You're right, Kajunfisher. I sometimes wish the moderators would delete some of the stupid posts I've made, but they leave them there and I look like an idiot in front of everyone (italics okay with you?) Happy crunching!
I hope that the higher power above will realize some of the things I've said and done were stupid and moderate those things as well. We all do things at some point in our lives that we regret doing or saying. It's human nature. I try to observe things from more than one perspective as best I can before I make an opinion or a statement. It's one of the things that I learned from a leadership development course while I was in the service. -It may be something small to you, but it will mean the whole world to someone else.
You're right, Kajunfisher. I sometimes wish the moderators would delete some of the stupid posts I've made, but they leave them there and I look like an idiot in front of everyone (italics okay with you?) Happy crunching!
I hope that the higher power above will realize some of the things I've said and done were stupid and moderate those things as well. We all do things at some point in our lives that we regret doing or saying. It's human nature. I try to observe things from more than one perspective as best I can before I make an opinion or a statement. It's one of the things that I learned from a leadership development course while I was in the service. -It may be something small to you, but it will mean the whole world to someone else.
Happy crunching to you too :-)
In retrospect, I wish that I could have eaten a few words, (ahem, er, "asleep-at-the-switch user..." when a wiser mind later pointed out that the user had a problem of which he/she was not aware) back from a forum. We all slip sometimes. We all look around to see if anyone noticed, and pick ourselves back up.
microcraft
"The arc of history is long, but it bends toward justice" - MLK
I am an Alpha Tester for BOINC and there is a place that we report issues with any version that we test, per the testing guidelines.
I recommend that you read the BOINC Dev boards to stay aware of potential problems/issues before downloading any newer version.
Post your problems/bugs/issues with the BOINC client there.
Kajun,
Thanks for putting up that link.
I have checked those boards every couple of days, but found precious little regarding 5.2.1 there.
I noticed Metod's comment about upload delays. Running solely Einstein, I have not seen this behavior, nor did he, and I deemed it unnecessary to post a "me too" comment over there when I could add nothing further.
I crunched SetiClassic starting 2002-12-03 (under "microcraft") until May last year, when I took a hard look at it and concluded that my work there had been just about the most useless waste of electricity, except for contributing to a demonstration of the power of Distributed Computing. The likelihood of discovering, let alone communicating with an ETI is miniscule, given the probable span of lifeforms and the vast distance/time involved, *unless* they happen to be patroling our solar system and observing us, in which case, why would they possibly want communication with such a violent, erratic species unless we looked like food (nah, too polluted) or slaves (also unlikely, too unruly). (just my personal conclusion about MY time with Seti, NOT directed at other crunchers. People, don't even waste the flames here - this is a testing thread, and I refuse to rise to the bait) I'm not interested at this point in crunching for other projects, so my testing will be limited to the Windows/Boinc/Einstein combination, and of little interest to anyone outside the Einstein community, ergo the thread here.
Regards
Michael
microcraft
"The arc of history is long, but it bends toward justice" - MLK
As promised I'm back here with another BOINC 5.2.1 issue. Right now this is just my experience so don't feel up to finding a dev list to post anything formal - it's been a hard Monday doing jigsawpuzzle adverts for assorted freebe newspapers and I'm tired, but anyway that's not BOINC.
OK the issue I noticed is that although I have set the preferences for SETI@home and for Einstein@home to run one hour each, when I staggered through the door, I saw that SETI had hogged all the crunching time - I had to actually suspend seti manually to get Einstein running - this over 9 hours or so. It is possible that this has happened due to the malfunctioning password issue of my previous post. I will try detaching from Einstein shortly and then rejoing with the password to see if that makes any difference.
As promised I'm back here with another BOINC 5.2.1 issue. Right now this is just my experience so don't feel up to finding a dev list to post anything formal - it's been a hard Monday doing jigsawpuzzle adverts for assorted freebe newspapers and I'm tired, but anyway that's not BOINC.
