Welcome to MilkyWay@home

Invalidated WUs??

Message boards : Number crunching : Invalidated WUs??
Message board moderation

To post messages, you must log in.

AuthorMessage
WimTea

Send message
Joined: 16 Nov 07
Posts: 23
Credit: 4,774,710
RAC: 0
Message 30659 - Posted: 13 Sep 2009, 11:36:45 UTC
Last modified: 13 Sep 2009, 11:39:08 UTC

My host 46925 suddenly has the 9.somethng WUs ps_constrainted_82_2s_4 and gs_constrainted_82_2s_4 invalidated. I've been OC'ing this host a bit but can't think of why only these WUs are affected. Anyone else got these invalidated too?
I'll try running one without OC'ing in the meantime.
[edit] gotta ca(t)ch(e) one first ofcourse and I don't really feel like trashing a perfectly OK cache...
ID: 30659 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile [Russia] michs

Send message
Joined: 16 Oct 08
Posts: 18
Credit: 164,409,593
RAC: 0
Message 30666 - Posted: 13 Sep 2009, 13:42:46 UTC

I have many invalidated wu on different hosts.
Host 79041 one of then. crunch on gpu.
radeon 4850, cpu and gpu at stock speed.
windows 7 32bit, catalist 9.5


Host 102639 crunch on cpu.

changing client from 0.19 to 0.20 not help.

ID: 30666 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
John R. @ SETI.USA

Send message
Joined: 1 Jan 09
Posts: 15
Credit: 85,816,654
RAC: 0
Message 30667 - Posted: 13 Sep 2009, 14:16:17 UTC

I've been getting quite a few of these "invalid" WUs also.

Travis has posted that he has fixed the problem with that set of work units.

But, I'm still getting a few as of this morning.
ID: 30667 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
WimTea

Send message
Joined: 16 Nov 07
Posts: 23
Credit: 4,774,710
RAC: 0
Message 30669 - Posted: 13 Sep 2009, 15:15:15 UTC

Travis posted this in the 11th.
However on my host these units were from the 12th around 20:30 UT till this morning around 07:30 UTC.
Just this single host, my other hosts got non of these units.
Seems to be fine now though, guess it was sorted out.
ID: 30669 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Beyond
Avatar

Send message
Joined: 15 Jul 08
Posts: 383
Credit: 729,293,740
RAC: 0
Message 30791 - Posted: 14 Sep 2009, 9:47:44 UTC

I've just started getting about 20% invalid WUs in the last few hours. Never had many at all before. HD 4770 GPU.
ID: 30791 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
John Clark

Send message
Joined: 4 Oct 08
Posts: 1734
Credit: 64,228,409
RAC: 0
Message 30804 - Posted: 14 Sep 2009, 11:36:34 UTC

My CPU rig is getting 100% invalid WUs (both opti 0.19 and 0.20), and my GPU is validating OK on 0.20 but has not got work for the last 3 hours. ;((
Go away, I was asleep


ID: 30804 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile gigadisk

Send message
Joined: 24 Apr 08
Posts: 3
Credit: 44,584,709
RAC: 0
Message 30820 - Posted: 14 Sep 2009, 13:49:53 UTC
Last modified: 14 Sep 2009, 13:50:10 UTC

I have many Invalid WU with new 0.20 application (astronomy_0.20_SSE3.exe by Gipsel) with old 0.19 SSE 4.1 was all ok.

http://milkyway.cs.rpi.edu/milkyway/result.php?resultid=104404259
http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=102275165

And many others WU, have some one same problem with 0.20 app ?
ID: 30820 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Cluster Physik

Send message
Joined: 26 Jul 08
Posts: 627
Credit: 94,940,203
RAC: 0
Message 30825 - Posted: 14 Sep 2009, 14:50:35 UTC - in response to Message 30820.  

I have many Invalid WU with new 0.20 application (astronomy_0.20_SSE3.exe by Gipsel) with old 0.19 SSE 4.1 was all ok.

http://milkyway.cs.rpi.edu/milkyway/result.php?resultid=104404259
http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=102275165

And many others WU, have some one same problem with 0.20 app ?

