Keith:
with all the new versions and problems popping up in CPDN and Predictor, I've been following the developers mailing list. The 4.16 is brand new and I'm sure will be available here soon.
It's just that before updating all your clients to 4.15 it might be better to wait for 4.16, or for a comment by the project admins that they'd rather have 4.15 installed.
> Damn
>
> 4.14 and 4.15 had no problems...but 4.16 had the following...basically it
> screwed cpdn
>
EDIT: corrected a little mistake, thanks Genes for pointing this out
AFAIK, this *SHOULD* be a relatively safe method of updating BOINC for Windows, I haven't tested it yet because I've never had any problems (I'm still on 4.14, I'm moving to 4.16 as soon as it appears on the E@H download page).
NOTE: I use the | instead of the backslash here (gets removed by the forum)
suspend/quit BOINC
use task manager to verify all science processes are gone
copy entire BOINC folder i.e. "C:|Program Files|BOINC|" -> "C:|Program Files|BOINC415|"
update BOINC to 4.16
when you get an error like just reported
suspend/quit BOINC
use task manager to verify all science processes are killed (if not, kill them)
rename "C:|Program Files|BOINC|" -> "C:|Program Files|BOINC416|"
rename "C:|Program Files|BOINC415| -> "C:|Program Files|BOINC|"
Start BOINC and, to use the Guardian of Forever's words: "All is as it was" (or it should be).
The BOINC416 folder is preserved, so please report what's in stderr.txt
Hmm, I'm on 4.15 at the moment, on both of my home machines. Since upgrading to 4.14, then 4.15, I've seen both of these reliable machines freeze in exactly the same way (although I'm not sure I can pin the blame on BOINC) -- you go to sleep, the machine is running, you look at the machine in the morning and it is dead. All you can do is push (and hold) the power button, and restart. No "Windows recovered from a horrible error" or anything, they are keeping mum.
So, with that in mind, I might as well try 4.16, but I'll re-order Marco's steps and copy the folder *AFTER* stopping Boinc and making sure everything's stopped.
--------------
EDIT - Two running OK on 4.16 so far.
Upgraded from 4.15 to 4.16. It killed three predictor WUs in my queue and one waiting to be returned.
I wasn't going to worry about that until I saw that it apparently brought back the dreaded, "there was work, but it was commited to other platforms" message from Predictor when trying to get more work.
Went back to 4.15 and restored my backup. WUs are back, returned the one waiting and Predictor is currently downloading more work.
Be advised if you run Predictor, you may have trouble getting work.
And yes, I've been running BOINC since back in beta, so I know how to install a new CC.
BOINC Client v4.16 released
)
Silly question time... Marco
Do we get the new version from
http://einstein.phys.uwm.edu/download.html ?
I noticed the version there is still 4.15.
> Silly question time...
)
> Silly question time... Marco
>
> Do we get the new version from
>
> http://einstein.phys.uwm.edu/download.html ?
>
> I noticed the version there is still 4.15.
>
Keith:
with all the new versions and problems popping up in CPDN and Predictor, I've been following the developers mailing list. The 4.16 is brand new and I'm sure will be available here soon.
It's just that before updating all your clients to 4.15 it might be better to wait for 4.16, or for a comment by the project admins that they'd rather have 4.15 installed.
- Marco
Team Canada
Thanks for that Marco.
)
Thanks for that Marco.
4.16 downloads
)
4.16 downloads
I am tempted to D/L 4.16
)
I am tempted to D/L 4.16 Ric
The last 2 versions have not worked for me yet (the uploading errors)
I will check back tomorrow and see if they say it is bug free.
