Welcome to MilkyWay@home

Server Trouble

Message boards : News : Server Trouble
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 12 · 13 · 14 · 15 · 16 · 17 · 18 . . . 22 · Next

AuthorMessage
Septimus

Send message
Joined: 8 Nov 11
Posts: 205
Credit: 2,882,853
RAC: 272
Message 72749 - Posted: 13 Apr 2022, 9:58:17 UTC - in response to Message 72744.  
Last modified: 13 Apr 2022, 10:00:50 UTC

I may be jumping to the wrong conclusion but I would say no NBODY Simulations are getting validated. My Validation Inconclusive Q is hundreds long, I am sure many will have thousands. I did a random check on every page of my backlog and they all show my task as inconclusive, a second task has a number but shows as unsent. The NBODY q of simulation tasks is always around 13.8 million yet at any one time over half a million are being processed, surely with that sort of volume it should go down. The number of tasks waiting validation was down to 554 when it last looked so it cant contain any Simulation tasks as that would be a significant number. Have stopped processing them, only doing separation. Separation tasks currently are getting validated within seconds of completion now so that is back to normal.
ID: 72749 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile mikey
Avatar

Send message
Joined: 8 May 09
Posts: 3315
Credit: 519,943,543
RAC: 22,328
Message 72751 - Posted: 13 Apr 2022, 10:35:57 UTC - in response to Message 72744.  



On a related topic, I would suggest that someone writes a script that checks the number of tasks waiting to go out for each project and turns off the work generator for a project if it exceeds some pre-chosen limit, turning it back on again when the number of tasks falls below a lower limit (chosen to avoid constant stop-start) - that could then be a cron job (or equivalent) running (say) every 15 or 30 minutes. It won't solve the current problem but it would mean a recurrence could be avoided without the need for constant supervision! (I presume there isn't such a capability built into the generators already...)


That should be a main part of the boinc server software. Surely there's a setting to say what size each queue should be? Who writes this crap?


In another post Tom said it was a setting and they are using it now.
ID: 72751 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Keith Myers
Avatar

Send message
Joined: 24 Jan 11
Posts: 696
Credit: 540,021,725
RAC: 86,722
Message 72764 - Posted: 13 Apr 2022, 19:32:52 UTC

BOINC server software has a default size for the feeder buffer. It can be configured larger if necessary if it is getting emptied too fast on every connection. But that swells the database and the database tracking of sent/returned/awaiting validation/deletion transactions take longer times.

So you can't dramatically increase the size of the feeder buffer willy-nilly without slowing down other parts of the server.

It's all a balancing act depending on the hardware and scheduler connections.
ID: 72764 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,142,956
RAC: 2
Message 72772 - Posted: 14 Apr 2022, 6:22:13 UTC - in response to Message 72764.  

BOINC server software has a default size for the feeder buffer. It can be configured larger if necessary if it is getting emptied too fast on every connection. But that swells the database and the database tracking of sent/returned/awaiting validation/deletion transactions take longer times.

So you can't dramatically increase the size of the feeder buffer willy-nilly without slowing down other parts of the server.

It's all a balancing act depending on the hardware and scheduler connections.
Something bad was happening in MW, since every time there was a problem, instead of the usual 1000/10000 tasks, it jumped to several million of each. Yet another Boinc bug I'd guess.
ID: 72772 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Bill F
Avatar

Send message
Joined: 4 Jul 09
Posts: 85
Credit: 16,692,188
RAC: 4,497
Message 72807 - Posted: 15 Apr 2022, 4:11:11 UTC - in response to Message 72772.  

BOINC server software has a default size for the feeder buffer. It can be configured larger if necessary if it is getting emptied too fast on every connection. But that swells the database and the database tracking of sent/returned/awaiting validation/deletion transactions take longer times.

So you can't dramatically increase the size of the feeder buffer willy-nilly without slowing down other parts of the server.

It's all a balancing act depending on the hardware and scheduler connections.
Something bad was happening in MW, since every time there was a problem, instead of the usual 1000/10000 tasks, it jumped to several million of each. Yet another Boinc bug I'd guess.
project forum.