As mentioned here there are a lot of broken WUs around lately. All application versions will create invalid results with those WUs. That it happend as you switched to the new application versions is just coincidence.

Michs explicitly stated above that he experienced those errors first while still running 0.19 and updated only after that to see if 0.20 does better, but it can't given those faulty WUs.
ID: 30825 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Bruce
Avatar

Send message
Joined: 28 Apr 08
Posts: 1415
Credit: 2,716,428
RAC: 0
Message 30850 - Posted: 14 Sep 2009, 17:56:57 UTC
Last modified: 14 Sep 2009, 17:57:48 UTC

I'm getting INVALID results now too, just a few, but I've not noticed any b4 yesterday. Now there seems to be at least a couple new ones every time I check.
;-(
ID: 30850 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Beyond
Avatar

Send message
Joined: 15 Jul 08
Posts: 383
Credit: 729,293,740
RAC: 0
Message 30876 - Posted: 14 Sep 2009, 20:16:30 UTC - in response to Message 30791.  

I've just started getting about 20% invalid WUs in the last few hours. Never had many at all before. HD 4770 GPU.

Solved for me anyway. This was the BOINC message I was getting:

Milkyway@home 9/14/2009 1:35:24 PM Output file de_constrainted_82_2s_5_228058_1252953073_0_0 for task de_constrainted_82_2s_5_228058_1252953073_0 absent

Seems it was caused when I switched to BOINC v6.10.4. Going back to v6.10.3 solved the problem.

ID: 30876 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile MC707
Avatar

Send message
Joined: 5 Jun 09
Posts: 2
Credit: 1,704
RAC: 0
Message 30920 - Posted: 15 Sep 2009, 2:17:18 UTC

I am also getting invalid WUs. Never happened to me before, never happens to me on any other project. MW should solve this fast... two of those work units took 150 mins each, which could gracefully be used for my rosetta or my WCG. Anyway, hope they solve that soon, I like this project and I wouldn't want to leave it behind :(
ID: 30920 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
John R. @ SETI.USA

Send message
Joined: 1 Jan 09
Posts: 15
Credit: 85,816,654
RAC: 0
Message 30926 - Posted: 15 Sep 2009, 3:58:09 UTC

As posted on the front page..."Let me know if you're having any issues with the *_6 workunits."

I did find one workunit of the *_6 variety that was given the dreaded "invalid" label.

Travis might need to do a little more tweaking to get it all straightened out.

A whole lotta wasted CPU cycles the last couple of days.
ID: 30926 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
John Clark

Send message
Joined: 4 Oct 08
Posts: 1734
Credit: 64,228,409
RAC: 0
Message 30948 - Posted: 15 Sep 2009, 12:31:22 UTC - in response to Message 30926.  

As posted on the front page..."Let me know if you're having any issues with the *_6 workunits."

I did find one workunit of the *_6 variety that was given the dreaded "invalid" label.

Travis might need to do a little more tweaking to get it all straightened out.

A whole lotta wasted CPU cycles the last couple of days.


Using CPUs I have to agree - see here
Go away, I was asleep


ID: 30948 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
John Clark

Send message
Joined: 4 Oct 08
Posts: 1734
Credit: 64,228,409
RAC: 0
Message 30952 - Posted: 15 Sep 2009, 13:38:20 UTC
Last modified: 15 Sep 2009, 13:39:46 UTC

On my CPU only rig which is the one giving all the invalid WU results, on the _82_2s_6_ WUs, despite seemingly doing a normal amount of crunching, I am trying an experiment.

I am looking at the MW WU cache for _82_2s_6_ work and aborting them. There seems to be a large number of _82_3s_6_ WUs as well, but I am leaving these alone.

My proof of the invalid work will be if the Account results for that rig show no more invalid results posted.
Go away, I was asleep


ID: 30952 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote

Message boards : Number crunching : Invalidated WUs??

©2024 Astroinformatics Group