OK the issue I noticed is that although I have set the preferences for SETI@home and for Einstein@home to run one hour each, when I staggered through the door, I saw that SETI had hogged all the crunching time - I had to actually suspend seti manually to get Einstein running - this over 9 hours or so. It is possible that this has happened due to the malfunctioning password issue of my previous post. I will try detaching from Einstein shortly and then rejoing with the password to see if that makes any difference.
Gray
Gray,
It is possible that the issue you mention can be attributed to the correct operation of the scheduler, and goes back to the 4.x Boinc clients.
If you had a work queue full of Seti WUs when you added the Einstein project, the scheduler may be giving preference to Seti in order to establish equilibrium. That is, it is designed to enter Earliest Deadline First mode when it senses that your workload approaches full.
Did you notice something similar when you added a new project while running the previous Boinc version?
I think this may settle down after a few days of crunching, if the scheduler is left alone to do it's thing. Over-riding it's decision tends to delay the proper settling-in process. It will also depend upon what share values you have set for the various projects you run, and allot time according to those settings.
I may have done an inadequate job of describing the scheduler activity, or may have made an incorrect assumption; I have no experience in running multiple projects, and have learned from studying solutions that have been posted by more-experienced folks here. Perhaps one of them will drop by and give a better answer.
Please keep checking in with your reports. :-)
Hope it's helpful!
Michael
microcraft
"The arc of history is long, but it bends toward justice" - MLK
As promised I'm back here with another BOINC 5.2.1 issue. Right now this is just my experience so don't feel up to finding a dev list to post anything formal - it's been a hard Monday doing jigsawpuzzle adverts for assorted freebe newspapers and I'm tired, but anyway that's not BOINC.
OK the issue I noticed is that although I have set the preferences for SETI@home and for Einstein@home to run one hour each, when I staggered through the door, I saw that SETI had hogged all the crunching time - I had to actually suspend seti manually to get Einstein running - this over 9 hours or so. It is possible that this has happened due to the malfunctioning password issue of my previous post. I will try detaching from Einstein shortly and then rejoing with the password to see if that makes any difference.
Gray
Gray,
It is possible that the issue you mention can be attributed to the correct operation of the scheduler, and goes back to the 4.x Boinc clients.
If you had a work queue full of Seti WUs when you added the Einstein project, the scheduler may be giving preference to Seti in order to establish equilibrium. That is, it is designed to enter Earliest Deadline First mode when it senses that your workload approaches full.
Did you notice something similar when you added a new project while running the previous Boinc version?
I think this may settle down after a few days of crunching, if the scheduler is left alone to do it's thing. Over-riding it's decision tends to delay the proper settling-in process. It will also depend upon what share values you have set for the various projects you run, and allot time according to those settings.
I may have done an inadequate job of describing the scheduler activity, or may have made an incorrect assumption; I have no experience in running multiple projects, and have learned from studying solutions that have been posted by more-experienced folks here. Perhaps one of them will drop by and give a better answer.
Please keep checking in with your reports. :-)
Hope it's helpful!
Michael
Hello Michael
Yes that info WAS useful - I had forgotten that ability. I did have 35-odd seti@home WUs listed before I joined Einstein (seti seems to have a somewhat over-sized idea of the power of this lil box).
What I have done, since my last post, now that I have a functioning password for Einstein is to upload what was crunched, unattach from everything, uninstall BOINC, trash BOINC folder and reinstall BOINC, then re-attach to Einstein first. I also now have a password for Climate, so have attached to that one
- the deadline is a realistic year away (should manage that I reckon :) ) - instead of seti@home. Watch this space. I will leave the system running through the night (thank goodness for Zalman fans...) and let you know what happens Tuesday evening after this slave returns home.
Just come across this thread. The following problem I have been reporting in the SETI Beta message boards, but thought I would post it here to add my experience with CC 5.2.1.
