I have a problem with versions 4.14, .15 and .16: I cannot update any projects - that is, the schedulars don't respond (S@H, CPDN, E@H, Predictor). WU'S download, run, and upload just fine.
I have tried this on 2 different computers, same result.
When I restore to 4.13 the problem disappears.
Running Win2000 SP4 on both boxes. Suggestions appreciated....
Updated 4.14 to 4.16 on one machine, everything went smooth and is running fine, but Predictor has yet to request another WU. I also just updated my prefs (both general and E@H) and that works fine as well.
I upgraded 1 PC last night to v4.16 with no problems so this morning I tried another one and it got hosed, it trashed the WU's in progress and all the waiting work units when BOINC restarted after the new version installation & gave me new ones ... ???
2005-01-21 07:40:58 - Starting BOINC client version 4.16 for windows_intelx86
Einstein@Home - 2005-01-21 07:40:58 - Project prefs: no separate prefs for home; using your defaults
--- - 2005-01-21 07:40:58 - H1_0121.4__0121.6_0.1_T01_Test02: error in element
--- - 2005-01-21 07:40:58 - Can't parse workunit in state file
--- - 2005-01-21 07:40:58 - State file has different major version (0.00); resetting projects
Einstein@Home - 2005-01-21 07:40:58 - Resetting project
Einstein@Home - 2005-01-21 07:40:58 - Host ID is 5409
--- - 2005-01-21 07:40:58 - General prefs: from Einstein@Home (last modified 2005-01-19 23:22:04)
--- - 2005-01-21 07:40:58 - General prefs: no separate prefs for home; using your defaults
--- - 2005-01-21 07:40:59 - Insufficient work; requesting more
Einstein@Home - 2005-01-21 07:40:59 - Requesting 344710 seconds of work
Einstein@Home - 2005-01-21 07:40:59 - Sending request to scheduler: http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi
Einstein@Home - 2005-01-21 07:41:51 - Scheduler RPC to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi succeeded
Einstein@Home - 2005-01-21 07:41:51 - Project prefs: no separate prefs for home; using your defaults
Einstein@Home - 2005-01-21 07:41:51 - Started download of einstein_4.72_windows_intelx86.exe
Einstein@Home - 2005-01-21 07:41:51 - Started download of einstein_4.72_windows_intelx86.pdb
Einstein@Home - 2005-01-21 07:42:05 - Finished download of einstein_4.72_windows_intelx86.exe
Einstein@Home - 2005-01-21 07:42:05 - Throughput 132474 bytes/sec
Einstein@Home - 2005-01-21 07:42:07 - Finished download of einstein_4.72_windows_intelx86.pdb
Einstein@Home - 2005-01-21 07:42:07 - Throughput 200490 bytes/sec
Einstein@Home - 2005-01-21 07:42:07 - Starting result H1_0121.4__0121.7_0.1_T08_Test02_0 using einstein version 4.72
Einstein@Home - 2005-01-21 07:42:07 - Starting result H1_0121.4__0121.8_0.1_T08_Test02_0 using einstein version 4.72
PS: Out of my 7 PC's 4 of them installed the v4.16 ok & 3 of them gave me the above error's, I only lost work on the 1 PC though ...
> I upgraded 1 PC last night to v4.16 with no problems so this morning I tried
> another one and it got hosed, it trashed the WU's in progress and all the
> waiting work units when BOINC restarted after the new version installation
> & gave me new ones ... ???
>
I had the same problem. BOINC 4.16 restarted the project, dumped the WU's, then downloaded E@H 4.72. Running fine otherwise.
> Damn
>
> 4.14 and 4.15 had no problems...but 4.16 had the following...basically it
> screwed cpdn
>
> - Starting BOINC client version 4.16 for windows_intelx86
> - Version Change Detected (4.15 -> 4.16); running CPU benchmarks
> - Scheduler RPC to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi
> succeeded
> - Can't resolve hostname climateprediction.net
> (valid name, no data record of requested type)
> - Scheduler op: init_master_fetch failed.
