Big change in GoofyxGrid@Home

Message boards : News : Big change in GoofyxGrid@Home

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · Next

AuthorMessage
Dirk Broer

Send message
Joined: 14 Nov 15
Posts: 103
Credit: 4,317,480
RAC: 6,390
Message 1705 - Posted: 15 Apr 2017, 9:52:30 UTC - in response to Message 1703.  
Last modified: 15 Apr 2017, 9:55:16 UTC

About Enigma@Home - I know from his admin that he is make some serious changes to daemons in project which is reason for this pause in work unit generation.



And about uncommunicative BOINC project admins: You don't answer the forum on Radioactive@Home either. The validator doesn't do a thing over there.
ID: 1705 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Dirk Broer

Send message
Joined: 14 Nov 15
Posts: 103
Credit: 4,317,480
RAC: 6,390
Message 1706 - Posted: 15 Apr 2017, 12:10:18 UTC - in response to Message 1704.  

About Enigma@Home - I know from his admin that he is make some serious changes to daemons in project which is reason for this pause in work unit generation.


So far he not very communicative about it. There's only mentioning of server maintenance on the 18th.


And that was concerning server maintenance in 2013....
ID: 1706 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Steve Hawker*

Send message
Joined: 10 Nov 15
Posts: 68
Credit: 1,699,860
RAC: 1,268
Message 1708 - Posted: 15 Apr 2017, 14:55:45 UTC
Last modified: 15 Apr 2017, 15:00:17 UTC

Gentlemen,

First I want you to know that I do appreciate the effort required to run a project.

But, please consider what people are seeing from BOINC@Poland:

1. Enigma grinds to a halt without any communication.
2. Radioactive was vibrant but now looks like going the way of QCN: no support, no communication and no-one to keep the thing running. Not to mention the disastrous kits. No OS X support.
3. DrugDiscovery@Home. Running one day. Completely dead the next. Lots of "he said this, she said that" on the forum, users left high and dry. No OS X support.
4. Goofyxgrid. Warning of project transition in 1-2 weeks. Work generation stops immediately. No response on that. OS X support ceased.
5. Universe runs WUs for a short time then that application is never seen again. Work runs out constantly. No OS X support.

We need consistency and communication. Consider PrimeGrid which is the gold standard for BOINC. Regular, active communication from admins. Minimum 30 day notice if a subproject is going to end AND that schedule is rigidly adhered to. Work is almost always available for all subprojects.

Please, if you get this GCC running, please, please, please be more like PrimeGrid and remember that credits and badges aren't the only things that people are chasing and for some, that's not even a consideration. Consistency and communication is the glue that keeps it all working.

Thank you.
ID: 1708 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
ChertseyAl

Send message
Joined: 2 Nov 15
Posts: 1
Credit: 266,710
RAC: 1,640
Message 1709 - Posted: 16 Apr 2017, 12:02:09 UTC - in response to Message 1703.  

Firstly yes, Goofyx will move project to GCC (http://gridcomputingcenter.org/).
Secondary - yes, it is still BOINC :)

About Enigma@Home - I know from his admin that he is make some serious changes to daemons in project which is reason for this pause in work unit generation.


ID: 1709 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Chris Skull

Send message
Joined: 28 Jan 17
Posts: 9
Credit: 715,120
RAC: 1,693
Message 1711 - Posted: 17 Apr 2017, 8:00:50 UTC

I think they try to build a commercial company where scientists have to pay for setting up a Boinc project.
On other projects i see scientists, students and maybe professors... but here we have managers and CEO's and contracts... that are terms of commercial company's.
However... commercial company's need clients, without communication you will not get many of them. ;-)
Greetz,
Chris

ID: 1711 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Steve Hawker*

Send message
Joined: 10 Nov 15
Posts: 68
Credit: 1,699,860
RAC: 1,268
Message 1712 - Posted: 17 Apr 2017, 15:03:00 UTC - in response to Message 1711.  

I think they try to build a commercial company where scientists have to pay for setting up a Boinc project.
On other projects i see scientists, students and maybe professors... but here we have managers and CEO's and contracts... that are terms of commercial company's.
However... commercial company's need clients, without communication you will not get many of them. ;-)


Charging money to set up and administer projects is one thing. Charging people for our resources is altogether something I want no part of.

The difficulty is knowing what is going on.
ID: 1712 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile bcavnaugh
Avatar

Send message
Joined: 8 May 16
Posts: 13
Credit: 766,550
RAC: 1,031
Message 1713 - Posted: 17 Apr 2017, 23:50:17 UTC - in response to Message 1708.  

Gentlemen,

First I want you to know that I do appreciate the effort required to run a project.

But, please consider what people are seeing from BOINC@Poland:

1. Enigma grinds to a halt without any communication.
2. Radioactive was vibrant but now looks like going the way of QCN: no support, no communication and no-one to keep the thing running. Not to mention the disastrous kits. No OS X support.
3. DrugDiscovery@Home. Running one day. Completely dead the next. Lots of "he said this, she said that" on the forum, users left high and dry. No OS X support.
4. Goofyxgrid. Warning of project transition in 1-2 weeks. Work generation stops immediately. No response on that. OS X support ceased.
5. Universe runs WUs for a short time then that application is never seen again. Work runs out constantly. No OS X support.

