I think I've made too many news threads. I've updated the scheduler one more time, I think I might have tracked down the issue. Let me know if you're still getting bad workunits or not getting workunits.

---

LOL...

Well, you can start as many new news threads as you like, as long as progress is being made! ;-)

So far nothing seems to have gotten broken again, but it may take awhile for another CPU request slot to come around on my single core CPU hosts. Some are pretty slow, and if you miss the chance to pull work when your turn comes up you have to go through the round robin list of projects again.

OTOH, since we have new nBody to chew I enabled CPU work on my big gun battleships again, so I should be able to tell if CPU MW tasks even have a chance of making it to the old timers.

As you can see by event log in the first few lines I requested tasks and it says that I don't need any.
There are none on my list so I closed down my malariacontrol tasks to force milkyway to send me tasks and now at the end few lines it is saying no tasks available!
Which is right?
All it wants to do is defer communication!


12/06/2013 03:55:53 | Milkyway@Home | update requested by user
12/06/2013 03:55:58 | Milkyway@Home | Sending scheduler request: Requested by user.
12/06/2013 03:55:58 | Milkyway@Home | Not requesting tasks: don't need
12/06/2013 03:56:00 | Milkyway@Home | Scheduler request completed
12/06/2013 03:58:45 | malariacontrol.net | task wu_3001_510_1131389_0_1370573890_1 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_2899_517_1136821_0_1370973915_0 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_3152_173_265705_0_1370972948_0 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_3001_511_1131389_0_1370573890_1 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_3255_514_1136823_0_1370974158_0 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_2903_515_1136808_0_1370972985_0 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_3255_509_1136823_0_1370974158_0 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_3255_510_1136823_0_1370974158_0 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_1068_414_1136822_0_1370974150_0 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_3255_504_1136823_0_1370974158_0 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_3255_506_1136823_0_1370974158_0 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_1068_411_1136822_0_1370974150_0 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_3300_317_1136920_0_1370980038_0 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_1068_501_1137031_0_1370987890_1 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_3300_505_1131567_0_1370587277_2 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_2903_524_1137030_0_1370987681_1 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_3300_506_1131567_0_1370587277_2 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_2903_522_1131615_0_1370591021_4 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_3301_518_1131618_0_1370591360_2 suspended by user
12/06/2013 03:58:46 | Milkyway@Home | Sending scheduler request: To fetch work.
12/06/2013 03:58:46 | Milkyway@Home | Requesting new tasks for CPU
12/06/2013 03:58:48 | Milkyway@Home | Scheduler request completed: got 0 new tasks
12/06/2013 03:58:48 | Milkyway@Home | No tasks sent
12/06/2013 03:58:48 | Milkyway@Home | No tasks are available for MilkyWay@Home
12/06/2013 03:58:48 | Milkyway@Home | No tasks are available for Milkyway@Home Separation
12/06/2013 03:58:48 | Milkyway@Home | No tasks are available for the applications you have selected.

As you can see by event log in the first few lines I requested tasks and it says that I don't need any.
There are none on my list so I closed down my malariacontrol tasks to force milkyway to send me tasks and now at the end few lines it is saying no tasks available!
Which is right?
All it wants to do is defer communication!


12/06/2013 03:55:53 | Milkyway@Home | update requested by user
12/06/2013 03:55:58 | Milkyway@Home | Sending scheduler request: Requested by user.
12/06/2013 03:55:58 | Milkyway@Home | Not requesting tasks: don't need
12/06/2013 03:56:00 | Milkyway@Home | Scheduler request completed
12/06/2013 03:58:45 | malariacontrol.net | task wu_3001_510_1131389_0_1370573890_1 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_2899_517_1136821_0_1370973915_0 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_3152_173_265705_0_1370972948_0 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_3001_511_1131389_0_1370573890_1 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_3255_514_1136823_0_1370974158_0 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_2903_515_1136808_0_1370972985_0 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_3255_509_1136823_0_1370974158_0 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_3255_510_1136823_0_1370974158_0 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_1068_414_1136822_0_1370974150_0 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_3255_504_1136823_0_1370974158_0 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_3255_506_1136823_0_1370974158_0 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_1068_411_1136822_0_1370974150_0 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_3300_317_1136920_0_1370980038_0 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_1068_501_1137031_0_1370987890_1 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_3300_505_1131567_0_1370587277_2 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_2903_524_1137030_0_1370987681_1 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_3300_506_1131567_0_1370587277_2 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_2903_522_1131615_0_1370591021_4 suspended by user
12/06/2013 03:58:45 | malariacontrol.net | task wu_3301_518_1131618_0_1370591360_2 suspended by user
12/06/2013 03:58:46 | Milkyway@Home | Sending scheduler request: To fetch work.
12/06/2013 03:58:46 | Milkyway@Home | Requesting new tasks for CPU
12/06/2013 03:58:48 | Milkyway@Home | Scheduler request completed: got 0 new tasks
12/06/2013 03:58:48 | Milkyway@Home | No tasks sent
12/06/2013 03:58:48 | Milkyway@Home | No tasks are available for MilkyWay@Home
12/06/2013 03:58:48 | Milkyway@Home | No tasks are available for Milkyway@Home Separation
12/06/2013 03:58:48 | Milkyway@Home | No tasks are available for the applications you have selected.

