Have you excluded the BOINC directory from beeing scanned for viruses? Result start failure might be caused by that.
Or does your windows version require to explicite set the executable files to be allowed to be executed?
I don't have any anti-virus program running in Vista.
OK, then you need to make sure the account you're running BOINC from has enough access privileges to be able to execute, create, rename, and/or delete files in the BOINC directories.
Another thought here is how do you have BOINC installed? As a Service, Single User, or All Users? This can have a bearing on what type of permissions are present when BOINC tries to run.
One other question (regarding a comment on your other thread about this problem), if you were seeking an install and forget DC setup, why are you running it on a Beta OS?
Well I removed BOINC again. After doing so, I went and checked if the directory and files were still there. To my surprise, they were. So I deleted the directory.
I then reloaded BOINC again.
This time it appears to be running OK, so far.
Why it stopped running in the first place is a mystery. It seemed to run OK for a while on Vista w/no problems.
Now to the question of my running one of the optimized versions? Yes. Could that be the issue?
Now to go through and do a clean install on the rest of my computers.
I wish to appologize for my bringing up the problem, and to thank everyone here for their input.
Have you excluded the BOINC directory from beeing scanned for viruses? Result start failure might be caused by that.
Or does your windows version require to explicite set the executable files to be allowed to be executed?
I don't have any anti-virus program running in Vista.
OK, then you need to make sure the account you're running BOINC from has enough access privileges to be able to execute, create, rename, and/or delete files in the BOINC directories.
Another thought here is how do you have BOINC installed? As a Service, Single User, or All Users? This can have a bearing on what type of permissions are present when BOINC tries to run.
One other question (regarding a comment on your other thread about this problem), if you were seeking an install and forget DC setup, why are you running it on a Beta OS?
Alinator
I do use this one, w/it dual core A64, as my primary system for running games and such. It's also the fastest, relative to Einstein@Home.
First off, no problemo about bringing up questions about BOINC and science app weirdness. That's what these forums are all about. ;-)
Second, sorry if I got a little testy earlier, no offense intended. :-)
Quote:
Well I removed BOINC again. After doing so, I went and checked if the directory and files were still there. To my surprise, they were. So I deleted the directory.
Yes, that's normal for the uninstaller to leave the specific project related files and directory structure intact. This is to avoid losing any work onboard when updates to BOINC itself occur (as opposed to the automatic science app updates if running in the standard configuration, which are transparent to the end user).
Quote:
I then reloaded BOINC again.
This time it appears to be running OK, so far.
Why it stopped running in the first place is a mystery. It seemed to run OK for a while on Vista w/no problems.
Now to the question of my running one of the optimized versions? Yes. Could that be the issue?
I don't think this was the root cause for the Vista problem, since it looks like you were running optimized for S4 and not S5. This definitely looks like it was a Vista specific problem with S5 on that one machine.
One thing you may want to consider would be to enable more detailed auditing. I haven't worked much with Vista, but if it's anything like other NT cored OS's, MS has only minimal auditing enabled by default. Of course this might not be the case since it is Beta.
Also, you can keep an eye on what the S5 app is doing by taking a look at the stderr file in whichever slot directory EAH is currently running from. Just make sure you don't save it when you close the editor.
Quote:
Now to go through and do a clean install on the rest of my computers.
I don't know if you need to do that on your XP boxes, they seem to be running fine. This might be a case of don't go looking for trouble (cause you just might find it)! :-)
Quote:
I wish to appologize for my bringing up the problem, and to thank everyone here for their input.
On my Dual Core 6600 my Daily Quota of 2 * 32 = 64 is reached after 14 hours (one short WU is chrunched after 27 minutes).
Is it posible to increase the Daily Quota ?
afaik, 32 WUs per CPU is the maximum. If I were you I'd try to delete the data file (after finishing the WUs you have already downloaded) and hope for one with longer WUs.
afaik, 32 WUs per CPU is the maximum. If I were you I'd try to delete the data file (after finishing the WUs you have already downloaded) and hope for one with longer WUs.
If I were you I'd try to delete the data file (after finishing the WUs you have already downloaded) and hope for one with longer WUs.
I tried that once. It just noticed the file was missing and downloaded it again. At minimum, to make this work, I suspect you'd need to purge a reference to the major data file in the client state file. I've not tried that, and I don't suggest it.
I think it has been said here that the server process biases things to make it more likely that fast machines get the big results, and slow machines the small ones, but it can take a while.
Last night we had a power failure, and BOINC lost a lot of state when my UPS quit in the middle of my shut down. My HT P4 Gallatin gots its first short results in many weeks this when things started back up. I expect them to stop coming after we get through that datafile, and I'm already getting long results from one or two other datafiles.
If I were you I'd try to delete the data file (after finishing the WUs you have already downloaded) and hope for one with longer WUs.
I tried that once. It just noticed the file was missing and downloaded it again. At minimum, to make this work, I suspect you'd need to purge a reference to the major data file in the client state file. I've not tried that, and I don't suggest it.