Got a similiar error when installing the 4.15 boinc on a new machine. It was a proxy problem. I'm behind an HTTP proxy and the internal DNS is not resolving for external domains (only the proxy itself can do that, but a ping or nslookup won't work).
For some reason the proxy settings where gone. I used an older client_state.xml, which I drop into the boinc folder after installing a new fresh client. (I saved and edited it, removed everything except for the proxy configs, and stored in my install folder on the LAN). After entering teh proxy settings again, it worked again.
So are you using a proxy, it might be this. I don't know why the proxy settings where not read correctly from the old client_state.xml the [proxy] segment in the new one looked the same, after reentering the proxy data (proxy.domain.nl at port 80, username and password for the proxy entered too). It used to work for older clients that way.
Getting an unusual situation with 4.16... Or maybe it's just an update bug:
I've just updated, and the total time to completion for all my units (whether they're untouched or 93% done (CP) give the TOTAL time to completion, so my CP unit says I've been working on it for 839 hours and that it's 867 hours to completion.
Will all these numbers fix once they resume or is this just another oddity of 4.16??
> Getting an unusual situation with 4.16... Or maybe it's just an update bug:
>
> I've just updated, and the total time to completion for all my units (whether
> they're untouched or 93% done (CP) give the TOTAL time to completion, so my CP
> unit says I've been working on it for 839 hours and that it's 867 hours to
> completion.
>
> Will all these numbers fix once they resume or is this just another oddity of
> 4.16??
After it resumes, it should show the correct time remaining again.
I have updated to 4.16 with no visible signs of problems. I do run ProteinPredictor as well as all the others so it will be interesting to see when I need more work if this happens or I get the error.
No probs with predictor for
)
No probs with predictor for me
Installed the new client and five minutes later got some work from predictor
I have a problem with
)
I have a problem with versions 4.14, .15 and .16: I cannot update any projects - that is, the schedulars don't respond (S@H, CPDN, E@H, Predictor). WU'S download, run, and upload just fine.
I have tried this on 2 different computers, same result.
When I restore to 4.13 the problem disappears.
Running Win2000 SP4 on both boxes. Suggestions appreciated....
I just tried an update and it
)
I just tried an update and it succeeded fine. Running 4.16, but on XPSP2. Have a Win2K machine at the office I can try this on tomorrow. Anyone else?
Updated 4.14 to 4.16 on one
)
Updated 4.14 to 4.16 on one machine, everything went smooth and is running fine, but Predictor has yet to request another WU. I also just updated my prefs (both general and E@H) and that works fine as well.
- Marco
Team Canada
I upgraded 1 PC last night to
)
I upgraded 1 PC last night to v4.16 with no problems so this morning I tried another one and it got hosed, it trashed the WU's in progress and all the waiting work units when BOINC restarted after the new version installation & gave me new ones ... ???