We need consistency and communication. Consider PrimeGrid which is the gold standard for BOINC. Regular, active communication from admins. Minimum 30 day notice if a subproject is going to end AND that schedule is rigidly adhered to. Work is almost always available for all subprojects.

Please, if you get this GCC running, please, please, please be more like PrimeGrid and remember that credits and badges aren't the only things that people are chasing and for some, that's not even a consideration. Consistency and communication is the glue that keeps it all working.

Thank you.


You Forgot MindModeling@Home Beta
Funny that you can still run tasks but cannot logon cannot contact anyone cannot post on the Forum no one to email to ask.
Crunching@EVGA The Number One Team in the BOINC Community.\r\nFolding@EVGA The Number One Team in the Folding@Home Community
ID: 1713 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Dirk Broer

Send message
Joined: 14 Nov 15
Posts: 103
Credit: 4,317,480
RAC: 6,390
Message 1714 - Posted: 18 Apr 2017, 0:11:56 UTC - in response to Message 1713.  

MindModeling@Home (Beta) is based in Dayton, Ohio at the University of Dayton Research Institute and Wright State University. It has nothing to do with the projects that are supported by/through BOINC@Poland and that are to be incorporated in GCC.
I tested tonight and could both login and post on their (MindModeling) forum.
ID: 1714 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Dirk Broer

Send message
Joined: 14 Nov 15
Posts: 103
Credit: 4,317,480
RAC: 6,390
Message 1715 - Posted: 18 Apr 2017, 20:56:57 UTC

The good news about this transition-to-be is that at least Radioactive@Home and Universe@Home are running alright now, and that the Universe@Home forum does its best to keep the volunteers informed.
As it is run by the same krzyszp that also takes the honours here when Goofyx is not around, we're glad to see that efforts are being made to get things going.
ID: 1715 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile [BOINC@Poland]goofyx
Project administrator
Project developer
Project scientist
Help desk expert

Send message
Joined: 8 Nov 15
Posts: 554
Credit: 9,507,504
RAC: 7,812
Message 1716 - Posted: 9 May 2017, 9:54:25 UTC - in response to Message 1715.  
Last modified: 9 May 2017, 9:56:51 UTC

I would like to apologize to everyone.
For the last month, I spent more time at the hospital than at home, which is why I delayed running GCC servers. I hope everything catch up ASAP.
My plan for the Monkeys is as follows:
1. Today and tomorrow I run NCI monkeys
2. Then the CPU monkeys <- platforms : windows 64, linux 64, arm 32 (odroid-xu4)
ID: 1716 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Sergey Kovalchuk

Send message
Joined: 25 Dec 15
Posts: 38
Credit: 3,518,590
RAC: 1,806
Message 1717 - Posted: 9 May 2017, 10:01:30 UTC - in response to Message 1716.  
Last modified: 9 May 2017, 10:03:27 UTC

Received v1 & v3, but no v2 & v4

Problem with 32 bit apps on 64 bit OS
ID: 1717 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Skivelitis2
Avatar

Send message
Joined: 12 Nov 15
Posts: 49
Credit: 3,368,000
RAC: 12,092
Message 1718 - Posted: 9 May 2017, 11:51:10 UTC
Last modified: 9 May 2017, 12:01:24 UTC

Apps not flagged nci and are taking up a full core between them (v1 and v3). Both Win and Linux.
Edit: This behavior is intermittent, occurring on and off throughout the run.
ID: 1718 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile [BOINC@Poland]goofyx
Project administrator
Project developer
Project scientist
Help desk expert

Send message
Joined: 8 Nov 15
Posts: 554
Credit: 9,507,504
RAC: 7,812
Message 1719 - Posted: 9 May 2017, 21:41:06 UTC - in response to Message 1717.  
Last modified: 9 May 2017, 22:13:38 UTC

Received v1 & v3, but no v2 & v4

All NCI apps v1,v2,v3,v4 works
ID: 1719 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
mmonnin

Send message
Joined: 28 Oct 16
Posts: 28
Credit: 4,740,400
RAC: 31,987
Message 1720 - Posted: 10 May 2017, 0:45:28 UTC - in response to Message 1719.  
Last modified: 10 May 2017, 0:46:44 UTC

Received v1 & v3, but no v2 & v4

All NCI apps v1,v2,v3,v4 works


Not quite. After 1st downloading all 4 tasks abort the 1st time. The next set take at least 1 CPU core. Sometimes I've seen it take 4 tasks away from a CPU. Pause all GCC tasks and the CPU work comes back. Resume GCC and they both run together with GCC as NCI.

I saw this on 3 separate Linux machines. On my windows machine it took 3 pause/resumes to get all CPU tasks running with GCC. I'm not sure what will happen in one hour when the next come. They might just keep the CPU if a user doesn't interrupt.