My Athlon 64 3500 (single-core) usually crunches long WU, but recently I've received some short ones - it seems to happens erratically, for 10 to 14 hours, then I receive long WU again. I don't think it has something to do with any files. Indeed, I do suspect that the respnsability lays upon the ratio "granted credits" Vs computing time OR total number of WU crunched.
Seems to me that after a long serie of crunched WU, all in pending state (I speak about 1,000 pending credits or more) my aforesaid computer is switched to short WU.
It's only a guess !
"Entia non sunt multiplicandam praeter necessitatem" (OKHAM)
I know it is a bit off-topic but what is the Athlon 3500+ like? Because I'm going to get one of those... I had a look at your profile but that didn't tell me much as you're running linux and we all know the benchmarks are completely different under linux than under Windows...
RE: RE: -185 =
)
OK, then you need to make sure the account you're running BOINC from has enough access privileges to be able to execute, create, rename, and/or delete files in the BOINC directories.
Another thought here is how do you have BOINC installed? As a Service, Single User, or All Users? This can have a bearing on what type of permissions are present when BOINC tries to run.
One other question (regarding a comment on your other thread about this problem), if you were seeking an install and forget DC setup, why are you running it on a Beta OS?
Alinator
Well I removed BOINC again.
)
Well I removed BOINC again. After doing so, I went and checked if the directory and files were still there. To my surprise, they were. So I deleted the directory.
I then reloaded BOINC again.
This time it appears to be running OK, so far.
Why it stopped running in the first place is a mystery. It seemed to run OK for a while on Vista w/no problems.
Now to the question of my running one of the optimized versions? Yes. Could that be the issue?
Now to go through and do a clean install on the rest of my computers.
I wish to appologize for my bringing up the problem, and to thank everyone here for their input.
RE: RE: RE: -185 =
)
I do use this one, w/it dual core A64, as my primary system for running games and such. It's also the fastest, relative to Einstein@Home.
First off, no problemo about
)
First off, no problemo about bringing up questions about BOINC and science app weirdness. That's what these forums are all about. ;-)
Second, sorry if I got a little testy earlier, no offense intended. :-)
Yes, that's normal for the uninstaller to leave the specific project related files and directory structure intact. This is to avoid losing any work onboard when updates to BOINC itself occur (as opposed to the automatic science app updates if running in the standard configuration, which are transparent to the end user).
I don't think this was the root cause for the Vista problem, since it looks like you were running optimized for S4 and not S5. This definitely looks like it was a Vista specific problem with S5 on that one machine.
One thing you may want to consider would be to enable more detailed auditing. I haven't worked much with Vista, but if it's anything like other NT cored OS's, MS has only minimal auditing enabled by default. Of course this might not be the case since it is Beta.
Also, you can keep an eye on what the S5 app is doing by taking a look at the stderr file in whichever slot directory EAH is currently running from. Just make sure you don't save it when you close the editor.
I don't know if you need to do that on your XP boxes, they seem to be running fine. This might be a case of don't go looking for trouble (cause you just might find it)! :-)
Alinator
On my Dual Core 6600 my Daily
)
On my Dual Core 6600 my Daily Quota of 2 * 32 = 64 is reached after 14 hours (one short WU is chrunched after 27 minutes).
Is it posible to increase the Daily Quota ?
afaik, 32 WUs per CPU is the
)
afaik, 32 WUs per CPU is the maximum. If I were you I'd try to delete the data file (after finishing the WUs you have already downloaded) and hope for one with longer WUs.
RE: afaik, 32 WUs per CPU
)
Thank you, Annika.
RE: If I were you I'd try
)
I tried that once. It just noticed the file was missing and downloaded it again. At minimum, to make this work, I suspect you'd need to purge a reference to the major data file in the client state file. I've not tried that, and I don't suggest it.
I think it has been said here that the server process biases things to make it more likely that fast machines get the big results, and slow machines the small ones, but it can take a while.
Last night we had a power failure, and BOINC lost a lot of state when my UPS quit in the middle of my shut down. My HT P4 Gallatin gots its first short results in many weeks this when things started back up. I expect them to stop coming after we get through that datafile, and I'm already getting long results from one or two other datafiles.
RE: RE: If I were you I'd
)
My Athlon 64 3500 (single-core) usually crunches long WU, but recently I've received some short ones - it seems to happens erratically, for 10 to 14 hours, then I receive long WU again. I don't think it has something to do with any files. Indeed, I do suspect that the respnsability lays upon the ratio "granted credits" Vs computing time OR total number of WU crunched.
Seems to me that after a long serie of crunched WU, all in pending state (I speak about 1,000 pending credits or more) my aforesaid computer is switched to short WU.
It's only a guess !
"Entia non sunt multiplicandam praeter necessitatem"
(OKHAM)
I know it is a bit off-topic
)
I know it is a bit off-topic but what is the Athlon 3500+ like? Because I'm going to get one of those... I had a look at your profile but that didn't tell me much as you're running linux and we all know the benchmarks are completely different under linux than under Windows...