2022, 0:51:13 UTC - in response to Message 106323.  

There is this 1600x1200 Eizo FlexScan
https://www.amazon.com/Eizo-FlexScan-S2133-BK-Monitor-1600x1200/dp/B00KDMMKD0
ID: 106324 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Peter Hucker
Avatar

Send message
Joined: 12 Aug 06
Posts: 1479
Credit: 7,138,291
RAC: 10,433
Message 106325 - Posted: 29 May 2022, 0:52:46 UTC - in response to Message 106324.  

There is this 1600x1200 Eizo FlexScan
https://www.amazon.com/Eizo-FlexScan-S2133-BK-Monitor-1600x1200/dp/B00KDMMKD0
A thousand dollars for a 21" screen? What planet do you come from? I have a 20" 4:3 here at that resolution I bought for $60.
ID: 106325 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Zarck

Send message
Joined: 18 Sep 05
Posts: 39
Credit: 7,240,817
RAC: 545
Message 106348 - Posted: 5 Jun 2022, 17:04:02 UTC

I don't know what the problem is, the rosetta units advance in calculation but the load of the cpus remains at zero... as if nothing was calculated...

Tested with Boinc Windows 7.16.20 (x64) with or without VM.

A solution ?

Thanks in advance for the reply.
ID: 106348 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
kotenok2000
Avatar

Send message
Joined: 22 Feb 11
Posts: 183
Credit: 103,927
RAC: 5
Message 106349 - Posted: 5 Jun 2022, 17:13:58 UTC - in response to Message 106348.  

I think you can open virtualbox manager gui and show vm screens and if vm is stuck you can try to reset vm.
ID: 106349 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile robertmiles

Send message
Joined: 16 Jun 08
Posts: 1217
Credit: 13,359,660
RAC: 640
Message 106353 - Posted: 7 Jun 2022, 23:05:48 UTC

I have a Rosetta Python task with a problem I've never seen before.

It's been running (or claiming to be) for hours, but it's still showing no CPU time and no memory used so far.


Application rosetta python projects 1.03 (vbox64)
Name aaam-PHE_pp-mNMVAL-ACPC-AMACBEN2_2_2554686_6
State Running
Received 6/4/2022 6:13:14 PM
Report deadline 6/7/2022 6:13:13 PM
Estimated computation size 80,000 GFLOPs
CPU time ---
CPU time since checkpoint ---
Elapsed time ---
Estimated time remaining 08:00:00
Fraction done 0.000%
Virtual memory size 0 bytes
Working set size 0 bytes
Directory slots/6
Executable vboxwrapper_26203_windows_x86_64.exe


I've decided to let it reach its deadline, then abort it.
ID: 106353 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Peter Hucker
Avatar

Send message
Joined: 12 Aug 06
Posts: 1479
Credit: 7,138,291
RAC: 10,433
Message 106364 - Posted: 8 Jun 2022, 17:02:17 UTC - in response to Message 106353.  

I've never checked the memory, but no CPU time is the common bug everyone is getting, just send it back and get on with another. It's not going to be of any use sitting there not using the CPU, you're just wasting a core.

Hey, we're actually nearing the end of the pythons! Hopefully either the next batch will work better, or preferably they won't use VB anymore.
ID: 106364 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Sid Celery

Send message
Joined: 11 Feb 08
Posts: 1902
Credit: 35,032,850
RAC: 3,911
Message 106367 - Posted: 9 Jun 2022, 9:45:51 UTC - in response to Message 106364.  

no CPU time is the common bug everyone is getting

Correct.
As soon as you see it happening, don't wait for something to change - it never EVER does.
Abort it on sight every time
ID: 106367 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Peter Hucker
Avatar

Send message
Joined: 12 Aug 06
Posts: 1479
Credit: 7,138,291
RAC: 10,433
Message 106371 - Posted: 9 Jun 2022, 19:38:06 UTC - in response to Message 106367.  

no CPU time is the common bug everyone is getting

Correct.
As soon as you see it happening, don't wait for something to change - it never EVER does.
Abort it on sight every time
Sounds like an instruction to a soldier.
ID: 106371 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
.clair.

