I wasn\'t aware of any delay, but the servers are being moved this week so things might not be working properly at the moment.

---

I'm a volunteer and my views are my own.
News and Announcements and FAQ

ok, the second wu has much more than 10 trickles now but only ten are shown at the website.

Maybe a error only the first ten wu\'s are shown or what it ever may be.
Is at trickle 10 a special checkpoint?
Are the data send from my pc received by cpdn servers or not?, what is going on?

---

The same for me, only the first 10 trickles are shown (and got credit).

:-)

Hi Tomcat

I\'ve just checked on the trickles for my own two models and they all show including for 21 April.

You have 4 models running simultaneously.

http://climateapps2.oucs.ox.ac.uk/cpdnboinc/results.php?hostid=613470

#1, Result ID 6451219 seems to show the correct trickles (about 26000 timesteps per trickle or model year, so that looks right). But the graph shows two years repeated, 1925 and 1929. This gives the impression that the model has completed 10 full years, not 8. It trickled on 21 April, so I think this model is OK.

#2, Result ID 6450435 shows 10 years completed on the graph and 10 trickles with the correct timestep for trickle #10. But this model hasn\'t trickled since 15 April. This model hadcm3inct_cnj2_1920_160_05865535_2 is a problem. The messages you posted above show it has uploaded its second 10-year upload. This means it\'s completed at least 20 years to 1941, but only 10 trickles show.

What a fast computer!

#3, Result ID 6447157 also shows 10 years completed on the graph and 10 trickles with the correct timestep for trickle #10. It hasn\'t trickled since 19 April. Has it been running or not?

#4, Result ID 6440177 shows 4 years completed on the graph (some of the years are repeated until the model has done 10 years) and 4 trickles with the correct timestep for trickle #4. The last trickle is 21 April, so I think this model is OK.

You need to tell us whether only #2 Result 6450435 is a problem, or #3 also.

At the end of each 10 model years, there\'s a bigger upload in addition to the normal trickle. You can see it in the Transfers tab of boinc manager and can watch it upload. When the upload has finished, the Transfers window is empty again.

I will ask one of the programmers to look at this thread, Tomcat, but that will probably not be possible until Monday or later because our cpdn php forum isn\'t working and they may consider that a greater priority.

Vielen Dank für Ihre Engelsgeduld!

---

Cpdn news

Hi S@NL EJG

This is the problem model on your dual-core? (I don\'t think you have any current models on the other computer.)

http://climateapps2.oucs.ox.ac.uk/cpdnboinc/result.php?resultid=6449610

This model still doesn\'t show any trickles since 17 April. It\'s running on Windows Vista whereas Tomcat\'s problem model is on a Mac.

Could you please post any messages relating to trickles after 17 April.

---

Cpdn news

Indeed, I\'ve only one model running at this time (Core2Duo, Windows Vista).

Below are the messages of the last day, just after a restart of Boinc. Since the start of the model (also after April 17) everything looks fine at my side.


21/04/2007 10:50:24||Host location: home
21/04/2007 10:50:24||General prefs: using separate prefs for home
21/04/2007 10:50:24|climateprediction.net|Restarting task hadcm3inct_cndc_1920_160_05865329_1 using hadcm3i version 540
21/04/2007 11:32:40|climateprediction.net|Sending scheduler request: To send trickle-up message
21/04/2007 11:32:40|climateprediction.net|(not requesting new work or reporting completed tasks)
21/04/2007 11:32:45|climateprediction.net|Scheduler RPC succeeded [server version 509]
21/04/2007 21:17:07|climateprediction.net|Sending scheduler request: To send trickle-up message
21/04/2007 21:17:07|climateprediction.net|(not requesting new work or reporting completed tasks)
21/04/2007 21:17:12|climateprediction.net|[file_xfer] Started upload of file hadcm3inct_cndc_1920_160_05865329_1_2.zip
21/04/2007 21:17:12|climateprediction.net|Scheduler RPC succeeded [server version 509]
21/04/2007 21:18:23|climateprediction.net|[file_xfer] Finished upload of file hadcm3inct_cndc_1920_160_05865329_1_2.zip
21/04/2007 21:18:23|climateprediction.net|[file_xfer] Throughput 75807 bytes/sec
22/04/2007 07:00:46|climateprediction.net|Sending scheduler request: To send trickle-up message
22/04/2007 07:00:46|climateprediction.net|(not requesting new work or reporting completed tasks)
22/04/2007 07:01:52|climateprediction.net|Scheduler request failed: HTTP internal server error
22/04/2007 07:01:52|climateprediction.net|Deferring communication for 1 min 0 sec
22/04/2007 07:01:52|climateprediction.net|Reason: scheduler request failed
22/04/2007 07:02:53|climateprediction.net|Sending scheduler request: To send trickle-up message
22/04/2007 07:02:53|climateprediction.net|(not requesting new work or reporting completed tasks)
22/04/2007 07:02:58|climateprediction.net|Scheduler RPC succeeded [server version 509]

