Several of the HadAM3P tasks that I've completed recently all show at the top of the task details page, "No data for result.."

Here is one example:
http://climateapps2.oucs.ox.ac.uk/cpdnboinc/result.php?resultid=8799790

Trickles show but there are no graphs. I'm sure my result zip files were uploaded successfully. What does this mean? Did the results get lost?

Several of the HadAM3P tasks that I've completed recently all show at the top of the task details page, "No data for result.."

Here is one example:
http://climateapps2.oucs.ox.ac.uk/cpdnboinc/result.php?resultid=8799790

Trickles show but there are no graphs. I'm sure my result zip files were uploaded successfully. What does this mean? Did the results get lost?

I think it's comment text in the PHP that's not tagged properly and is then promoted to the top of the visible page. It's been passed on to the project but nothing has happened thus far.

In any event it's benign.

OK, I'll ignore the error message. But the fact that there are no graphs for several results.... Is this just an artifact of the upload server difficulties, slow processing, or something else?

OK, I'll ignore the error message. But the fact that there are no graphs for several results.... Is this just an artifact of the upload server difficulties, slow processing, or something else?

We've not had an explanation for that though, again, "it's been noticed": if the model completes then the graphs seem to be there (for HADSM3/MH at least), but the intermediate ones are missing.

I'll bump these two problems and see what happens: the first is easily neglected but also easily fixed, the second may be because something's been turned off until the new server is functioning properly - which I imagine is very much the priority at the moment.

OK, I'll ignore the error message. But the fact that there are no graphs for several results.... Is this just an artifact of the upload server difficulties, slow processing, or something else?

I see your computer is hidden, but you are using LINUX.

If you look at the list of tasks, those for HADAM3P are shown as "success" with 72,000 time steps instead of 72,096, and with a credit granted of 1980.00 instead of 1,982.64. They have failed to do the post processing time step that follows the 72,096 time step.

This is the same problem that was happening with Mac OSX using version 6.07, which has now been upgraded to 6.08 and should now be corrected (as from yesterday).

Keith

The data used for the graphs is in the last trickle.

"No data for result.." is because not all of the trickles have been received.

The data used for the graphs is in the last trickle.

"No data for result.." is because not all of the trickles have been received.

Having looked at other LINUX crunchers in the top computers, I found no others with the same problem, that all mac users were having.

I had monitored the last two "faulty" results on my Mac.
They appeared on the graphics view to complete right up to 72,096.
Then started another T/step of 96 as the post processing step, which failed to count down, and the task was marked as 100% and completed, with next task being started.
Even though the crunching appeared to go to 72,096, the list of tasks only showed 72,000.

It seems strange that this one linux consistently had the same problem as the Macs.

Keith

Based on the applications page, the Linux HadAM3P application has not be updated, but I see the MacOSX one has. Does anyone know what they changed in the new MacOS version of the application? What libraries does the post-processing use? Perhaps the application is using an older library that has not been regression tested (or missing) on the newer unix-based OS. I'm running the latest Fedora 11 linux.

Unfortunately, I don't have access to a linux box at work right now. But could someone run the linker info like this:
ldd hadam3p_se_6.06_i686-pc-linux-gnu.so

When I get home tonight (EST), I'll check to see what 32-bit libraries I have installed. Perhaps I'm missing an older 32-bit library that CPDN needs.

Does anyone know what they changed in the new MacOS version of the application?

Some of the linking has been changed: it's a re-build rather than a code change.

OK, so I did some research as I said earlier. Here is the shared library readout:
$ ldd hadam3p_se_6.06_i686-pc-linux-gnu.so
linux-gate.so.1 => (0x001f0000)
libz.so.1 => /lib/libz.so.1 (0x00e23000)
libstdc++.so.6 => /usr/lib/libstdc++.so.6 (0x001f1000)
libnsl.so.1 => /lib/libnsl.so.1 (0x00dd8000)
libm.so.6 => /lib/libm.so.6 (0x00857000)
libgcc_s.so.1 => /lib/libgcc_s.so.1 (0x0063b000)
libc.so.6 => /lib/libc.so.6 (0x002de000)
/lib/ld-linux.so.2 (0x00947000)

There is also an "/lib/i686/nosegneg/libm.so.6" on my system.