Damn 4.14 and 4.15 had no
)
Damn
4.14 and 4.15 had no problems...but 4.16 had the following...basically it screwed cpdn
--- - 2005-01-20 20:15:39 - Starting BOINC client version 4.16 for windows_intelx86
climateprediction.net - 2005-01-20 20:15:39 - Project prefs: using your defaults
Einstein@Home - 2005-01-20 20:15:39 - Project prefs: using your defaults
--- - 2005-01-20 20:15:39 - Version Change Detected (4.15 -> 4.16); running CPU benchmarks
climateprediction.net - 2005-01-20 20:15:39 - Host ID is xxxxxx
Einstein@Home - 2005-01-20 20:15:39 - Host ID is xxxx
--- - 2005-01-20 20:15:39 - General prefs: from climateprediction.net (last modified 2005-01-12 09:07:11)
--- - 2005-01-20 20:15:39 - General prefs: using your defaults
--- - 2005-01-20 20:15:45 - Running CPU benchmarks
--- - 2005-01-20 20:16:46 - Benchmark results:
--- - 2005-01-20 20:16:46 - Number of CPUs: 2
--- - 2005-01-20 20:16:46 - 1355 double precision MIPS (Whetstone) per CPU
--- - 2005-01-20 20:16:46 - 2510 integer MIPS (Dhrystone) per CPU
--- - 2005-01-20 20:16:46 - Finished CPU benchmarks
climateprediction.net - 2005-01-20 20:16:47 - Resuming computation for result 1x4g_100110498_2 using hadsm3 version 4.04
climateprediction.net - 2005-01-20 20:16:47 - Resuming computation for result 2yjr_000159480_1 using hadsm3 version 4.04
Einstein@Home - 2005-01-20 20:16:47 - Deferring computation for result H1_0054.9__0055.4_0.1_T00_Test02_3
Einstein@Home - 2005-01-20 20:16:47 - Deferring computation for result H1_0054.9__0055.0_0.1_T01_Test02_2
climateprediction.net - 2005-01-20 20:17:26 - Sending request to scheduler: http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi
climateprediction.net - 2005-01-20 20:17:31 - Scheduler RPC to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi succeeded
climateprediction.net - 2005-01-20 20:17:31 - Project prefs: using your defaults
--- - 2005-01-20 20:17:45 - Can't resolve hostname climateprediction.net (valid name, no data record of requested type)
climateprediction.net - 2005-01-20 20:17:45 - Scheduler op: init_master_fetch failed.
climateprediction.net - 2005-01-20 20:29:37 - Unrecoverable error for result 2yjr_000159480_1 ( - exit code -5 (0xfffffffb))
climateprediction.net - 2005-01-20 20:29:37 - Computation for result 2yjr_000159480 finished
Einstein@Home - 2005-01-20 20:29:37 - Restarting result H1_0054.9__0055.4_0.1_T00_Test02_3 using einstein version 4.71
climateprediction.net - 2005-01-20 20:29:40 - Started upload of 2yjr_000159480_1_1.zip
climateprediction.net - 2005-01-20 20:29:40 - Started upload of 2yjr_000159480_1_2.zip
climateprediction.net - 2005-01-20 20:29:46 - Finished upload of 2yjr_000159480_1_1.zip
climateprediction.net - 2005-01-20 20:29:46 - Throughput 256 bytes/sec
climateprediction.net - 2005-01-20 20:29:46 - Started upload of 2yjr_000159480_1_3.zip
climateprediction.net - 2005-01-20 20:29:49 - Finished upload of 2yjr_000159480_1_2.zip
climateprediction.net - 2005-01-20 20:29:49 - Throughput 7857 bytes/sec
climateprediction.net - 2005-01-20 20:29:49 - Started upload of 2yjr_000159480_1_4.zip
climateprediction.net - 2005-01-20 20:29:52 - Finished upload of 2yjr_000159480_1_3.zip
climateprediction.net - 2005-01-20 20:29:52 - Throughput 254 bytes/sec
climateprediction.net - 2005-01-20 20:29:52 - Started upload of 2yjr_000159480_1_5.zip
climateprediction.net - 2005-01-20 20:29:55 - Finished upload of 2yjr_000159480_1_4.zip
climateprediction.net - 2005-01-20 20:29:55 - Throughput 254 bytes/sec
climateprediction.net - 2005-01-20 20:30:01 - Finished upload of 2yjr_000159480_1_5.zip
climateprediction.net - 2005-01-20 20:30:01 - Throughput 3460 bytes/sec
climateprediction.net - 2005-01-20 20:30:17 - Unrecoverable error for result 1x4g_100110498_2 ( - exit code -5 (0xfffffffb))
climateprediction.net - 2005-01-20 20:30:17 - Computation for result 1x4g_100110498 finished
Einstein@Home - 2005-01-20 20:30:17 - Restarting result H1_0054.9__0055.0_0.1_T01_Test02_2 using einstein version 4.71
