YIKES, Stick. I didn't mean to step on your toes - I don't know anything further to look at, as you've already covered more than I would have thought of.
Michael,
If you were trying to step on my toes, you MISSED badly. I am very glad you jumped in. However, if we have truly run your string, too, then I hope we attract some more help.
I also remembered reading some things about ports and firewalls recently - so I did a keyword search on "port". There were a bunch of hits. I copied this one from Ageless:
Quote:
TCP port 80 outbound from BOINC.exe is the only port that should be needed to communicate with the project servers.
TCP Ports 1043 and 31416 inbound for BOINC.exe and outbound for BOINCMgr.exe are used for remote control of the daemon (boinc.exe) by the manager (BOINCMgr.exe). It should not be nessecary to allow these ports to be accessed from internet.
Most OSes can use 127.0.0.1 without opening a port on the host, but some OSes will not allow the distinction between the loopback (127.0.0.1) address and a general internet address, and the machine must allow these ports to be open to the LAN.
However, my gut feeling is that it's not a firewall problem.
I'm having a similar problem, and don't understand a fair bit of what's written below. I have four machines all hooked into a local network - the main PC, the old clunker up in the attic that is the fileserver/backup, the Linux machine, and the laptop (left on to keep it synched for when I go out and about).
All found are running BOINC, all four have (as far as I can tell) the same configuration and all four are "subscribed" to Einstein. But only two actually RUN Einstein - the main PC, and the laptop. The Linux box reports "Deferring communication with project for ... hours" and has done that for weeks. Meanwhile, it is happily crunching away at Seti.
The attic machine seems to talk to the Einstein server, but never does any work for it.
Should I care? What should I do to get the other two to work?
I'm having a similar problem, and don't understand a fair bit of what's written below. I have four machines all hooked into a local network - the main PC, the old clunker up in the attic that is the fileserver/backup, the Linux machine, and the laptop (left on to keep it synched for when I go out and about).
All found are running BOINC, all four have (as far as I can tell) the same configuration and all four are "subscribed" to Einstein. But only two actually RUN Einstein - the main PC, and the laptop. The Linux box reports "Deferring communication with project for ... hours" and has done that for weeks. Meanwhile, it is happily crunching away at Seti.
The attic machine seems to talk to the Einstein server, but never does any work for it.
Should I care? What should I do to get the other two to work?
Since your Einstein preferences allow "viewing" the Einstein data from your computers, we can tell a lot from the results you have returned. For instance, I can tell that your Linux box downloads a work unit every day and fails it immediately. (I haven't looked up the "failure" codes yet but that can be easily done and should provide insight into finding a fix. But, because the Linux box has failing all the units it downloads, Einstein has limited it's quota to 1 per day. That's basically why you get the "Deferring communication . . ." messages. As to the "old clunker", it appears it is getting a work unit every now and then, but it may not be fast enough to complete processing before the 2 week deadline. Your results for it indicate that speed may a possible problem. Is it also doing Seti? Your Intel(R) Pentium(R) 4 CPU 3.20GHz looks like it's doing splendidly. And the Intel(R) Celeron(TM) CPU 1133MHz is showing 1 unit in progress (I presume it's the laptop which you say is also working). I will get back to you on the error codes for the Linux - but, since I am a Windows guy, we may have to wait on a Linux expert to respond with a possible fix.
The Linux box reports "Deferring communication with project for ... hours" and has done that for weeks.
All your viewable Linux box results report the following:
Quote:
4.43
Couldn't start or resume: -108
Exit Status: -185 (0xffffff47)
The BOINC Wiki says the following for this type message:
Quote:
The BOINC Daemon was not able to start the Science Application to begin or resume processing on the Result. This could be the result of an error in the Science Application or corruption on the Disk Drive leading to errors in the Work Unit Data File(s), Result Data File(s), or the State File.
Hopefully, a Linux user will see this soon and advise you what to do.