The error on the tasks are all the same:
Unable to load library hadam3p_se_6.06_i686-pc-linux-gnu.so
dlopen error: 138932776

Perhaps the finish script is not returning to the main project directory and is staying in the ./slots/X directory?

Can anyone explain why HadAM3P models fail to finish on my computer?

Here is a perfect example workunit. Computer #124 (wow an original cruncher) and my computer #979394 did the same model. His finished at timestep 72,096, but mine finished at 72,000. And both the temperature and precipitation graphs are missing for mine. He's running an Intel Q9450 w/ Linux 2.6.25.20-0.4 and I'm running an Intel i7 920 w/ Linux 2.6.29.5-191. So, why do my tasks fail to finish?

Computer #859124 (on that same workunit) is a Mac, so I'm interested to see if that computer finishes the task correctly.

Can anyone explain why HadAM3P models fail to finish on my computer?
.........

DJStarfox

I think you will find that that task for a Mac is running version 6.07 and will finish at 72,000.
However, I have started 2 tasks on the updated version of 6.08 and should complete at 72,096.
Watch this space on Tuesday 28th July!!! Should have finished at 72,096 by then (or the day before).
You might like to see thread "Good news for Mac users. HadAM3P Latest News???" with more information.

My Mac ran normally on version 6.06, had the 72,000 problem after the 6.07 update, and is now running 6.08 for the first time.

It does seem strange that other LINUX systems do not have a 72,000 problem while yours does.
I did pass on this detail, and it was then reported to Tolu 4 days ago.

Keith

Hello!

This WU ID 6351662 has 3 finished tasks, processed on 3 different OS (Darwin, Linux and XP - my task). All 3 are marked as "Success", but all 3 have 72000 problem.

By accident I saw the end of process.
What I know exactly:
1. the last 72096 step was reached;
2. post processing has at least started;
3. 3 zip files were sent.
That is all - I decided the task is OK, and I did not fix additional details.

Now I have 3 ways:
1. just forget this task, because the situation is hopeless OR... in reality the task is finished successful, no matter, what is shown on the task page due missing last trickle;
2. restore the task from backup and try to finish it, because this task crushed, but the situation is not hopeless;
3. restore the task from backup and try to finish it, watch the end of process careful, write a short report here, what I saw, extract system messages and post them here, do something more.

Dear project team!
Which way I must go or which way is the best?

---

First off, please quote the Task Id when reporting problems, not the Work unit ID. It reduces the time needed to search through the computers to find yours.
This is the task in question.

As you say, it didn't upload the final trickle.
I don't think that there's much point in re-running from a backup, which will most likely end the say way, but please keep the backup in case we come up with something.
I'll pass this on to the project people for consideration.

In the meantime, you can go to your climateprediction.net preferences on your account page, and select a different model type to download.

And keep an eye on this thread. You can subscribe to it and get an email when a new post is made here. IF you have email notification turned on. :)

---

Backups: Here

First off, please quote the Task Id when reporting problems, not the Work unit ID. It reduces the time needed to search through the computers to find yours.

I understand - this time I quoted WU purposely, because my goal was to show the global problem, not my own. 3 parallel tasks on 3 different OS crushed exactly at the same point - looking from this position minimum quorum 3, required for this WU, is reached, just kidding. :-)
And, of course... Thank You very much for Your quick reply (as ever) and recommendations - CPDN forum is the best!

---

I was downloading something completely unrelated the other night, when I realized my system was lacking the ncompress tool. This is creates (and expands) .Z compressed files on unix-like systems. I remember this utility being on all my other Linux systems (before this last upgrade). Could that be the problem?

Update: From my earlier post about "ncompress" library...
Adding this package did not help. Two more HadAM3P tasks ended before the last trickle. If the re-linking of the HadAM3P binary fixed the issue with the MacOS version, could this also be done with the Linux version?

If not, then could the following questions be answered? What linker and parameters do you guys use to build the executable? What versions of shared libraries did you use to link against?

If it is an dlopen() error from the program source code, would ldd even report it? I think it would not as it cannot know the first argument to the function.

Does the user who runs BOINC have read/execute on all library files that come with BOINC and CPDN?