Peter you continually bash the BOINC software on this forum. The Milkyway project does not write, maintain or modify the basic BOINC software for Clients or Servers. This is all controlled and developed by the BOINC organization out of Berkely Ca.

Here is a URL to their forum I politely suggest that you redirect your BOINC suggestions and criticisms to some where they can be addressed. They serve no purpose here. Create yourself an account and do some studying.

https://boinc.berkeley.edu/forum_index.php

Sincerely
Bill F
In October of 1969 I took an oath to support and defend the Constitution of the United States against all enemies, foreign and domestic;
There was no expiration date.


ID: 72807 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,142,956
RAC: 2
Message 72812 - Posted: 15 Apr 2022, 6:20:31 UTC - in response to Message 72807.  
Last modified: 15 Apr 2022, 6:21:02 UTC

Peter you continually bash the BOINC software on this forum. The Milkyway project does not write, maintain or modify the basic BOINC software for Clients or Servers. This is all controlled and developed by the BOINC organization out of Berkely Ca.
I never suggested it was the fault of MW, contrarywise, I'm saying it's someone else's fault and not this project.

Here is a URL to their forum I politely suggest that you redirect your BOINC suggestions and criticisms to some where they can be addressed. They serve no purpose here. Create yourself an account and do some studying.

https://boinc.berkeley.edu/forum_index.php
Between my 6 accounts there I've been banned 38 times. I'm even ruder there. Their rubbish programmers can't handle criticism.
ID: 72812 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Ralph Little

Send message
Joined: 30 Jul 16
Posts: 6
Credit: 58,669,504
RAC: 0
Message 72864 - Posted: 16 Apr 2022, 0:09:58 UTC

Things have been much better over the last couple of days.
Getting plenty of GPU separation work now. :)
ID: 72864 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,142,956
RAC: 2
Message 72873 - Posted: 16 Apr 2022, 9:12:50 UTC - in response to Message 72864.  

Things have been much better over the last couple of days.
Getting plenty of GPU separation work now. :)
Maybe he's overvolting the server? I did that successfully with my electric toothbrush and unsuccessfully with my electric razor.
ID: 72873 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
San-Fernando-Valley

Send message
Joined: 13 Apr 17
Posts: 256
Credit: 604,411,638
RAC: 0
Message 72874 - Posted: 16 Apr 2022, 9:25:18 UTC - in response to Message 72873.  

... and unsuccessfully with my electric razor.

... did you forget to turn it on ?
ID: 72874 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,142,956
RAC: 2
Message 72875 - Posted: 16 Apr 2022, 9:35:38 UTC - in response to Message 72874.  

... and unsuccessfully with my electric razor.
... did you forget to turn it on ?
It damaged the bearings in the motor, it still works but with a terrible whining noise.
ID: 72875 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
San-Fernando-Valley

Send message
Joined: 13 Apr 17
Posts: 256
Credit: 604,411,638
RAC: 0
Message 72876 - Posted: 16 Apr 2022, 9:36:54 UTC - in response to Message 72875.  

... and unsuccessfully with my electric razor.
... did you forget to turn it on ?
It damaged the bearings in the motor, it still works but with a terrible whining noise.

... poor thing ...
ID: 72876 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Fabio

Send message
Joined: 9 Jul 20
Posts: 1
Credit: 34,582,247
RAC: 0
Message 72883 - Posted: 16 Apr 2022, 10:58:56 UTC - in response to Message 72749.  

same for me, only separation results are validated and the list of "validation inconclusive" it getting longer. And there are only N-Body tasks added to that list during the last two weeks. I only do separation tasks from now on, too.
ID: 72883 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,142,956
RAC: 2
Message 72884 - Posted: 16 Apr 2022, 11:07:45 UTC - in response to Message 72883.  

same for me, only separation results are validated and the list of "validation inconclusive" it getting longer. And there are only N-Body tasks added to that list during the last two weeks. I only do separation tasks from now on, too.
Yesterday I had 5 of 1587 validated Nbody, today I have 354 of 2179. They're going through.
ID: 72884 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
San-Fernando-Valley

Send message
Joined: 13 Apr 17
Posts: 256
Credit: 604,411,638
RAC: 0
Message 72885 - Posted: 16 Apr 2022, 11:13:48 UTC - in response to Message 72883.  