Since your Einstein preferences allow "viewing" the Einstein data from your computers, we can tell a lot from the results you have returned. For instance, I can tell that your Linux box downloads a work unit every day and fails it immediately.
Ah! That triggered a memory. Linux (or at least the incarnation I'm running) is paranoid, and puts very high security levels on everything the "root system" creates or just works on. It's a right pain, because it causes all sorts of strange errors.
So I sign up for Einstein, which works fine for a while. Then I do something else, and the "control" file gets edited, whereupon Linux changes it to root access only, read-only. That is very difficult when the file in question is the datafile it's supposed to be writing to and working on.
I've just changed the ownership to "me" and the permissions on the file to "all can read and write". I think that may fix that, when the embargo on new work comes off.
Quote:
As to the "old clunker", it appears it is getting a work unit every now and then, but it may not be fast enough to complete processing before the 2 week deadline. Your results for it indicate that speed may a possible problem. Is it also doing Seti?
It wasn't, until this morning. On the basis it wasn't doing anything with Einstein, I added Seti to it. It's a P200MMX, so a fair bit slower than the other machines, but it works fine as a fileserver. I'll leave it for a bit to see how it does.
Quote:
Your Intel(R) Pentium(R) 4 CPU 3.20GHz looks like it's doing splendidly.
That's the desktop, relatively new, and the busy machine!
Quote:
And the Intel(R) Celeron(TM) CPU 1133MHz is showing 1 unit in progress (I presume it's the laptop which you say is also working).
Exactly so. I'm away this weekend, taking the laptop with me, so I just set it to "no new work" so that it can clear itself.
Quote:
I will get back to you on the error codes for the Linux - but, since I am a Windows guy, we may have to wait on a Linux expert to respond with a possible fix.
Thanks - I suspect I've fixed that. The ownership and permissions of the Seti and the Einstein folders are now the same. I think that will prove to be the problem.
ok, stick, i really have an firewall (zonealarm) but it worked until now even with the firewall...so i don´t know if it is really the problem. besides i don´t know how to open or close any ports... but there is s.th. else: i changed a few days before my browser from opera to firefox. maybe this is the problem?
at the moment it looks like this when i pull the update button:
>>14.10.2005 10:07:09||request_reschedule_cpus: project op
14.10.2005 10:07:11|Einstein@Home|Sending scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi
14.10.2005 10:07:11|Einstein@Home|Requesting 0 seconds of work, returning 0 results
14.10.2005 10:07:12|Einstein@Home|Scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi succeeded
14.10.2005 11:03:08||Insufficient work; requesting more
14.10.2005 12:03:09||Insufficient work; requesting more>>
i don´t know...
bye, soho
Quote:
Quote:
Quote:
thanks stick,
for your help.
But do you think thats normal:
>>13.10.2005 11:13:45||Starting BOINC client version 4.45 for windows_intelx86
13.10.2005 11:13:45||Data directory: C:\\Programme\\BOINC
13.10.2005 11:13:45|Einstein@Home|Computer ID: 421093; location: home; project prefs: default
13.10.2005 11:13:45||General prefs: from Einstein@Home (last modified 2005-10-11 23:24:31)
13.10.2005 11:13:45||General prefs: no separate prefs for home; using your defaults
13.10.2005 11:13:45||Remote control not allowed; using loopback address
13.10.2005 11:13:45||Insufficient work; requesting more
13.10.2005 12:13:46||Insufficient work; requesting more
13.10.2005 13:13:47||Insufficient work; requesting more
13.10.2005 14:13:47||Insufficient work; requesting more
13.10.2005 15:13:48||Insufficient work; requesting more
13.10.2005 16:13:49||Insufficient work; requesting more>>???
i wonder why boinc knows that it needs more work to do but doesn´t upload a new WU...
i did three WU´s and everything worked fine until now...
Quote:
Quote:
stick,
i´m not running other projects. so it is wasted time to wait and to have nothing to work...