Very strange, looking into this one.

---

Hello.
I have a problem again with resources at GPU tasks. Now resources are :0.98 CPU + 1 ATI GPU (with opencl).

Hello,

I'm very happy that you put your time in this! As we say in Holland, "Petje af" which means "Top of the head to you".


However, I'm not getting any WU's at all.
The log-file tell's me the following:

12/06/2013 07:00:50 | Milkyway@Home | update requested by user
12/06/2013 07:00:53 | Milkyway@Home | Sending scheduler request: Requested by user.
12/06/2013 07:00:53 | Milkyway@Home | Requesting new tasks for CPU
12/06/2013 07:00:55 | Milkyway@Home | Scheduler request completed: got 0 new tasks
12/06/2013 07:00:55 | Milkyway@Home | No tasks sent

I hope you can fix this.............;-)

A big THANK YOU..

Greetings,
André Barnier from the Netherlands.

Regards dodgy WU's - I'm receiving units that are in excess of 2,000 hours. Account = Lonely - team Picard. As a relative newbie - just returned - it seems strange that these have less days to complete than can be achieved even with 24/7 rig time. Oddly enough, these same units complete in a few hours which raises the Q, are they completing correctly?

---

a) Maybe Error in *VALIDATION* ?
b) or was it a *bad* application ver. 1.08 which created invalid results?
c) OR (worst case) could it be an error in latest "1.18 mt" app ???

Please, investigate: WU id = 369440881

WU, that was correctly computed on app.ver 1.08 on stable computers (for example on my old i5 notebook: Hobitin)

Somehow FAILED to validate

*against* the WU finished correctly on app.ver "1.18 (mt)".

Here is the link:
http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=369440881

*Namaste*

---

Melwen - Child of the Fangorn Forest
Rig "BRISINGR" [ASUS G73-JH, i7 720QM 1.73, 4x2GB DDR3 1333 CL7, ATi HD5870M 1GB GDDR5],bought on 2011-02-24

Most AMD cards seem to work fine here. But my 6970s only create Computation errors since they run the OpenCL tasks again.

http://milkyway.cs.rpi.edu/milkyway/results.php?hostid=496290

They worked fine before with OpenCL.

I run the 7.0.64 (64bit) version of Boinc, and use 13.4 WHQL as AMD driver. I wonder why it worked before, with the same setup (about one week ago) and now it won't.
My idea was, Milkyway@home made changes, because the OpenCL apps were not working properly any more.

Just got another set of WUs (1.02 OpenCL), all finished with computation errors. But run-times seem a little short with 27 seconds.

Removed and reattached the project, and will see how it is going.

Regards

Sebastian*

PS: 160 Errors so far


Update:

Still computation error on the 6970. I brought a 6950 online, also computation Errors.
http://milkyway.cs.rpi.edu/milkyway/results.php?hostid=522312

@Travis:

My HD 4850 (WinXP/Cat. 12.1) is still working after your changes...everything's fine here. Just want to let you know.

Best wishes

Two issues:

1. mt still want every core. if you must do mt, default should be 1/2 available cores. not sure why you like mt, they are not efficient.

2. with an mt running, shuts down one of the gpus running.

if you must keep mt going, please allow user to disable in preferences or automatically disable if you have more than one gpu.

Running 2 ATI Radeons HD 7970 now without Errors. Great.
But still no WU´s for the CPUs since several hours. I mean on all my MilkyWay-PCs (1 PC with the Radeons, 2 PCs with NVIDIA GTX570 each).