Send message
Joined: 2 Jan 07
Posts: 202
Credit: 23,702,139
RAC: 4,186
Message 106372 - Posted: 10 Jun 2022, 0:15:56 UTC

one of the `fun` things about the Linux command line is the `kill` command
A great way to deal will zombie processes
and rosetta workunits .
ID: 106372 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Peter Hucker
Avatar

Send message
Joined: 12 Aug 06
Posts: 1479
Credit: 7,138,291
RAC: 10,433
Message 106373 - Posted: 10 Jun 2022, 0:34:09 UTC
Last modified: 10 Jun 2022, 0:34:40 UTC

You mean exactly like pressing "end task" in Windows Task Manager?

And wouldn't you be better letting Boinc kill it? Then it knows to start another one.
ID: 106373 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Michael E.

Send message
Joined: 5 Apr 08
Posts: 16
Credit: 1,720,091
RAC: 9
Message 106374 - Posted: 10 Jun 2022, 2:59:27 UTC - in response to Message 106367.  

I did an experiment with one of these tasks that seem to never finish. I let it run. Elapsed time is now 4 days 5:12:xx hours with 00:00:01 seconds remaining. I am curious if it will ever finish and was curious about what the credits might look like, but the task took too long.

The Rosetta BOINC status says the task has: "timed out - no response". Task id is: aagb-FPS_pp-mNMPHE-mACPenC13T-B3LEU_pp_3_2601503_6_0

As Sid says, if the time needed for a "Rosetta python projects 1.03 (vbox64)" is much more than is expected, stop the task.
ID: 106374 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 1 Dec 05
Posts: 1664
Credit: 6,533,949
RAC: 1,745
Message 106375 - Posted: 10 Jun 2022, 7:06:37 UTC - in response to Message 106278.  

Ah, this is the problem. The Python book has only one use:


It's a pity, the new version of Python seems great
ID: 106375 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
.clair.

Send message
Joined: 2 Jan 07
Posts: 202
Credit: 23,702,139
RAC: 4,186
Message 106377 - Posted: 10 Jun 2022, 18:34:58 UTC - in response to Message 106373.  

You mean exactly like pressing "end task" in Windows Task Manager?
And wouldn't you be better letting Boinc kill it? Then it knows to start another one.

Yes , true , and it iz wot I do , Just commenting and being a silly bint .
ID: 106377 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Peter Hucker
Avatar

Send message
Joined: 12 Aug 06
Posts: 1479
Credit: 7,138,291
RAC: 10,433
Message 106378 - Posted: 10 Jun 2022, 23:45:46 UTC - in response to Message 106375.  

Ah, this is the problem. The Python book has only one use:
It's a pity, the new version of Python seems great
PyPy ROTFPMSL! Is that the one for kids?
ID: 106378 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Peter Hucker
Avatar

Send message
Joined: 12 Aug 06
Posts: 1479
Credit: 7,138,291
RAC: 10,433
Message 106379 - Posted: 10 Jun 2022, 23:46:33 UTC - in response to Message 106377.  

You mean exactly like pressing "end task" in Windows Task Manager?
And wouldn't you be better letting Boinc kill it? Then it knows to start another one.

Yes , true , and it iz wot I do , Just commenting and being a silly bint .
You sound like a fun bint, what are you wearing? ;-)
ID: 106379 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
.clair.

Send message
Joined: 2 Jan 07
Posts: 202
Credit: 23,702,139
RAC: 4,186
Message 106380 - Posted: 11 Jun 2022, 0:24:36 UTC - in response to Message 106379.  
Last modified: 11 Jun 2022, 0:27:29 UTC

You mean exactly like pressing "end task" in Windows Task Manager?
And wouldn't you be better letting Boinc kill it? Then it knows to start another one.