Soho,
You might read through this thread: Message boards : Problems and Bug Reports : BOINC is currently suspended - not exactly the same problem as yours but some similarities. You might get some ideas to try.
Stick
Sorry about the extra post - I seem to have a nervous "trigger" finger.
No, I don't think it's normal. And, quite frankly I'm stumped. That's why I pointed you to the other thread. One of the last suggestions there was to do a reinstall of BOINC using the "Repair" option. It seemed to work for that problem - but, it's really just "a shot in the dark" from me (although, I don't think it would hurt to try). As far as Michael's firewall question goes, I had dismissed the idea, because you seem to be communicating. But, it may be worth looking into anyway. I know BOINC uses some ports in addition to the standard port 80. If it's the firewall, it might be hanging one of the other ports. I don't if I can offer any more help, but our posts seem to have attracted some attention. I know Michael has been very helpful in other situations - so I will leave you in his very capable hands.
ok, stick, i really have an firewall (zonealarm) but it worked until now even with the firewall...so i don´t know if it is really the problem. besides i don´t know how to open or close any ports... but there is s.th. else: i changed a few days before my browser from opera to firefox. maybe this is the problem?
at the moment it looks like this when i pull the update button:
>>14.10.2005 10:07:09||request_reschedule_cpus: project op
14.10.2005 10:07:11|Einstein@Home|Sending scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi
14.10.2005 10:07:11|Einstein@Home|Requesting 0 seconds of work, returning 0 results
14.10.2005 10:07:12|Einstein@Home|Scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi succeeded
14.10.2005 11:03:08||Insufficient work; requesting more
14.10.2005 12:03:09||Insufficient work; requesting more>>
i don´t know...
bye, soho
Soho,
Michael and Stick have been doing a great job trying to help you (and others) in this thread. One of the difficulties, particularly when people chime in with "similar" problems, is really getting the accurate picture of all facets of your "particular" problem. In some ways, adding to an existing "similar" thread sometimes adds confusion because you tend to "mix together" symptoms from various "different" problems. I've found that I've had to read carefully and exclude some posts so as to focus on your particular problem.
Early on you mentioned that you were only running EAH. Is this really true? Do you by any chance have another project like Orbit or LHC which currently has no work? If you go to your projects tab of BOINC Manager, is there definately only one project? If you look in the projects folder inside your BOINC folder do you see ONLY the folder "einstein.phys.uwm.edu"?
Depending on your answer to this, I might be able to help.
Yes, gary, i have only one project running, E@H. in the folder is also just this one. i was running climateprediction a few weeks ago, but left it and erased all
data. it didn´t run without problems and so i changed to einstein because i thought it is a easier one...
maybe it is just my pc which makes the problems...and the fact that i am no pc-freak...
thanks and bye, soho.
Quote:
Quote:
ok, stick, i really have an firewall (zonealarm) but it worked until now even with the firewall...so i don´t know if it is really the problem. besides i don´t know how to open or close any ports... but there is s.th. else: i changed a few days before my browser from opera to firefox. maybe this is the problem?
at the moment it looks like this when i pull the update button:
>>14.10.2005 10:07:09||request_reschedule_cpus: project op
14.10.2005 10:07:11|Einstein@Home|Sending scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi
14.10.2005 10:07:11|Einstein@Home|Requesting 0 seconds of work, returning 0 results
14.10.2005 10:07:12|Einstein@Home|Scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi succeeded
14.10.2005 11:03:08||Insufficient work; requesting more
14.10.2005 12:03:09||Insufficient work; requesting more>>
i don´t know...
bye, soho
Soho,
Michael and Stick have been doing a great job trying to help you (and others) in this thread. One of the difficulties, particularly when people chime in with "similar" problems, is really getting the accurate picture of all facets of your "particular" problem. In some ways, adding to an existing "similar" thread sometimes adds confusion because you tend to "mix together" symptoms from various "different" problems. I've found that I've had to read carefully and exclude some posts so as to focus on your particular problem.