The problem I have been having since 5.1.x, and now 5.2.1, is the HTTP Proxy. Whenever I try to use it I get a error code of 500. I can connect fine with CC 4.72, but the new proxy procedures do not seem to talk to my company's proxy. Does not matter which of the 5.x.x version CC I use, I get the the 500 error; switch back to version 4.72 CC and I can get through the proxy without any problems.
This is the error message that 5.x.x gives me:
10/17/2005 3:27:16 PM|Einstein@Home|Scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi failed with a return value of 500
10/17/2005 3:27:16 PM|Einstein@Home|No schedulers responded
10/17/2005 3:27:21 PM|Einstein@Home|Deferring communication with project for 54 seconds
Just come across this thread. The following problem I have been reporting in the SETI Beta message boards, but thought I would post it here to add my experience with CC 5.2.1.
The problem I have been having since 5.1.x, and now 5.2.1, is the HTTP Proxy. Whenever I try to use it I get a error code of 500. I can connect fine with CC 4.72, but the new proxy procedures do not seem to talk to my company's proxy. Does not matter which of the 5.x.x version CC I use, I get the the 500 error; switch back to version 4.72 CC and I can get through the proxy without any problems.
This is the error message that 5.x.x gives me:
10/17/2005 3:27:16 PM|Einstein@Home|Scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi failed with a return value of 500
10/17/2005 3:27:16 PM|Einstein@Home|No schedulers responded
10/17/2005 3:27:21 PM|Einstein@Home|Deferring communication with project for 54 seconds
RE: RE: think before you
)
I hope that the higher power above will realize some of the things I've said and done were stupid and moderate those things as well. We all do things at some point in our lives that we regret doing or saying. It's human nature. I try to observe things from more than one perspective as best I can before I make an opinion or a statement. It's one of the things that I learned from a leadership development course while I was in the service. -It may be something small to you, but it will mean the whole world to someone else.
Happy crunching to you too :-)
RE: RE: RE: think
)
In retrospect, I wish that I could have eaten a few words, (ahem, er, "asleep-at-the-switch user..." when a wiser mind later pointed out that the user had a problem of which he/she was not aware) back from a forum. We all slip sometimes. We all look around to see if anyone noticed, and pick ourselves back up.
microcraft
"The arc of history is long, but it bends toward justice" - MLK
I am an Alpha Tester for
)
I am an Alpha Tester for BOINC and there is a place that we report issues with any version that we test, per the testing guidelines.
I recommend that you read the BOINC Dev boards to stay aware of potential problems/issues before downloading any newer version.
Post your problems/bugs/issues with the BOINC client there.
RE: I am an Alpha Tester
)
Kajun,
Thanks for putting up that link.
I have checked those boards every couple of days, but found precious little regarding 5.2.1 there.
I noticed Metod's comment about upload delays. Running solely Einstein, I have not seen this behavior, nor did he, and I deemed it unnecessary to post a "me too" comment over there when I could add nothing further.
I crunched SetiClassic starting 2002-12-03 (under "microcraft") until May last year, when I took a hard look at it and concluded that my work there had been just about the most useless waste of electricity, except for contributing to a demonstration of the power of Distributed Computing. The likelihood of discovering, let alone communicating with an ETI is miniscule, given the probable span of lifeforms and the vast distance/time involved, *unless* they happen to be patroling our solar system and observing us, in which case, why would they possibly want communication with such a violent, erratic species unless we looked like food (nah, too polluted) or slaves (also unlikely, too unruly). (just my personal conclusion about MY time with Seti, NOT directed at other crunchers. People, don't even waste the flames here - this is a testing thread, and I refuse to rise to the bait) I'm not interested at this point in crunching for other projects, so my testing will be limited to the Windows/Boinc/Einstein combination, and of little interest to anyone outside the Einstein community, ergo the thread here.
Regards
Michael
microcraft
"The arc of history is long, but it bends toward justice" - MLK
Posting a "it happened to me
)
Posting a "it happened to me too" comment makes the developers aware that the situation is happening on more than one system.
