climateprediction.net home page
Posts by glaesum

Posts by glaesum

1) Message boards : Number crunching : hadcm3n upload error (Message 50847)
Posted 18 Nov 2014 by glaesum
Post:
Interesting. Another person posted about the same problem about a week ago.

However, that IS the re-start server, which isn't listed on the Server Status page. Although I thought that it was, about a year ago.

So, thanks for that. It's email time.

glad to be even slightly useful
2) Message boards : Number crunching : hadcm3n upload error (Message 50845)
Posted 18 Nov 2014 by glaesum
Post:
I'll send an email about it, but if you're talking about a different file, for ANZ, PNW, the re-start file (zip 13), or the trickle_up files, then you'll have to say so.



Thanks: it's the final completion zip for this unit:
18/11/2014 01:15:28 climateprediction.net Started upload of hadam3p_eu_lbnu_2013_1_008826339_0_13.zip
on an unattended pc which I'm leaving again before 10.00 GMT.
although I can access it remotely if really necessary.
I don't know how long it's been trying but last trickle was Nov12th, the pc isn't on very much.

3) Message boards : Number crunching : hadcm3n upload error (Message 50842)
Posted 17 Nov 2014 by glaesum
Post:
the upload server has been down since at least the middle of the weekend.
any news on repair?
4) Message boards : Number crunching : Reporting - Errors while computing - (Message 47725)
Posted 4 Dec 2013 by glaesum
Post:
Another one bites the dust...

Just after the 20th trickle went up, so that makes it at the 50% mark doesn't it?
I've got another long coupled model at 47%, will it get passed the 50% hurdle?

Let's see if things go better with the shorter models currently available.
5) Message boards : Number crunching : Reporting - Errors while computing - (Message 47587)
Posted 16 Nov 2013 by glaesum
Post:

I'll change "Only after computer has been idle for" to 0 minutes, it was on 3.00mins.
{not entirely sure what this latter setting actually means or really remember why it was on 3.00}


If you have selected "while computer is in use", the setting doesn't matter.


Thanks prof. Desty, I sensed some double speak in there...
6) Message boards : Number crunching : Reporting - Errors while computing - (Message 47576)
Posted 14 Nov 2013 by glaesum
Post:
The error list shows BOINC stopping a lot, indicative of the option: Suspend work if CPU usage is above being still set to the default of 25%.
Which is fine for other projects, but not here.
These programs DON'T like being interrupted at certain critical points.

So it's possible that you started to use the computer at that moment, the cpu load went above 25%, and BOINC, (and the model), stopped. In the case of the model, permanently.


Les, thanks for looking at things.

Current setting:

"Computing allowed"
1] while computer is in use
2] while processor usage is less than 0 percent

I'll change "Only after computer has been idle for" to 0 minutes, it was on 3.00mins.
{not entirely sure what this latter setting actually means or really remember why it was on 3.00}

also, applications are left in memory on suspend.

It's true, there does tend to be a whole lot of stuff running on the pc most times. I do need to get myself a new desktop pc which will share the workload of everything going on! :)

Pete
7) Message boards : Number crunching : Reporting - Errors while computing - (Message 47568)
Posted 14 Nov 2013 by glaesum
Post:
I've returned to the project after a bit of a gap.
A couple of models have errored out, one annoyingly close to the end - so I'll post a clip of the message log to gauge any opinions, I don't know what all the error codes mean.
I do most of the hygiene things anyway, though perhaps I should look at exempting Boinc data from MSE a/virus.
One model completed successfully and was stuck in the recent upload blockage and one of the others failed just after everything cleared. Just coincidence?
My tasks should be set to visible to see the stderr exit files <which I don't understand!>.

message log, starting with the successful model completing and reporting:

12/11/2013 07:30:19 climateprediction.net Started upload of hadcm3n_o1km_1980_40_008401621_1_4.zip
12/11/2013 07:30:21 climateprediction.net [error] Error reported by file upload server: Server is out of disk space
12/11/2013 07:30:21 climateprediction.net Temporarily failed upload of hadcm3n_o1km_1980_40_008401621_1_4.zip: transient upload error
12/11/2013 07:30:21 climateprediction.net Backing off 3 hr 30 min 42 sec on upload of hadcm3n_o1km_1980_40_008401621_1_4.zip
12/11/2013 11:01:04 climateprediction.net Started upload of hadcm3n_o1km_1980_40_008401621_1_4.zip
12/11/2013 11:04:50 climateprediction.net Finished upload of hadcm3n_o1km_1980_40_008401621_1_4.zip
12/11/2013 12:59:28 climateprediction.net task hadcm3n_7x75_1980_40_008454308_3 resumed by user
12/11/2013 13:02:59 climateprediction.net Restarting task hadcm3n_7x75_1980_40_008454308_3 using hadcm3n version 607
12/11/2013 21:39:31 climateprediction.net Sending scheduler request: To send trickle-up message.
12/11/2013 21:39:31 climateprediction.net Reporting 1 completed tasks, not requesting new tasks
12/11/2013 21:39:34 climateprediction.net Scheduler request completed
13/11/2013 01:15:22 climateprediction.net Task hadcm3n_o525_1940_40_008380310_2 exited with zero status but no 'finished' file
13/11/2013 01:15:22 climateprediction.net If this happens repeatedly you may need to reset the project.
13/11/2013 01:15:22 climateprediction.net Task hadcm3n_7x75_1980_40_008454308_3 exited with zero status but no 'finished' file
13/11/2013 01:15:22 climateprediction.net If this happens repeatedly you may need to reset the project.
13/11/2013 01:15:23 climateprediction.net Restarting task hadcm3n_o525_1940_40_008380310_2 using hadcm3n version 607
13/11/2013 01:15:24 climateprediction.net Restarting task hadcm3n_7x75_1980_40_008454308_3 using hadcm3n version 607
13/11/2013 01:16:28 climateprediction.net Task hadcm3n_ofqn_1900_40_008475522_1 exited with zero status but no 'finished' file
13/11/2013 01:16:28 climateprediction.net If this happens repeatedly you may need to reset the project.
13/11/2013 01:16:28 climateprediction.net Restarting task hadcm3n_ofqn_1900_40_008475522_1 using hadcm3n version 607
13/11/2013 01:20:33 climateprediction.net Sending scheduler request: To send trickle-up message.
13/11/2013 01:20:33 climateprediction.net Not reporting or requesting tasks
13/11/2013 01:20:37 climateprediction.net Scheduler request completed
13/11/2013 01:20:51 climateprediction.net Computation for task hadcm3n_ofqn_1900_40_008475522_1 finished
13/11/2013 01:20:51 climateprediction.net Output file hadcm3n_ofqn_1900_40_008475522_1_3.zip for task hadcm3n_ofqn_1900_40_008475522_1 absent
13/11/2013 01:20:51 climateprediction.net Output file hadcm3n_ofqn_1900_40_008475522_1_4.zip for task hadcm3n_ofqn_1900_40_008475522_1 absent
13/11/2013 09:25:47 climateprediction.net Sending scheduler request: To send trickle-up message.
13/11/2013 09:25:47 climateprediction.net Reporting 1 completed tasks, not requesting new tasks
13/11/2013 09:25:50 climateprediction.net Scheduler request completed


8) Message boards : Number crunching : NEW BOINC VERSION (Message 47566)
Posted 14 Nov 2013 by glaesum
Post:
If you're using a 6.x or earlier version please heed the warning about not being able to go back (unless you back up your client_state files).


What does 'not able to go back mean'? thanks
9) Message boards : Number crunching : New Tasks not being snapped up (Message 47234)
Posted 4 Oct 2013 by glaesum
Post:
Isn't it surprising that the available tasks queue is not reducing more quickly? I imagine that a lot volunteers either suspended CPDN or set 'no new tasks' when we were having lots of problems and haven't bothered to reset their account. ~ David


I guess I was one of those, I didn't have the patience to monitor what was going on from day to day. Great that everything is settling down again - there's work to do!
Anyway, I now have three models in my task box, though I do rest one or two of them for other projects for the occasional challenge.
My old desktop was getting a bit slow compared with most users and I just defragged the hard drive to death but ... there's a new powerful desktop on my shopping list for the next month and I can almost hear the revving noises adready! /pg
10) Message boards : Number crunching : Upload Failure (Message 44472)
Posted 27 Jun 2012 by glaesum
Post:
ok :(
I've fetched work for other active projects for a day or two and shut network activity while waiting for news. /pg
11) Message boards : Number crunching : Upload Failure (Message 44470)
Posted 27 Jun 2012 by glaesum
Post:
same here on my old machine with a European model. (just noticed boinc not running on the other machine so no news from that machine until it runs for a bit).
12) Message boards : Number crunching : Upload Failure (Message 44392)
Posted 13 Jun 2012 by glaesum
Post:
hi Greg, all's well. In the small hours I was just about to shut down network activity overnight when I found everything had finally uploaded ok. I think it was several people reporting success in the previous hours that got me slightly concerned while it looked like the servers all had green flags.

let's hope the long period of recurrent server problems is eventually overcome.
now I just need to clear the cache of SIMAPs before returning to my HADAM eu models! :)
13) Message boards : Number crunching : Upload Failure (Message 44382)
Posted 12 Jun 2012 by glaesum
Post:
Mine are still struggling. I have had two pairs of EU _4.zip and _5.zips waiting for some days. I have a couple of PNW models reporting fine on a laptop on the same network/router. Meanwhile I've suspended the cpdn EU wus to stop adding to the queue and running some SIMAP resends for June until things clear up. (I'm keeping one of the laptop PNW models going to confirm it still sends it's zips though they'll be infrequent.)