Early on you mentioned that you were only running EAH. Is this really true? Do you by any chance have another project like Orbit or LHC which currently has no work? If you go to your projects tab of BOINC Manager, is there definately only one project? If you look in the projects folder inside your BOINC folder do you see ONLY the folder "einstein.phys.uwm.edu"?
Depending on your answer to this, I might be able to help.
Will you please do this for me. In your BOINC folder you will find a file called "client_state.xml". Can you carefully open it with something like Windows notepad and look for the tags. You should certainly have one with Einstein stuff under it and you also may have one for CPDN. I want you to look for any "debt" values. In your version you should have both short term and long term debts. They may be abbreviated STD and LTD, I'm not familiar with your version. I'd like you to report any values you find anywhere in the file and also report what tag you found them under. When you finish just close the file without changing it in any way. Thanks.
Ah! That triggered a memory. Linux (or at least the incarnation I'm running) is paranoid, and puts very high security levels on everything the "root system" creates or just works on. It's a right pain, because it causes all sorts of strange errors.
So I sign up for Einstein, which works fine for a while. Then I do something else, and the "control" file gets edited, whereupon Linux changes it to root access only, read-only. That is very difficult when the file in question is the datafile it's supposed to be writing to and working on.
I've just changed the ownership to "me" and the permissions on the file to "all can read and write". I think that may fix that, when the embargo on new work comes off.
Thanks - I suspect I've fixed that. The ownership and permissions of the Seti and the Einstein folders are now the same. I think that will prove to be the problem.
Keef,
I noticed that your Linux box downloaded 4 units today - so whatever you did had some effect. However, they all failed immediately, again, just like the others (i.e. the exact same messages/exit codes).
RE: YIKES, Stick. I
)
Michael,
If you were trying to step on my toes, you MISSED badly. I am very glad you jumped in. However, if we have truly run your string, too, then I hope we attract some more help.
I also remembered reading some things about ports and firewalls recently - so I did a keyword search on "port". There were a bunch of hits. I copied this one from Ageless:
However, my gut feeling is that it's not a firewall problem.
I'm having a similar problem,
)
I'm having a similar problem, and don't understand a fair bit of what's written below. I have four machines all hooked into a local network - the main PC, the old clunker up in the attic that is the fileserver/backup, the Linux machine, and the laptop (left on to keep it synched for when I go out and about).
All found are running BOINC, all four have (as far as I can tell) the same configuration and all four are "subscribed" to Einstein. But only two actually RUN Einstein - the main PC, and the laptop. The Linux box reports "Deferring communication with project for ... hours" and has done that for weeks. Meanwhile, it is happily crunching away at Seti.
The attic machine seems to talk to the Einstein server, but never does any work for it.
Should I care? What should I do to get the other two to work?
Keef, Essex or Norfolk, England
RE: I'm having a similar
)
Since your Einstein preferences allow "viewing" the Einstein data from your computers, we can tell a lot from the results you have returned. For instance, I can tell that your Linux box downloads a work unit every day and fails it immediately. (I haven't looked up the "failure" codes yet but that can be easily done and should provide insight into finding a fix. But, because the Linux box has failing all the units it downloads, Einstein has limited it's quota to 1 per day. That's basically why you get the "Deferring communication . . ." messages. As to the "old clunker", it appears it is getting a work unit every now and then, but it may not be fast enough to complete processing before the 2 week deadline. Your results for it indicate that speed may a possible problem. Is it also doing Seti? Your Intel(R) Pentium(R) 4 CPU 3.20GHz looks like it's doing splendidly. And the Intel(R) Celeron(TM) CPU 1133MHz is showing 1 unit in progress (I presume it's the laptop which you say is also working). I will get back to you on the error codes for the Linux - but, since I am a Windows guy, we may have to wait on a Linux expert to respond with a possible fix.
RE: The Linux box reports
)
All your viewable Linux box results report the following:
The BOINC Wiki says the following for this type message:
Hopefully, a Linux user will see this soon and advise you what to do.
