I have been using the 4.46 beta (and its predecessor, the power application 4.36) for some time now, and I have run across an unusual error that does not seem to be related prolonged Internet access failure. Neither is the host overclocked or in an unstable condition.
Edit: I do not see this error often, but I thought I would post this on the off chance it might mean something.
2008-06-27 03:49:48.8194 [normal]: Built at: Feb 25 2008 17:11:17
2008-06-27 03:49:48.8194 [normal]: Start of BOINC application '..\\..\\projects\\einstein.phys.uwm.edu\\einstein_S5R3_4.46_windows_intelx86_1.exe'.
2008-06-27 03:49:48.9441 [debug]: Set up communication with graphics process.
2008-06-27 03:49:49.4116 [debug]: Reading SFTs and setting up stacks ... done
2008-06-27 03:50:48.9502 [normal]: INFO: Couldn't open checkpoint h1_0982.25_S5R3__3_S5R3b_4_0.cpt
2008-06-27 03:50:48.9502 [debug]: Total skypoints = 1202. Progress: 0,
$Revision: 1.119 $ REV:$Revision, OPT:3, SCVAR:9, SCTRIM:2, HOTVAR:3, HOTDIV:0, HGHPRE:7, HGHBAT:2
c
1, c
2, c
3, c
4, c
5, c
6, c
[snip]
680, c
681, c
682,
APP DEBUG: Application caught signal 8.
Stack trace of LAL functions in worker thread:
FstatVectToPeakGram at line 2075 of file HierarchicalSearch.c
(null) at line 0 of file (null)
At lowest level status code = 0, description: NO LAL ERROR REGISTERED
I don't see this on Windows very often, it's much more common on Linux due to the "lazy FPU restore" on certain kernels.
As for Windows I've read some reports where some badly programmed hardware drivers (typically printer drivers) leave the FPU stack in a bad state, which would cause such an exception.
I am aware that signal 8 == FPE, hence my statement of my host being non-overclocked and stable. I suspect the error could be one of those rare once-in-a-blue-moon occurrences - I haven't done any printing recently.
After instaling einstein_S5R3_4.46_windows_intelx86.zip I have ben unable to download work. I get the mesage "2008-09-10 20:14:07|Einstein@Home|Sending scheduler request: Requested by user. Requesting 0 seconds of work, reporting 0 completed tasks"
I am using Vista 64 and as fare I can understand the units I have finished so fare are ok.
Well, the first thing to keep in mind is the Beta and Power apps were for S5R3. We are now working S5R4, except for any 'resend' tasks required to cleanup and close out the R3 search.
So you have two choices:
1.) If you still want to run the Beta and/or Power apps and still be able to run the new work, you will have to add the references for the new application to your app_info file for EAH.
2.) Delete the app_info file and let BOINC pick up the stock apps as required. This will slow down the execution of any R3 resends you pick up some, but you won't have to worry about keeping up to date with any changes which come about for the application for the R4 search (probably the easiest way out).
After instaling einstein_S5R3_4.46_windows_intelx86.zip I have ben unable to download work. I get the mesage "2008-09-10 20:14:07|Einstein@Home|Sending scheduler request: Requested by user. Requesting 0 seconds of work, reporting 0 completed tasks"
I am using Vista 64 and as fare I can understand the units I have finished so fare are ok.
Should I delete the Beta aplication?
This app was written for the previous S5R3 run and is outdated by now, actually it is incompatible with current workunits. The current stock app is almost identical in performnace to this power app anyway, so even if it worked there would be no reason to use it. I advice so remove the app_info.xml file again which will cause the stock up to be downloaded automagically.
I have been using the 4.46
)
I have been using the 4.46 beta (and its predecessor, the power application 4.36) for some time now, and I have run across an unusual error that does not seem to be related prolonged Internet access failure. Neither is the host overclocked or in an unstable condition.
Edit: I do not see this error often, but I thought I would post this on the off chance it might mean something.