... I only do separation tasks from now on, too.

N-Body is doing fine now.

The idea, at this point of the game, is to try to work through the N-Body queue as fast as possible in order to
get things back to normal.
This has been described and talked about in many previous posts.

As you might have noticed, the number of crunchers doing N-Body tasks has gone up -
from around 100 to now over 2300 active users.
They are all trying to help.
ID: 72885 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Septimus

Send message
Joined: 8 Nov 11
Posts: 205
Credit: 2,882,853
RAC: 272
Message 72886 - Posted: 16 Apr 2022, 11:46:41 UTC - in response to Message 72885.  

... I only do separation tasks from now on, too.

N-Body is doing fine now.

The idea, at this point of the game, is to try to work through the N-Body queue as fast as possible in order to
get things back to normal.
This has been described and talked about in many previous posts.

As you might have noticed, the number of crunchers doing N-Body tasks has gone up -
from around 100 to now over 2300 active users.
They are all trying to help.


I had over 100 validated this morning.
ID: 72886 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
San-Fernando-Valley

Send message
Joined: 13 Apr 17
Posts: 256
Credit: 604,411,638
RAC: 0
Message 72887 - Posted: 16 Apr 2022, 12:02:16 UTC - in response to Message 72886.  

... same here ...
ID: 72887 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Kiska

Send message
Joined: 31 Mar 12
Posts: 94
Credit: 151,910,410
RAC: 12,497
Message 72888 - Posted: 16 Apr 2022, 12:02:56 UTC

ID: 72888 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile HRFMguy

Send message
Joined: 12 Nov 21
Posts: 236
Credit: 575,029,129
RAC: 37,181
Message 72905 - Posted: 16 Apr 2022, 20:27:23 UTC - in response to Message 72595.  

Regarding your five Radeon R9 280X cards, how do you like them? Sounds like you are pretty well pleased.
Love them. One of them I play games on.

R9 280x is ready to be installed. Two fans say Tri-X on them, third one says TOXIC. Time to bust out the Nitrile gloves! LOL. Peter, (or anybody) do you have a manual for this? Radeon help desk says no specific unit manuals, only generic. I guess my question is, what do the 6 LEDs on the back indicate? Also, what is the lifespan of the thermal paste? Probably should be changed anyway. This thing is a beast. Any specific guidance? The heat sinks and fans look squeaky clean, no dust at all.

OK, cover me! I'm goin' in! LOL
ID: 72905 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,142,956
RAC: 2
Message 72906 - Posted: 16 Apr 2022, 20:31:26 UTC - in response to Message 72905.  

Regarding your five Radeon R9 280X cards, how do you like them? Sounds like you are pretty well pleased.
Love them. One of them I play games on.

R9 280x is ready to be installed. Two fans say Tri-X on them, third one says TOXIC. Time to bust out the Nitrile gloves! LOL. Peter, (or anybody) do you have a manual for this? Radeon help desk says no specific unit manuals, only generic. I guess my question is, what do the 6 LEDs on the back indicate? Also, what is the lifespan of the thermal paste? Probably should be changed anyway. This thing is a beast. Any specific guidance? The heat sinks and fans look squeaky clean, no dust at all.

OK, cover me! I'm goin' in! LOL
The LEDs are temperature I think. I just monitor temperature with MSI Afterburner.

Thermal paste? Forever, I've never replaced it.
ID: 72906 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Tom Donlon
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 10 Apr 19
Posts: 408
Credit: 120,203,200
RAC: 0
Message 72917 - Posted: 17 Apr 2022, 14:40:16 UTC

https://grafana.kiska.pw/d/boinc/boinc?orgId=1&var-project=milkyway@home&from=now-72h&to=now&chunkNotFound=&refresh=1m


This is awesome. Thanks Kiska.
ID: 72917 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Previous · 1 . . . 12 · 13 · 14 · 15 · 16 · 17 · 18 . . . 22 · Next

Message boards : News : Server Trouble

©2024 Astroinformatics Group