2005-01-21 07:40:58 - Starting BOINC client version 4.16 for windows_intelx86
Einstein@Home - 2005-01-21 07:40:58 - Project prefs: no separate prefs for home; using your defaults
--- - 2005-01-21 07:40:58 - H1_0121.4__0121.6_0.1_T01_Test02: error in element
--- - 2005-01-21 07:40:58 - Can't parse workunit in state file
--- - 2005-01-21 07:40:58 - State file has different major version (0.00); resetting projects
Einstein@Home - 2005-01-21 07:40:58 - Resetting project
Einstein@Home - 2005-01-21 07:40:58 - Host ID is 5409
--- - 2005-01-21 07:40:58 - General prefs: from Einstein@Home (last modified 2005-01-19 23:22:04)
--- - 2005-01-21 07:40:58 - General prefs: no separate prefs for home; using your defaults
--- - 2005-01-21 07:40:59 - Insufficient work; requesting more
Einstein@Home - 2005-01-21 07:40:59 - Requesting 344710 seconds of work
Einstein@Home - 2005-01-21 07:40:59 - Sending request to scheduler: http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi
Einstein@Home - 2005-01-21 07:41:51 - Scheduler RPC to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi succeeded
Einstein@Home - 2005-01-21 07:41:51 - Project prefs: no separate prefs for home; using your defaults
Einstein@Home - 2005-01-21 07:41:51 - Started download of einstein_4.72_windows_intelx86.exe
Einstein@Home - 2005-01-21 07:41:51 - Started download of einstein_4.72_windows_intelx86.pdb
Einstein@Home - 2005-01-21 07:42:05 - Finished download of einstein_4.72_windows_intelx86.exe
Einstein@Home - 2005-01-21 07:42:05 - Throughput 132474 bytes/sec
Einstein@Home - 2005-01-21 07:42:07 - Finished download of einstein_4.72_windows_intelx86.pdb
Einstein@Home - 2005-01-21 07:42:07 - Throughput 200490 bytes/sec
Einstein@Home - 2005-01-21 07:42:07 - Starting result H1_0121.4__0121.7_0.1_T08_Test02_0 using einstein version 4.72
Einstein@Home - 2005-01-21 07:42:07 - Starting result H1_0121.4__0121.8_0.1_T08_Test02_0 using einstein version 4.72
PS: Out of my 7 PC's 4 of them installed the v4.16 ok & 3 of them gave me the above error's, I only lost work on the 1 PC though ...
> I upgraded 1 PC last night
)
> I upgraded 1 PC last night to v4.16 with no problems so this morning I tried
> another one and it got hosed, it trashed the WU's in progress and all the
> waiting work units when BOINC restarted after the new version installation
> & gave me new ones ... ???
>
I had the same problem. BOINC 4.16 restarted the project, dumped the WU's, then downloaded E@H 4.72. Running fine otherwise.
> Damn > > 4.14 and 4.15 had
)
> Damn
>
> 4.14 and 4.15 had no problems...but 4.16 had the following...basically it
> screwed cpdn
>
> - Starting BOINC client version 4.16 for windows_intelx86
> - Version Change Detected (4.15 -> 4.16); running CPU benchmarks
> - Scheduler RPC to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi
> succeeded
> - Can't resolve hostname climateprediction.net
> (valid name, no data record of requested type)
> - Scheduler op: init_master_fetch failed.
Got a similiar error when installing the 4.15 boinc on a new machine. It was a proxy problem. I'm behind an HTTP proxy and the internal DNS is not resolving for external domains (only the proxy itself can do that, but a ping or nslookup won't work).
For some reason the proxy settings where gone. I used an older client_state.xml, which I drop into the boinc folder after installing a new fresh client. (I saved and edited it, removed everything except for the proxy configs, and stored in my install folder on the LAN). After entering teh proxy settings again, it worked again.
So are you using a proxy, it might be this. I don't know why the proxy settings where not read correctly from the old client_state.xml the [proxy] segment in the new one looked the same, after reentering the proxy data (proxy.domain.nl at port 80, username and password for the proxy entered too). It used to work for older clients that way.
Roelof
Getting an unusual situation
)
Getting an unusual situation with 4.16... Or maybe it's just an update bug:
I've just updated, and the total time to completion for all my units (whether they're untouched or 93% done (CP) give the TOTAL time to completion, so my CP unit says I've been working on it for 839 hours and that it's 867 hours to completion.
Will all these numbers fix once they resume or is this just another oddity of 4.16??
> Getting an unusual
)
> Getting an unusual situation with 4.16... Or maybe it's just an update bug:
>
> I've just updated, and the total time to completion for all my units (whether
> they're untouched or 93% done (CP) give the TOTAL time to completion, so my CP
> unit says I've been working on it for 839 hours and that it's 867 hours to
> completion.
>
> Will all these numbers fix once they resume or is this just another oddity of
> 4.16??
After it resumes, it should show the correct time remaining again.
I have updated to 4.16 with
)
I have updated to 4.16 with no visible signs of problems. I do run ProteinPredictor as well as all the others so it will be interesting to see when I need more work if this happens or I get the error.