World Community Grid - View Thread

World Community Grid Forums

Category: Support

Forum: BOINC Agent Support

Thread: An OSX bug

Quick Go »

No member browsing this thread

Thread Status: Active
Total posts in this thread: 146

[ ]

Author

This topic has been viewed 12164 times and has 145 replies

Sekerob
Ace Cruncher
Joined: Jul 24, 2005
Post Count: 20043
Status: Offline


Re: An OSX bug

Actually in the world of Windows a service/daemon install still allows to launch the BM i.e. no need of the CL to perform the management. When done, simply exit BM to unload it and remove it as potential source of the hang, hung, hunged.

----------------------------------------

WCG

Global & Research > Make Proposal Help: Start Here!
Please help to make the Forums an enjoyable experience for All!

[Mar 24, 2008 3:02:20 PM]

Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline


Re: An OSX bug

I realize that. Most of my crunching PCs (dozens) are unfortunately running windows. I run boinc as a service on ALL (yes, all) of them. With the GUI install of the Mac (I only have one Mac) client, you can't shut down the Boinc Manager without stopping your crunching (anyone correct me if I'm wrong). Thats why I asked the question if any one else was running the CL version.

As I was writing this my dddt work unit locked up at 1.2xx% and had to stop and restart it. The dddt project is clearly broken for linux and mac and should be pulled until fixed.

[Mar 25, 2008 4:25:42 AM]

Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline


Re: An OSX bug

I'm pleased to report that the OSX bug is non-existent with the HCC project anymore. I've not had one stall since I switched to HCC only. It would appear that in fixing the page fault problems they've fixed the OSX bug.

I hope they might be able to examine the other projects and apply the same mojo to them as well.

-SS

Now I wish they would make HCC available for PPC Mac. As of now only DDDT and FAAH are available for PPC mac and they are not working properly.

[Mar 25, 2008 4:50:14 AM]

Bearcat
Master Cruncher
USA
Joined: Jan 6, 2007
Post Count: 2803
Status: Offline
Project Badges:

5 year badge for Human Proteome Folding - Phase 2

1 year badge for Discovering Dengue Drugs - Together

2 year badge for Nutritious Rice for the World

180 day badge for The Clean Energy Project

10 year badge for Help Fight Childhood Cancer

1 year badge for Influenza Antiviral Drug Search

5 year badge for Help Cure Muscular Dystrophy - Phase 2

45 day badge for Discovering Dengue Drugs - Together - Phase 2

5 year badge for The Clean Energy Project - Phase 2

5 year badge for Computing for Clean Water

5 year badge for Drug Search for Leishmaniasis

5 year badge for GO Fight Against Malaria

5 year badge for Computing for Sustainable Water

20 year badge for Mapping Cancer Markers

10 year badge for Uncovering Genome Mysteries

10 year badge for Outsmart Ebola Together

20 year badge for FightAIDS@Home - Phase 2

20 year badge for Smash Childhood Cancer

20 year badge for Microbiome Immunity Project

1 year badge for Africa Rainfall Project

20 year badge for OpenPandemics - COVID-19


Re: An OSX bug

Had 1 FAAH unit freeze up yesterday. Had to suspend boinc then restart it. So far things seem ok with Leopard. Hopefully this doesn't happen to often.

----------------------------------------

Crunching for humanity since 2007!

[Mar 30, 2008 3:01:47 AM]

Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline


Re: An OSX bug

Just wanted to note that I am suffering from this problem in a major way. In the last week since I've started crunching on a brand new 24" iMac I have had to intervene several times to abort stuck work units. I'm not entirely sure if any particular project is the culprit, but I have definitely noticed several stuck HPF2 workunits, and the occasional DDDT one.

My machine has wasted at least 2-3 days of CPU time in the last week because of this... work units appear stuck in BOINC, but in the Activity Monitor it shows that the process is using ~100% CPU allocation, and that the process has been running for way longer than BOINC claims. Also, I end up getting three work units running (the stuck one taking up entirely one core), and two other work units running on the other core at 50% each.