Result 100320026
- exit code -1073741680 (0xc0000090)
2008-06-27 03:49:48.8194 [normal]: Built at: Feb 25 2008 17:11:17
2008-06-27 03:49:48.8194 [normal]: Start of BOINC application '..\\..\\projects\\einstein.phys.uwm.edu\\einstein_S5R3_4.46_windows_intelx86_1.exe'.
2008-06-27 03:49:48.9441 [debug]: Set up communication with graphics process.
2008-06-27 03:49:49.4116 [debug]: Reading SFTs and setting up stacks ... done
2008-06-27 03:50:48.9502 [normal]: INFO: Couldn't open checkpoint h1_0982.25_S5R3__3_S5R3b_4_0.cpt
2008-06-27 03:50:48.9502 [debug]: Total skypoints = 1202. Progress: 0,
$Revision: 1.119 $ REV:$Revision, OPT:3, SCVAR:9, SCTRIM:2, HOTVAR:3, HOTDIV:0, HGHPRE:7, HGHBAT:2
c
1, c
2, c
3, c
4, c
5, c
6, c
[snip]
680, c
681, c
682,
APP DEBUG: Application caught signal 8.
Stack trace of LAL functions in worker thread:
FstatVectToPeakGram at line 2075 of file HierarchicalSearch.c
(null) at line 0 of file (null)
At lowest level status code = 0, description: NO LAL ERROR REGISTERED
]]>
Soli Deo Gloria
RE: APP DEBUG: Application
)
This is a floating point exception (FPE).
I don't see this on Windows very often, it's much more common on Linux due to the "lazy FPU restore" on certain kernels.
As for Windows I've read some reports where some badly programmed hardware drivers (typically printer drivers) leave the FPU stack in a bad state, which would cause such an exception.
What version of Windows are you running?
BM
BM
I am aware that signal 8 ==
)
I am aware that signal 8 == FPE, hence my statement of my host being non-overclocked and stable. I suspect the error could be one of those rare once-in-a-blue-moon occurrences - I haven't done any printing recently.
This host is running Windows XP SP3.
Soli Deo Gloria
After instaling
)
After instaling einstein_S5R3_4.46_windows_intelx86.zip I have ben unable to download work. I get the mesage "2008-09-10 20:14:07|Einstein@Home|Sending scheduler request: Requested by user. Requesting 0 seconds of work, reporting 0 completed tasks"
I am using Vista 64 and as fare I can understand the units I have finished so fare are ok.
Should I delete the Beta aplication?
Well, the first thing to keep
)
Well, the first thing to keep in mind is the Beta and Power apps were for S5R3. We are now working S5R4, except for any 'resend' tasks required to cleanup and close out the R3 search.
So you have two choices:
1.) If you still want to run the Beta and/or Power apps and still be able to run the new work, you will have to add the references for the new application to your app_info file for EAH.
2.) Delete the app_info file and let BOINC pick up the stock apps as required. This will slow down the execution of any R3 resends you pick up some, but you won't have to worry about keeping up to date with any changes which come about for the application for the R4 search (probably the easiest way out).
HTH,
Alinator
RE: After instaling
)
This app was written for the previous S5R3 run and is outdated by now, actually it is incompatible with current workunits. The current stock app is almost identical in performnace to this power app anyway, so even if it worked there would be no reason to use it. I advice so remove the app_info.xml file again which will cause the stock up to be downloaded automagically.
CU
Bikeman
RE: Requested by user.
)
Means, that your computer has enough work ATM, according to your resource share. This is not related to installing the beta-application.
E@H now processes S5R4, so installing beta-app. for S5R3 is no longer necessary.
Michael
edit: D'oh, Bikeman and Alinator were faster...
Team Linux Users Everywhere
At least we all agree
)
At least we all agree :-)
CU
Bikeman
LOL... No CBNC here!
)
LOL...
No CBNC here! :-D
Alinator
Thank you all. Hmm.... I
)
Thank you all.
Hmm.... I see.
So the Beta application should be removed and not possible to download and run at all.