It looks like this on one of the NVIDIA-PCs:

12.06.2013 15:46:12 | Milkyway@Home | Sending scheduler request: To fetch work.
12.06.2013 15:46:12 | Milkyway@Home | Requesting new tasks for CPU and NVIDIA
12.06.2013 15:46:14 | Milkyway@Home | Scheduler request completed: got 15 new tasks
12.06.2013 15:47:19 | Milkyway@Home | Sending scheduler request: To fetch work.
12.06.2013 15:47:19 | Milkyway@Home | Requesting new tasks for CPU and NVIDIA
12.06.2013 15:47:21 | Milkyway@Home | Scheduler request completed: got 8 new tasks
12.06.2013 15:48:26 | Milkyway@Home | Sending scheduler request: To fetch work.
12.06.2013 15:48:26 | Milkyway@Home | Requesting new tasks for CPU and NVIDIA
12.06.2013 15:48:28 | Milkyway@Home | Scheduler request completed: got 1 new tasks
12.06.2013 15:49:33 | Milkyway@Home | Sending scheduler request: To fetch work.
12.06.2013 15:49:33 | Milkyway@Home | Requesting new tasks for CPU and NVIDIA
12.06.2013 15:49:36 | Milkyway@Home | Scheduler request completed: got 0 new tasks
12.06.2013 15:49:36 | Milkyway@Home | No tasks sent
12.06.2013 15:49:36 | Milkyway@Home | This computer has reached a limit on tasks in progress

All incoming tasks are for GPU, not CPU.
What means ´This computer has reached a limit on tasks in Progress´ in detail?
Reached limits? On PCs set to ´no limits´ (if possible) in all categories?

Most AMD cards seem to work fine here. But my 6970s only create Computation errors since they run the OpenCL tasks again.

http://milkyway.cs.rpi.edu/milkyway/results.php?hostid=496290

They worked fine before with OpenCL.

I run the 7.0.64 (64bit) version of Boinc, and use 13.4 WHQL as AMD driver. I wonder why it worked before, with the same setup (about one week ago) and now it won't.
My idea was, Milkyway@home made changes, because the OpenCL apps were not working properly any more.

Just got another set of WUs (1.02 OpenCL), all finished with computation errors. But run-times seem a little short with 27 seconds.

Removed and reattached the project, and will see how it is going.

Regards

Sebastian*

PS: 160 Errors so far


Update:

Still computation error on the 6970. I brought a 6950 online, also computation Errors.
http://milkyway.cs.rpi.edu/milkyway/results.php?hostid=522312

What version of catalyst are you running, if you happen to know...

---

Most AMD cards seem to work fine here. But my 6970s only create Computation errors since they run the OpenCL tasks again.

http://milkyway.cs.rpi.edu/milkyway/results.php?hostid=496290

They worked fine before with OpenCL.

I run the 7.0.64 (64bit) version of Boinc, and use 13.4 WHQL as AMD driver. I wonder why it worked before, with the same setup (about one week ago) and now it won't.
My idea was, Milkyway@home made changes, because the OpenCL apps were not working properly any more.

Just got another set of WUs (1.02 OpenCL), all finished with computation errors. But run-times seem a little short with 27 seconds.

Removed and reattached the project, and will see how it is going.

Regards

Sebastian*

PS: 160 Errors so far


Update:

Still computation error on the 6970. I brought a 6950 online, also computation Errors.
http://milkyway.cs.rpi.edu/milkyway/results.php?hostid=522312

What version of catalyst are you running, if you happen to know...

I marked something red...

---

Most AMD cards seem to work fine here. But my 6970s only create Computation errors since they run the OpenCL tasks again.

http://milkyway.cs.rpi.edu/milkyway/results.php?hostid=496290

They worked fine before with OpenCL.

I run the 7.0.64 (64bit) version of Boinc, and use 13.4 WHQL as AMD driver. I wonder why it worked before, with the same setup (about one week ago) and now it won't.
My idea was, Milkyway@home made changes, because the OpenCL apps were not working properly any more.

Just got another set of WUs (1.02 OpenCL), all finished with computation errors. But run-times seem a little short with 27 seconds.

Removed and reattached the project, and will see how it is going.

Regards

Sebastian*

PS: 160 Errors so far


Update:

Still computation error on the 6970. I brought a 6950 online, also computation Errors.
http://milkyway.cs.rpi.edu/milkyway/results.php?hostid=522312

What version of catalyst are you running, if you happen to know...

I marked something red...

So before you were getting the opencl workunits and they were working, and it looks like now you're getting the 0.82 ATI (non-opencl) workunits and they're not working? At least looking at the server you should be getting 0.82 ATI (non-opencl) WUs.