climateprediction.net - 2005-01-20 20:30:18 - Started upload of 1x4g_100110498_2_1.zip
climateprediction.net - 2005-01-20 20:30:18 - Started upload of 1x4g_100110498_2_2.zip
climateprediction.net - 2005-01-20 20:30:24 - Finished upload of 1x4g_100110498_2_1.zip
climateprediction.net - 2005-01-20 20:30:24 - Throughput 254 bytes/sec
climateprediction.net - 2005-01-20 20:30:24 - Started upload of 1x4g_100110498_2_3.zip
climateprediction.net - 2005-01-20 20:30:27 - Finished upload of 1x4g_100110498_2_2.zip
climateprediction.net - 2005-01-20 20:30:27 - Throughput 3126 bytes/sec
climateprediction.net - 2005-01-20 20:30:27 - Started upload of 1x4g_100110498_2_4.zip
climateprediction.net - 2005-01-20 20:30:30 - Finished upload of 1x4g_100110498_2_3.zip
climateprediction.net - 2005-01-20 20:30:30 - Throughput 250 bytes/sec
climateprediction.net - 2005-01-20 20:30:30 - Started upload of 1x4g_100110498_2_5.zip
climateprediction.net - 2005-01-20 20:30:33 - Finished upload of 1x4g_100110498_2_4.zip
climateprediction.net - 2005-01-20 20:30:33 - Throughput 252 bytes/sec
climateprediction.net - 2005-01-20 20:30:39 - Finished upload of 1x4g_100110498_2_5.zip
climateprediction.net - 2005-01-20 20:30:39 - Throughput 3510 bytes/sec
Einstein@Home - 2005-01-20 21:30:17 - Pausing result H1_0054.9__0055.4_0.1_T00_Test02_3 (left in memory)
Einstein@Home - 2005-01-20 21:30:17 - Pausing result H1_0054.9__0055.0_0.1_T01_Test02_2 (left in memory)
Bugger,
Paul
> 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
>
EDIT: corrected a little mistake, thanks Genes for pointing this out
AFAIK, this *SHOULD* be a relatively safe method of updating BOINC for Windows, I haven't tested it yet because I've never had any problems (I'm still on 4.14, I'm moving to 4.16 as soon as it appears on the E@H download page).
NOTE: I use the | instead of the backslash here (gets removed by the forum)
suspend/quit BOINC
use task manager to verify all science processes are gone
copy entire BOINC folder i.e. "C:|Program Files|BOINC|" -> "C:|Program Files|BOINC415|"
update BOINC to 4.16
when you get an error like just reported
suspend/quit BOINC
use task manager to verify all science processes are killed (if not, kill them)
rename "C:|Program Files|BOINC|" -> "C:|Program Files|BOINC416|"
rename "C:|Program Files|BOINC415| -> "C:|Program Files|BOINC|"
Start BOINC and, to use the Guardian of Forever's words: "All is as it was" (or it should be).
The BOINC416 folder is preserved, so please report what's in stderr.txt
- Marco
Team Canada
Hmm, I'm on 4.15 at the
)
Hmm, I'm on 4.15 at the moment, on both of my home machines. Since upgrading to 4.14, then 4.15, I've seen both of these reliable machines freeze in exactly the same way (although I'm not sure I can pin the blame on BOINC) -- you go to sleep, the machine is running, you look at the machine in the morning and it is dead. All you can do is push (and hold) the power button, and restart. No "Windows recovered from a horrible error" or anything, they are keeping mum.
So, with that in mind, I might as well try 4.16, but I'll re-order Marco's steps and copy the folder *AFTER* stopping Boinc and making sure everything's stopped.
--------------
EDIT - Two running OK on 4.16 so far.
Just upgraded from 4.15 to
)
Just upgraded from 4.15 to 4.16 and everything is runnig fine
Upgraded from 4.15 to 4.16.
)
Upgraded from 4.15 to 4.16. It killed three predictor WUs in my queue and one waiting to be returned.
I wasn't going to worry about that until I saw that it apparently brought back the dreaded, "there was work, but it was commited to other platforms" message from Predictor when trying to get more work.
Went back to 4.15 and restored my backup. WUs are back, returned the one waiting and Predictor is currently downloading more work.
Be advised if you run Predictor, you may have trouble getting work.
And yes, I've been running BOINC since back in beta, so I know how to install a new CC.