As "only" the data trickle is missing, my guess would be some zlib thing to look for.


If this is a common problem, it would even be possible that everyone who already has crunched old models has that missing library from earlier downloads, but those who started with HadAM3p never received it.


edit : There are quite a few results with this problem (Google finds 246 entries) :

CPDN Monitor - Quit request from BOINC...
Unable to load library hadam3p_se_6.07_i686-apple-darwin.dylib
dlopen error: 3152085
17:09:08 (180): called boinc_finish

The interesting part is, that those I have checked have all been on Darwins (edit again ... now I found a few dlopen errors on Linuxes too)

72,000 problem is present on Windows machines too. Next day, next task - 9105039 - with this problem.
I don't know, is this useful or not, I captured all messages from trickle after Timestep 69,120 to end oft task.
They are (PC time was ~UTC+02:55)...
29/07/2009 09:00:37 climateprediction.net Sending scheduler request: To send trickle-up message.
29/07/2009 09:00:37 climateprediction.net Not reporting or requesting tasks
29/07/2009 09:00:42 climateprediction.net Scheduler request completed: got 0 new tasks
29/07/2009 14:13:16 climateprediction.net Sending scheduler request: To send trickle-up message.
29/07/2009 14:13:16 climateprediction.net Not reporting or requesting tasks
29/07/2009 14:13:21 climateprediction.net Scheduler request completed: got 0 new tasks
29/07/2009 14:23:22 climateprediction.net Sending scheduler request: To send trickle-up message.
29/07/2009 14:23:22 climateprediction.net Not reporting or requesting tasks
29/07/2009 14:23:27 climateprediction.net Scheduler request completed: got 0 new tasks
29/07/2009 14:23:30 climateprediction.net Computation for task hadam3p_md2t_1964_2_006118783_3 finished
29/07/2009 14:23:32 climateprediction.net Started upload of hadam3p_md2t_1964_2_006118783_3_1.zip
29/07/2009 14:23:32 climateprediction.net Started upload of hadam3p_md2t_1964_2_006118783_3_2.zip
29/07/2009 14:26:15 climateprediction.net Finished upload of hadam3p_md2t_1964_2_006118783_3_2.zip
29/07/2009 14:26:15 climateprediction.net Started upload of hadam3p_md2t_1964_2_006118783_3_3.zip
29/07/2009 14:26:19 climateprediction.net Finished upload of hadam3p_md2t_1964_2_006118783_3_3.zip
29/07/2009 14:26:46 climateprediction.net Finished upload of hadam3p_md2t_1964_2_006118783_3_1.zip
Looking to the messages, the final trickle was sent (or an attempt to send it was done), but (possible) it was refused by server...

---

It is a dlopen() error, but what file is it trying to open? That is the $1M question.