I'll post more details if anyone wants them.

[Apr 11, 2008 3:52:09 PM]

Sekerob
Ace Cruncher
Joined: Jul 24, 2005
Post Count: 20043
Status: Offline


Re: An OSX bug

There are beta's in preparation (see beta forum and Viktors posts) for DDDT and FAAH using a new BOINC API and also to make them ready for BOINC 6. Why one or the other machine does this it's not known. Think you reported before on your brand new cruncher this/other thread ? having issues but can't recollect presently what the OS was exactly and client.

so, is HCC running without hitch?

----------------------------------------

WCG

Global & Research > Make Proposal Help: Start Here!
Please help to make the Forums an enjoyable experience for All!

[Apr 11, 2008 3:59:23 PM]

Bearcat
Master Cruncher
USA
Joined: Jan 6, 2007
Post Count: 2803
Status: Offline
Project Badges:


Re: An OSX bug

How much ram do you have installed? What version of boinc are you running? Do you leave the connection open all the time? Would suggest try running 1 core only. Any other programs running in the backgroud (ie. widgets)?

----------------------------------------

Crunching for humanity since 2007!

[Apr 12, 2008 1:49:03 AM]

Bearcat
Master Cruncher
USA
Joined: Jan 6, 2007
Post Count: 2803
Status: Offline
Project Badges:


Re: An OSX bug

Forgot to ask this.. in energy saver, do you have the disk to sleep after a period of time?

----------------------------------------

Crunching for humanity since 2007!

[Apr 12, 2008 1:50:42 AM]

Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline


Re: An OSX bug

Hi Sekerob,

Cheers for the info. The other thread was about HPF2 units on my iMac seeming to take longer than they should, so it was claiming say 150 BOINC points, and getting about 120. If you look at the list of what other people we claiming/getting on the same work unit, very few others had such a disparity. It wasn't a serious problem, unlike this one.

Based on spec in comparison to other crunchers I have, my iMac should be making about 6000 WCG points a day. This problem (and slightly also the one from the other thread I mention above) means that I'm getting more like 3000 or less.

so, is HCC running without hitch?

I don't remember aborting an HCC work unit yet. I'll let you know if I do. I've just had to cancel a stuck dddt work unit that has been running for 22 hours of CPU time. Nearly an entire day on one core wasted with no points.

[Apr 12, 2008 10:58:42 AM]

Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline


Re: An OSX bug

How much ram do you have installed?

4 GiB.

What version of boinc are you running?

5.10.45

Do you leave the connection open all the time?

Yes.

Would suggest try running 1 core only. Any other programs running in the backgroud (ie. widgets)?

I can try it, but I'll be very surprised if it makes any difference. Of course, there is other stuff running, but only at a very low level as I'm not at work right now:


Processes:  66 total, 5 running, 62 sleeping... 229 threads     
  PID COMMAND      %CPU   TIME   #TH #PRTS #MREGS RPRVT  RSHRD  RSIZE  VSIZE
79696 wcg_hcc1_i  97.4% 73:26.76   3    20    156   38M- 1380K    40M+   71M+
42763 wcg_faah_a  91.5%  8:36:05   3    20     49   96M  1380K    99M   307M 
82485 top          3.7%  0:06.70   1    19     34  748K   200K  1340K-   19M
  220 Dock         1.4% 62:15.14   3   150    298 3760K    29M    12M   348M
  250 pmTool       1.4% 85:22.54   1    28     34  612K  4896K  1612K    28M
  237 Activity M   0.8% 56:44.37   6   116    308 6236K    44M    16M   412M
   92 WindowServ   0.6% 35:46.38   5   283    861 5216K    64M    60M   399M
.
etc
.

And when I'm actually at work and working on it, obviously there is more stuff running.

----------------------------------------
[Edit 2 times, last edit by Former Member at Apr 12, 2008 11:11:08 AM]

[Apr 12, 2008 11:09:30 AM]

[ ]