Thanks for the messages.

EJG\'s model has at least 10 years missing on both the trickle list and the graph, because the zip file whose number ends _2 is the second decadal upload sent in 1941, but only the 10 trickles from 1921-1931 show on his web page.

Both Tomcat and EJG received this message
Scheduler request failed: HTTP internal server error
before their server request succeeded. Could this be significant or not?

---

Cpdn news

This was the first server/connection error I had in a long time. When sending the trickles during the last days I never had any connection problems (as far as I know) and despite that they did not show up on the web page.
As far as I can see in Boinc all trickles succeeded (except this one time server error, but the next try one minute later was fine), as did the file uploads.

:-)

EJG and Tomcat

Did you update your version of boinc at the same time that the trickles stopped for these models?

---

Cpdn news

I did not upgrade Boinc during this model. It was a fresh install of Boinc on a new PC.

:-)

thanks for that information. I think I\'ve found the culprit -- the model id info (BOINC appid) wasn\'t put in the cpdn_trickle program. I\'ve now bumped it up to 20 slots and recompiled cpdn_trickle so that should last us for years (version 5.40 hadcm3i is #6). I should have originally just used a std::vector I guess for the data struct, oh well. and all this 24 hours after my great \"going away party\" last night! Smile

Tomcat\'s and EJG\'s trickles now appear up-to-date, but their graphs still only show results to 1931. Maybe the graphs update at a different time and we\'ll have to be patient.

---

Cpdn news

All trickles do show now, thanks for the effort (even during the weekend, I feel somewhat guilty :-) ) . We will keep an eye on the graph.

One thing, I noticed a couple of error messages at the bottom of the result page:

Warning: fopen(../cache/e0/result.php_field%3D%26resultid%3D6449610%26phase%3D) [function.fopen]: failed to open stream: No such file or directory in /var/www/boinc/projects/cpdnboinc/html/inc/cache.inc on line 205

Warning: fwrite(): supplied argument is not a valid stream resource in /var/www/boinc/projects/cpdnboinc/html/inc/cache.inc on line 208

Warning: fclose(): supplied argument is not a valid stream resource in /var/www/boinc/projects/cpdnboinc/html/inc/cache.inc on line 209

ok thank you mo.v and Carl, I\'m looking forward and hoping all will be good next week :)
I was just writing the answer for mo.v as Carl posted.

...
#1, Result ID 6451219... so I think this model is OK.

it has not reached the ten trickles yet, but will do soon.

#2, Result ID 6450435 shows 10 years completed on the graph and 10 trickles with the correct timestep for trickle #10. But this model hasn\'t trickled since 15 April. This model hadcm3inct_cnj2_1920_160_05865535_2 is a problem. The messages you posted above show it has uploaded its second 10-year upload. This means it\'s completed at least 20 years to 1941, but only 10 trickles show.

What a fast computer!

MacPro 2.66 with \"only\" 2 DualCore ;) Now, Apple has one with 2 Quadcores -3GHZ! - this would be fast :)

#3, Result ID 6447157 also shows 10 years completed on the graph and 10 trickles with the correct timestep for trickle #10. It hasn\'t trickled since 19 April. Has it been running or not?

...
You need to tell us whether only #2 Result 6450435 is a problem, or #3 also.
...
Vielen Dank für Ihre Engelsgeduld!

the problem are #2 and #3!
- but I hope Carl has been solved it
so thank you very much to you and Carl!

---

I see the same error messages on all the cpdn members\' results web pages, so solving one problem may have created another. I\'ll tell the programmers about this. The really important thing is to protect the data for the models and make sure all the uploads get to Oxford.

---

Cpdn news