Yes , true , and it iz wot I do , Just commenting and being a silly bint .
You sound like a fun bint, what are you wearing? ;-)

At the moment I am not wearing my one piece `avatar` na`vi suit , with tail , though the hair is real ,
I never was as good looking as Neytiri , or blue ;-)
ID: 106380 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Peter Hucker
Avatar

Send message
Joined: 12 Aug 06
Posts: 1479
Credit: 7,138,291
RAC: 10,433
Message 106381 - Posted: 11 Jun 2022, 0:40:11 UTC - in response to Message 106380.  

You mean exactly like pressing "end task" in Windows Task Manager?
And wouldn't you be better letting Boinc kill it? Then it knows to start another one.

Yes , true , and it iz wot I do , Just commenting and being a silly bint .
You sound like a fun bint, what are you wearing? ;-)

At the moment I am not wearing my one piece `avatar` na`vi suit
If you're not wearing it then.... oh!

, with tail , though the hair is real ,I never was as good looking as Neytiri , or blue ;-)
I'm not so sure that's a suit. Isn't that just her natural skin?
https://james-camerons-avatar.fandom.com/wiki/Na%27vi
ID: 106381 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Sid Celery

Send message
Joined: 11 Feb 08
Posts: 1902
Credit: 35,032,850
RAC: 3,911
Message 106382 - Posted: 11 Jun 2022, 7:59:08 UTC - in response to Message 106374.  

I did an experiment with one of these tasks that seem to never finish. I let it run. Elapsed time is now 4 days 5:12:xx hours with 00:00:01 seconds remaining. I am curious if it will ever finish and was curious about what the credits might look like, but the task took too long.

The Rosetta BOINC status says the task has: "timed out - no response". Task id is: aagb-FPS_pp-mNMPHE-mACPenC13T-B3LEU_pp_3_2601503_6_0

As Sid says, if the time needed for a "Rosetta python projects 1.03 (vbox64)" is much more than is expected, stop the task.

I'm rarely confident of anything regarding computers, but because I work away 2-4 consecutive days every week I routinely come back to one or two tasks in progress for whole days.
With one solitary exception, they show reduced CPU runtimes of 1 or 2hrs down to as little as 20-something seconds. Not one ever restarts once stalled
ID: 106382 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Peter Hucker
Avatar

Send message
Joined: 12 Aug 06
Posts: 1479
Credit: 7,138,291
RAC: 10,433
Message 106385 - Posted: 12 Jun 2022, 3:08:05 UTC

You might be able to leave it running more than the number of cores you have, then if some stall your PC is still flat out while you're away. Not sure if you can do this for tasks designed for a single core, but I do it with multicore Cosmology tasks:

<app_config>
   <app_version>
       <app_name>camb_boinc2docker</app_name>
       <plan_class>vbox64_mt</plan_class>
       <avg_ncpus>5</avg_ncpus>
       <cmdline>--nthreads 6</cmdline>
   </app_version>
</app_config>


This tells the actual app to use 6 cores per task, but the Boinc scheduler they only use 5.

Or perhaps there's another way to tell Boinc you have more cores than you really do?
ID: 106385 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Greg_BE
Avatar

Send message
Joined: 30 May 06
Posts: 5652
Credit: 5,622,096
RAC: 87
Message 106390 - Posted: 15 Jun 2022, 6:46:18 UTC

I just got back from vacation and restarted my system. I didn't check it and today I did and I see this stupid project dropped me from VM jobs until I reset it again a bit ago. This project really enjoys kicking people off its VM stuff.
ID: 106390 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Previous · 1 . . . 214 · 215 · 216 · 217 · 218 · 219 · 220 . . . 257 · Next

Message boards : Number crunching : Problems and Technical Issues with Rosetta@home



©2022 University of Washington
https://www.bakerlab.org