RE: Since your Einstein
)
Ah! That triggered a memory. Linux (or at least the incarnation I'm running) is paranoid, and puts very high security levels on everything the "root system" creates or just works on. It's a right pain, because it causes all sorts of strange errors.
So I sign up for Einstein, which works fine for a while. Then I do something else, and the "control" file gets edited, whereupon Linux changes it to root access only, read-only. That is very difficult when the file in question is the datafile it's supposed to be writing to and working on.
I've just changed the ownership to "me" and the permissions on the file to "all can read and write". I think that may fix that, when the embargo on new work comes off.
It wasn't, until this morning. On the basis it wasn't doing anything with Einstein, I added Seti to it. It's a P200MMX, so a fair bit slower than the other machines, but it works fine as a fileserver. I'll leave it for a bit to see how it does.
That's the desktop, relatively new, and the busy machine!
Exactly so. I'm away this weekend, taking the laptop with me, so I just set it to "no new work" so that it can clear itself.
Thanks - I suspect I've fixed that. The ownership and permissions of the Seti and the Einstein folders are now the same. I think that will prove to be the problem.
Thanks!
Keef, Essex or Norfolk, England
ok, stick, i really have an
)
ok, stick, i really have an firewall (zonealarm) but it worked until now even with the firewall...so i don´t know if it is really the problem. besides i don´t know how to open or close any ports... but there is s.th. else: i changed a few days before my browser from opera to firefox. maybe this is the problem?
at the moment it looks like this when i pull the update button:
>>14.10.2005 10:07:09||request_reschedule_cpus: project op
14.10.2005 10:07:11|Einstein@Home|Sending scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi
14.10.2005 10:07:11|Einstein@Home|Requesting 0 seconds of work, returning 0 results
14.10.2005 10:07:12|Einstein@Home|Scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi succeeded
14.10.2005 11:03:08||Insufficient work; requesting more
14.10.2005 12:03:09||Insufficient work; requesting more>>
i don´t know...
bye, soho
RE: ok, stick, i really
)
Soho,
Michael and Stick have been doing a great job trying to help you (and others) in this thread. One of the difficulties, particularly when people chime in with "similar" problems, is really getting the accurate picture of all facets of your "particular" problem. In some ways, adding to an existing "similar" thread sometimes adds confusion because you tend to "mix together" symptoms from various "different" problems. I've found that I've had to read carefully and exclude some posts so as to focus on your particular problem.
Early on you mentioned that you were only running EAH. Is this really true? Do you by any chance have another project like Orbit or LHC which currently has no work? If you go to your projects tab of BOINC Manager, is there definately only one project? If you look in the projects folder inside your BOINC folder do you see ONLY the folder "einstein.phys.uwm.edu"?
Depending on your answer to this, I might be able to help.
Cheers,
Gary.
Yes, gary, i have only one
)
Yes, gary, i have only one project running, E@H. in the folder is also just this one. i was running climateprediction a few weeks ago, but left it and erased all
data. it didn´t run without problems and so i changed to einstein because i thought it is a easier one...
maybe it is just my pc which makes the problems...and the fact that i am no pc-freak...
thanks and bye, soho.
Will you please do this for
)
Will you please do this for me. In your BOINC folder you will find a file called "client_state.xml". Can you carefully open it with something like Windows notepad and look for the tags. You should certainly have one with Einstein stuff under it and you also may have one for CPDN. I want you to look for any "debt" values. In your version you should have both short term and long term debts. They may be abbreviated STD and LTD, I'm not familiar with your version. I'd like you to report any values you find anywhere in the file and also report what tag you found them under. When you finish just close the file without changing it in any way. Thanks.
Cheers,
Gary.
RE: Ah! That triggered a
)
Keef,
I noticed that your Linux box downloaded 4 units today - so whatever you did had some effect. However, they all failed immediately, again, just like the others (i.e. the exact same messages/exit codes).
Stick