---

GRR I ANGRY NOW, ONLY ERRORS :


498234788 380502780 189547 12 Jun 2013, 5:19:24 UTC 12 Jun 2013, 9:41:58 UTC Fehler beim Berechnen 43.09 1.06 --- MilkyWay@Home v1.02 (opencl_amd_ati)
498234782 380502776 189547 12 Jun 2013, 5:19:24 UTC 12 Jun 2013, 5:21:52 UTC Fehler beim Berechnen 28.83 1.03 --- MilkyWay@Home v1.02 (opencl_amd_ati)
498124033 380427089 189547 12 Jun 2013, 2:10:05 UTC 12 Jun 2013, 5:19:24 UTC Fehler beim Berechnen 42.89 0.97 --- MilkyWay@Home v1.02 (opencl_amd_ati)
498124032 380427088 189547 12 Jun 2013, 2:10:05 UTC 12 Jun 2013, 5:19:24 UTC Fehler beim Berechnen 42.68 0.98 --- MilkyWay@Home v1.02 (opencl_amd_ati)
498123621 380426807 189547 12 Jun 2013, 2:09:00 UTC 12 Jun 2013, 5:19:24 UTC Fehler beim Berechnen 43.76 0.94 --- MilkyWay@Home v1.02 (opencl_amd_ati)
498123234 380038971 189547 12 Jun 2013, 2:07:58 UTC 12 Jun 2013, 5:19:24 UTC Fehler beim Berechnen 54.82 1.05 --- MilkyWay@Home v1.02 (opencl_amd_ati)
498123231 380038968 189547 12 Jun 2013, 2:07:58 UTC 12 Jun 2013, 5:19:24 UTC Fehler beim Berechnen 59.69 1.06 --- MilkyWay@Home v1.02 (opencl_amd_ati)
498122876 380395756 189547 12 Jun 2013, 2:06:55 UTC 12 Jun 2013, 5:19:24 UTC Fehler beim Berechnen 55.63 1.00 --- MilkyWay@Home v1.02 (opencl_amd_ati)
498122875 380387872 189547 12 Jun 2013, 2:06:55 UTC 12 Jun 2013, 5:19:24 UTC Fehler beim Berechnen 54.74 1.09 --- MilkyWay@Home v1.02 (opencl_amd_ati)
498122558 380411009 189547 12 Jun 2013, 2:05:51 UTC 12 Jun 2013, 5:19:24 UTC Fehler beim Berechnen 54.81 1.01 --- MilkyWay@Home v1.02 (opencl_amd_ati)
498122181 380416989 189547 12 Jun 2013, 2:04:47 UTC 12 Jun 2013, 5:19:24 UTC Fehler beim Berechnen 42.70 0.95 --- MilkyWay@Home v1.02 (opencl_amd_ati)
498121855 380425785 189547 12 Jun 2013, 2:03:44 UTC 12 Jun 2013, 5:19:24 UTC Fehler beim Berechnen 54.68 1.17 --- MilkyWay@Home v1.02 (opencl_amd_ati)
498121854 380425784 189547 12 Jun 2013, 2:03:44 UTC 12 Jun 2013, 5:19:24 UTC Fehler beim Berechnen 59.77 0.95 --- MilkyWay@Home v1.02 (opencl_amd_ati)
498121458 380425529 189547 12 Jun 2013, 2:02:41 UTC 12 Jun 2013, 5:19:24 UTC Fehler beim Berechnen 41.74 1.06 --- MilkyWay@Home v1.02 (opencl_amd_ati)
498121171 379528436 189547 12 Jun 2013, 2:01:34 UTC 12 Jun 2013, 5:19:24 UTC Fehler beim Berechnen 42.68 0.90 --- MilkyWay@Home v1.02 (opencl_amd_ati)
498121170 379518575 189547 12 Jun 2013, 2:01:34 UTC 12 Jun 2013, 5:19:24 UTC Fehler beim Berechnen 55.02 1.06 --- MilkyWay@Home v1.02 (opencl_amd_ati)
498120896 380425188 189547 12 Jun 2013, 2:00:29 UTC 12 Jun 2013, 5:19:24 UTC Fehler beim Berechnen 49.90 0.90 --- MilkyWay@Home v1.02 (opencl_amd_ati)
498120895 380425187 189547 12 Jun 2013, 2:00:29 UTC 12 Jun 2013, 5:19:24 UTC Fehler beim Berechnen 49.73 0.97 --- MilkyWay@Home v1.02 (opencl_amd_ati)
498120522 380424927 189547 12 Jun 2013, 1:59:25 UTC 12 Jun 2013, 5:19:24 UTC Fehler beim Berechnen 55.87 1.06 --- MilkyWay@Home v1.02 (opencl_amd_ati)


AND I RUN WITH BM 7.0.64 AND ATI 13.4 DRIVER. I CHANGED PROJECT NOW.