They are NOT labled as NCI in BM (besides the app name). They do not behave as NCI should. Whatever code has changed when the CPU app was introduced needs to reverted.

I'm suspending GCC on all computers until its fixed.
ID: 1720 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Dr Who Fan
Avatar

Send message
Joined: 4 Dec 15
Posts: 104
Credit: 1,273,380
RAC: 3,008
Message 1721 - Posted: 10 May 2017, 4:30:46 UTC - in response to Message 1720.  
Last modified: 10 May 2017, 4:31:16 UTC


..........
Not quite. After 1st downloading all 4 tasks abort the 1st time. The next set take at least 1 CPU core. Sometimes I've seen it take 4 tasks away from a CPU. Pause all GCC tasks and the CPU work comes back. Resume GCC and they both run together with GCC as NCI.

I saw this on 3 separate Linux machines. On my windows machine it took 3 pause/resumes to get all CPU tasks running with GCC. I'm not sure what will happen in one hour when the next come. They might just keep the CPU if a user doesn't interrupt.

They are NOT labled as NCI in BM (besides the app name). They do not behave as NCI should. Whatever code has changed when the CPU app was introduced needs to reverted.

I'm suspending GCC on all computers until its fixed.


SAME HERE across 5 computers !!!
STOPPED RECEIVING MORE TASKS UNTIL PROJECT FIXED

ID: 1721 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
JugNut

Send message
Joined: 25 Apr 16
Posts: 23
Credit: 5,530,670
RAC: 60,299
Message 1722 - Posted: 10 May 2017, 6:43:51 UTC - in response to Message 1719.  

Hi good to see the GoofyX team back in business :)

But....
1 Could you please give the validator a kick!! Many of us(myself included) have completed plenty of the latest tasks but so far none have validated, they are all stuck in "Validation pending"..

2 As mentioned these tasks are "not" running as NCI.
One solution is to abort the tasks that are not running in nci mode in the hope that the next lot of tasks received will work as planned. Strangely this does work sometimes but it's only a temporary fix at best.

A better solution that was mentioned once was to use an app_config like this one....

app_config>
<app_version>
<app_name>monkeys_v1</app_name>
<avg_ncpus>0.000001</avg_ncpus>
</app_version>
<app_version>
<app_name>monkeys_v2</app_name>
<avg_ncpus>0.000001</avg_ncpus>
</app_version>
<app_version>
<app_name>monkeys_v3</app_name>
<avg_ncpus>0.000001</avg_ncpus>
</app_version>
<app_version>
<app_name>monkeys_v4</app_name>
<avg_ncpus>0.000001</avg_ncpus>
</app_version>
</app_config>
(+1 to the guy who originally suggested it:)

This above solution works well & after it's installed you'll won't have to worry about this problem again...
Err well at least until a new monkeys apps is released, Then just update it with the new apps name/s.

Of course it should go without saying that the best solution is a real fix from our friendly dev's. But hey what do I know? :)

Crunch-on..
ID: 1722 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Sergey Kovalchuk

Send message
Joined: 25 Dec 15
Posts: 38
Credit: 3,518,590
RAC: 1,806
Message 1723 - Posted: 10 May 2017, 7:48:15 UTC - in response to Message 1722.  

This above solution works well & after it's installed you'll won't have to worry about this problem again...
Err well at least until a new monkeys apps is released, Then just update it with the new apps name/s.

Do not set <avg_ncpus> options for new CPU applications
You will be crowded with a million monkeys
ID: 1723 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Ivan

Send message
Joined: 24 Dec 15
Posts: 10
Credit: 503,580
RAC: 721
Message 1724 - Posted: 10 May 2017, 10:07:41 UTC

The tasks are not running as NCI on my huawey P8
ID: 1724 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
skgiven

Send message
Joined: 11 Aug 16
Posts: 1
Credit: 1,192,310
RAC: 3,399
Message 1726 - Posted: 10 May 2017, 11:19:57 UTC - in response to Message 1724.  
Last modified: 10 May 2017, 11:24:00 UTC

Things to do:

Modify the NCI apps to instruct BM that they are NCI and don't use the CPU.
At present Boinc thinks these use a full CPU when in fact they don't. Problem is that stops the CPU being used for other work.
Temporary fix is to accept only NCI work from here (deselecting the CPU work, just in case any work turns up for these apps) and use the app_config.xml below.

Stop sending X86 work to X64 Linux rigs; the tasks all fail.
ID: 1726 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Conan

Send message
Joined: 9 Nov 15
Posts: 63
Credit: 1,659,603
RAC: 4,200
Message 1728 - Posted: 10 May 2017, 11:32:37 UTC

It is a pity that our host totals were not carried over and have been wiped, we start from scratch, at least the overall total came across.
Least of the projects worries I suppose, get all the errors fixed first.

Conan
ID: 1728 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Previous · 1 · 2 · 3 · 4 · Next

Message boards : News : Big change in GoofyxGrid@Home


©2017 Polish BOINC Foundation