Here are the permissions of my /boinc/projects/climateprediction.net/

[size=8]total 214220
drwxrwxr-x. 5 boinc   boinc     4096 2009-07-29 15:48 .
drwxrwxr-x. 7 boinc   boinc     4096 2009-06-05 13:45 ..
-rw-rw-r--. 1 boinc   boinc  1573376 2008-08-29 07:57 globe.rgb
-rwxr-xr-x. 1 boinc   boinc  2551172 2009-06-19 13:59 hadam3_6.01_i686-pc-linux-gnu
-rw-r--r--. 1 boinc   boinc 32002981 2009-06-19 14:19 hadam3_data_6.01_i686-pc-linux-gnu.zip
-rwxr-xr-x. 1 boinc   boinc 16405162 2009-06-19 14:09 hadam3_graphics_6.01_i686-pc-linux-gnu
drwxrwx--x. 2 boinc   boinc     4096 2009-07-11 07:08 hadam3h_g_089se_005a_005a_2
drwxrwx--x. 2 boinc   boinc     4096 2009-06-20 10:23 hadam3h_g_90s08_005a_005a_2
-rwxr-xr-x. 1 boinc   boinc  1391705 2009-07-10 22:00 hadam3p_6.06_i686-pc-linux-gnu
-rw-r--r--. 1 boinc   boinc 83311850 2009-07-10 23:34 hadam3p_data_6.06_i686-pc-linux-gnu.zip
-rwxr-xr-x. 1 boinc   boinc 16399234 2009-07-10 22:09 hadam3p_graphics_6.06_i686-pc-linux-gnu
-rwxrwxr-x. 1 boinc   boinc  1950297 2009-03-06 08:34 hadam3p_se_6.06_i686-pc-linux-gnu.so
-rw-r--r--. 1 boinc   boinc  1837353 2009-07-10 22:01 hadam3p_se_6.06_i686-pc-linux-gnu.zip
-rwxrwxr-x. 1 boinc   boinc  5504520 2009-03-06 08:53 hadam3p_um_6.06_i686-pc-linux-gnu
-rw-r--r--. 1 boinc   boinc  2246011 2009-07-10 22:11 hadam3p_um_6.06_i686-pc-linux-gnu.zip
-rwxrwxr-x. 1 boinc   boinc  5102878 2008-08-22 06:45 hadam3_um_6.01_i686-pc-linux-gnu
-rw-r--r--. 1 boinc   boinc  3069209 2009-06-19 14:11 hadam3_um_6.01_i686-pc-linux-gnu.zip
-rw-r--r--. 1 boinc   boinc     3782 2009-06-13 10:36 hadcm3_40.png
-rw-r--r--. 1 boinc   boinc    13708 2009-06-13 10:36 hadcm3_banner_290.png
-rw-r--r--. 1 boinc   boinc    49890 2009-06-13 10:36 hadcm3_ss_290_1.png
-rw-r--r--. 1 boinc   boinc    48077 2009-06-13 10:36 hadcm3_ss_290_2.png
-rw-r--r--. 1 boinc   boinc    41929 2009-06-13 10:36 hadcm3_ss_290_3.png
-rwxr-xr-x. 1 boinc   boinc  3098846 2008-07-26 13:51 hadcm3trans_se_6.04_i686-pc-linux-gnu
-rwxrwxr-x. 1 boinc   boinc  5073074 2008-10-21 08:46 hadcm3trans_um_6.04_i686-pc-linux-gnu
-rwxr-xr-x. 1 boinc   boinc  2573442 2009-07-03 15:58 hadsm3mh_6.03_i686-pc-linux-gnu
-rw-r--r--. 1 boinc   boinc    80223 2009-07-03 16:01 hadsm3mh_data_6.03_i686-pc-linux-gnu.zip
-rwxr-xr-x. 1 boinc   boinc 16403992 2009-07-03 16:06 hadsm3mh_graphics_6.03_i686-pc-linux-gnu
-rwxrwxr-x. 1 boinc   boinc  2955099 2008-08-29 07:27 hadsm3mh_se_6.03_i686-pc-linux-gnu
-rw-r--r--. 1 boinc   boinc  2153701 2009-07-03 15:59 hadsm3mh_se_6.03_i686-pc-linux-gnu.zip
-rwxrwxr-x. 1 boinc   boinc 10050414 2008-08-29 07:26 hadsm3mh_um_6.03_i686-pc-linux-gnu
-rw-r--r--. 1 boinc   boinc  3361911 2009-07-03 16:01 hadsm3mh_um_6.03_i686-pc-linux-gnu.zip
-rw-r--r--. 1 boinc   boinc       76 2009-07-29 07:17 slideshow_hadcm3_00
-rw-r--r--. 1 boinc   boinc       74 2009-07-29 07:17 slideshow_hadcm3_01
-rw-r--r--. 1 boinc   boinc       74 2009-07-29 07:17 slideshow_hadcm3_02
-rw-r--r--. 1 boinc   boinc       74 2009-07-29 07:17 slideshow_hadcm3_03
-rw-r--r--. 1 boinc   boinc       76 2009-07-29 07:17 slideshow_hadcm3i_00
-rw-r--r--. 1 boinc   boinc       74 2009-07-29 07:17 slideshow_hadcm3i_01
-rw-r--r--. 1 boinc   boinc       74 2009-07-29 07:17 slideshow_hadcm3i_02
-rw-r--r--. 1 boinc   boinc       74 2009-07-29 07:17 slideshow_hadcm3i_03
-rw-r--r--. 1 boinc   boinc       68 2009-07-29 07:17 stat_icon
drwxrwxr-x. 2 boinc   boinc     4096 2009-07-03 16:14 txf
[/size]