---

Member of Charity Team
´

I still don't get any CPU work units:

13-06-12 13:28:31 | Milkyway@Home | update requested by user
13-06-12 13:28:32 | Milkyway@Home | Sending scheduler request: Requested by user.
13-06-12 13:28:32 | Milkyway@Home | Requesting new tasks for CPU
13-06-12 13:28:34 | Milkyway@Home | Scheduler request completed: got 0 new tasks
13-06-12 13:28:34 | Milkyway@Home | No tasks sent
13-06-12 13:28:34 | Milkyway@Home | No tasks are available for MilkyWay@Home
13-06-12 13:28:34 | Milkyway@Home | No tasks are available for Milkyway@Home Separation
13-06-12 13:28:34 | Milkyway@Home | No tasks are available for the applications you have selected.
13-06-12 13:28:34 | Milkyway@Home | Tasks for NVIDIA GPU are available, but your preferences are set to not accept them

It seem that there are no units available? My computer used to chew a lot of Separation work units, but the project status page (http://milkyway.cs.rpi.edu/milkyway/server_status.php) also shows none for Separation.

I just got some new tasks (yay) but...they were immediately aborted (not by me!).

one of those:

name of package: de_separation_20_2s_sscon_1_1371001450_382271
recieved: 12.06.2013 19:57:18 CEST


in my protocol I see the following:

Mi 12 Jun 2013 19:59:32 CEST | Milkyway@Home | [error] Missing coprocessor for task ps_separation_79_DR8_rev_2_1371001450_383161_0; aborting

one for each task I received.

Also, before those, there are these two:
Mi 12 Jun 2013 20:01:46 CEST | | App version needs OpenCL but GPU doesn't support it
Mi 12 Jun 2013 20:01:46 CEST | Milkyway@Home | [error] App version uses non-existent ATI GPU


which is complete bullshit since my GPU is there and it's working, driver is fully functional no errors in Xorg.0.log.

*edit
I didn't have the SDK installed...but it did work before without it. Strange.

Most AMD cards seem to work fine here. But my 6970s only create Computation errors since they run the OpenCL tasks again.

http://milkyway.cs.rpi.edu/milkyway/results.php?hostid=496290

They worked fine before with OpenCL.

I run the 7.0.64 (64bit) version of Boinc, and use 13.4 WHQL as AMD driver. I wonder why it worked before, with the same setup (about one week ago) and now it won't.
My idea was, Milkyway@home made changes, because the OpenCL apps were not working properly any more.

Just got another set of WUs (1.02 OpenCL), all finished with computation errors. But run-times seem a little short with 27 seconds.

Removed and reattached the project, and will see how it is going.

Regards

Sebastian*

PS: 160 Errors so far


Update:

Still computation error on the 6970. I brought a 6950 online, also computation Errors.
http://milkyway.cs.rpi.edu/milkyway/results.php?hostid=522312



What version of catalyst are you running, if you happen to know...

I was getting the non openCL WUs (.82) a few days ago, and they were working. Now i am getting the openCL WUs again, and they are not working on the 69xx cards.
The 1.02 openCL WUs were working several weeks ago, when there were no changes on the servers.

I still don't get any CPU work units:

13-06-12 13:28:31 | Milkyway@Home | update requested by user
13-06-12 13:28:32 | Milkyway@Home | Sending scheduler request: Requested by user.
13-06-12 13:28:32 | Milkyway@Home | Requesting new tasks for CPU
13-06-12 13:28:34 | Milkyway@Home | Scheduler request completed: got 0 new tasks
13-06-12 13:28:34 | Milkyway@Home | No tasks sent
13-06-12 13:28:34 | Milkyway@Home | No tasks are available for MilkyWay@Home
13-06-12 13:28:34 | Milkyway@Home | No tasks are available for Milkyway@Home Separation
13-06-12 13:28:34 | Milkyway@Home | No tasks are available for the applications you have selected.
13-06-12 13:28:34 | Milkyway@Home | Tasks for NVIDIA GPU are available, but your preferences are set to not accept them

It seem that there are no units available? My computer used to chew a lot of Separation work units, but the project status page (http://milkyway.cs.rpi.edu/milkyway/server_status.php) also shows none for Separation.

There shouldn't be any separation workunits, those should be under milkyway, as far as I know.

---