If there isn't a thread related to your issue, create a thread.
Hello Michael As promised
)
Hello Michael
As promised I'm back here with another BOINC 5.2.1 issue. Right now this is just my experience so don't feel up to finding a dev list to post anything formal - it's been a hard Monday doing jigsawpuzzle adverts for assorted freebe newspapers and I'm tired, but anyway that's not BOINC.
OK the issue I noticed is that although I have set the preferences for SETI@home and for Einstein@home to run one hour each, when I staggered through the door, I saw that SETI had hogged all the crunching time - I had to actually suspend seti manually to get Einstein running - this over 9 hours or so. It is possible that this has happened due to the malfunctioning password issue of my previous post. I will try detaching from Einstein shortly and then rejoing with the password to see if that makes any difference.
Gray
RE: Hello Michael As
)
Gray,
It is possible that the issue you mention can be attributed to the correct operation of the scheduler, and goes back to the 4.x Boinc clients.
If you had a work queue full of Seti WUs when you added the Einstein project, the scheduler may be giving preference to Seti in order to establish equilibrium. That is, it is designed to enter Earliest Deadline First mode when it senses that your workload approaches full.
Did you notice something similar when you added a new project while running the previous Boinc version?
I think this may settle down after a few days of crunching, if the scheduler is left alone to do it's thing. Over-riding it's decision tends to delay the proper settling-in process. It will also depend upon what share values you have set for the various projects you run, and allot time according to those settings.
I may have done an inadequate job of describing the scheduler activity, or may have made an incorrect assumption; I have no experience in running multiple projects, and have learned from studying solutions that have been posted by more-experienced folks here. Perhaps one of them will drop by and give a better answer.
Please keep checking in with your reports. :-)
Hope it's helpful!
Michael
microcraft
"The arc of history is long, but it bends toward justice" - MLK
RE: RE: Hello Michael As
)
Hello Michael
Yes that info WAS useful - I had forgotten that ability. I did have 35-odd seti@home WUs listed before I joined Einstein (seti seems to have a somewhat over-sized idea of the power of this lil box).
What I have done, since my last post, now that I have a functioning password for Einstein is to upload what was crunched, unattach from everything, uninstall BOINC, trash BOINC folder and reinstall BOINC, then re-attach to Einstein first. I also now have a password for Climate, so have attached to that one
- the deadline is a realistic year away (should manage that I reckon :) ) - instead of seti@home. Watch this space. I will leave the system running through the night (thank goodness for Zalman fans...) and let you know what happens Tuesday evening after this slave returns home.
Gray
Just come across this thread.
)
Just come across this thread. The following problem I have been reporting in the SETI Beta message boards, but thought I would post it here to add my experience with CC 5.2.1.
The problem I have been having since 5.1.x, and now 5.2.1, is the HTTP Proxy. Whenever I try to use it I get a error code of 500. I can connect fine with CC 4.72, but the new proxy procedures do not seem to talk to my company's proxy. Does not matter which of the 5.x.x version CC I use, I get the the 500 error; switch back to version 4.72 CC and I can get through the proxy without any problems.
This is the error message that 5.x.x gives me:
10/17/2005 3:27:16 PM|Einstein@Home|Scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi failed with a return value of 500
10/17/2005 3:27:16 PM|Einstein@Home|No schedulers responded
10/17/2005 3:27:21 PM|Einstein@Home|Deferring communication with project for 54 seconds
You like Myst? Uru Live returns! www.urulive.com
RE: Just come across this
)
Again, these problems should be reported on the http://boinc.berkeley.edu/dev/ site.
It says there:
Please use these boards only for messages related to BOINC software: questions, bug reports, feature requests, etc.
Posting to these boards requires creating an account. You don't need an account to read.
so just go there and read, or post a problem, if you don't see it described anywhere!
So nobody are obliged to be content with the help they can get here, there are many qualified people out there to help!