Using Greg's test [msg: June 6th 8:02utc]
http://climateapps2.oerc.ox.ac.uk/cpdnboinc/forum_thread.php?id=7300&nowrap=true#44309

My browser fails the first test and won't connect to the upload server page but passes the second 'tracert' test with no '*'s and only 16 to 18 cmd lines.
Not sure what that all means... ...not quite tecchie enough to understand. :)

/pg

from Greg: "Open another tab in your browser and go to this address: http://cpdn-upload2.oerc.ox.ac.uk.

The response should be a page that says
Climate Prediction.net Upload Server

This server is part of the Climate Prediction.net project. Please visit climateprediction.net to participate.
If you don't get that, there is a connection problem - perhaps a blacklisting. To check, open a command prompt (Win+R cmd.exe) and run
tracert cpdn-upload2.oerc.ox.ac.uk

The trace should complete in less than a minute, listing fewer than 30 numbered lines and with not too many '*'s in the output."

14) Message boards : Number crunching : had3pam_eu models not uploading (Message 43810)
Posted 16 Feb 2012 by glaesum
Post:
I suspended network activity when I spotted the backlog.

anyway all now cleared and up to date - it was all rather slow and stop/start, let's hope all the trickles and the one finished wu reporting all got assimilated properly.
15) Message boards : Number crunching : screensaver (Message 43646)
Posted 6 Jan 2012 by glaesum
Post:
Thanks for this old tip: it still works so is perhaps worth highlighting.

I've had some regional hadam3p models recently after a rest from cpdn and had also forgotten how to get the savepoint countdown to display (there used to be a hidden keystroke - a "Z" I think - bit that doesn't seem to do it anymore).
16) Message boards : Number crunching : Server can't open log file (Message 42644)
Posted 19 Jul 2011 by glaesum
Post:
"The trickles aren't showing because of the huge backlog of data on the upload servers that needs to be processed, at the same time as tens of thousands of computers want to download new work."

right now I can see trickles for Jul 10 & Jul 11 - so that's some activity, though not sure if any headway is being made into the backlog. how's things with others? /p

17) Message boards : Number crunching : Server can't open log file (Message 42622)
Posted 15 Jul 2011 by glaesum
Post:
well, my first log message for days "Scheduler request completed" without the dreaded "Scheduler request failed: HTTP internal server error" was at 15:55 BST and no sign of any recent trickle files in the data folder.

They don't seem to be showing on the database yet - I expect that's the huge backlog to process. At least there's some sign of life - well done for getting it up for the weekend!

I'll let another model run now as well as the long coupled models. /p
18) Message boards : Cafe CPDN : Milestones Thread (Message 38978)
Posted 23 Feb 2010 by glaesum
Post:
{mo.v said} You have an excellent record of completed models! Your computer\'s very reliable.

thanks Mo, I do try to nurse my models carefully - I did do the occasional back-up (not often enough I know) of the old long coupled models and only aborted a couple when I was getting that weird completion deadline in the early 1900s. One slab model which finished at my end but seemed to have been orphaned by missing the final trickle in the upload servers - it now lives on as a ghost.

I\'ve just added a new host which won\'t be turned on that much but it\'s powerful enough to progress the short HadAM models - I got one to start it off but it\'s a pity there are none left in stock on the server right now. /pg
19) Message boards : Cafe CPDN : Milestones Thread (Message 38855)
Posted 3 Feb 2010 by glaesum
Post:
not so much life in the milestones thread, pity...

well, I finally passed 200,000 a couple of days ago plodding away with my old pc (+over 80k on BBC-CE); some hope on the horizon for a more powerful machine soon. /pg
20) Message boards : Number crunching : Long shutdowns and keeping HADSM model alive (Message 38326)
Posted 19 Nov 2009 by glaesum
Post:
your advice ended up pretty well.

I\'ve got to 92% before my long shutdown, so two phases have reported. If I hadn\'t gone away for a weekend I might have actually finished the wu.

anyway, thanks for keeping me crunching confidently. hope everyone has happy break at the end of the year. /pg


Next 20

©2024 climateprediction.net