Having been without Broadband for a couple of weeks, my hadsm34.13 application completed. Since regaining a broadband connection the transfer will not restart. Only 1.11% (0.02Mb) of the first zip file is transferred before it hangs and later fails. Any ideas?

---

What is the message you get?

An example of text that appears under messages:-

21/12/2005 14:10:29|climateprediction.net|Started upload of 40xy_200209730_0_4.zip
21/12/2005 14:15:33|climateprediction.net|Temporarily failed upload of 40xy_200209730_0_5.zip: -103
21/12/2005 14:15:33|climateprediction.net|Backing off 28 minutes and 14 seconds on upload of file 40xy_200209730_0_5.zip
21/12/2005 14:16:31|climateprediction.net|Temporarily failed upload of 40xy_200209730_0_4.zip: -103
21/12/2005 14:16:31|climateprediction.net|Backing off 1 hours, 50 minutes, and 38 seconds on upload of file 40xy_200209730_0_4.zip

---

I have since investigated this further and found a number of message threads discussing the problem. It seems it relates to the failure to connect to the receiving server due to it being \"down\". Previously, the Berne server has experienced a power outage causing participants to experience similar errors.

An important piece of information is the abandonment of the upload following two weeks of failed uploads. The message link below includes a five point resolution to this problem from Thyme Lawn, which involves resetting the timer.

http://www.climateprediction.net/board/viewtopic.php?p=29490&highlight=

From what I understand, this resets the timer to zero, allowing a further two weeks for the relavant server to be recovered.

I wait in hope....

---

Yes, according to the Server Stats page, there\'s a problem at Oxford. One says \'disabled\', which is a manual thing, so hopefully they will all be back before too long.

And now the Stats page says that they\'re all running. :)

I have upgraded to version 5.2.13 and the transfer figures all say 100%, but status still indicates \'retry in xx.xx.xx\'.

The following messages are being displayed.

22/12/2005 10:03:33|climateprediction.net|Resuming computation for result 3fex_100181553_1 using hadsm3 version 413
22/12/2005 10:03:33|climateprediction.net|Deferring computation for result sulphur_ae87_000484999_0
22/12/2005 10:17:29|climateprediction.net|Fetching master file
22/12/2005 10:18:49|climateprediction.net|Master file fetch failed
22/12/2005 10:18:49|climateprediction.net|Too many backoffs - fetching master file
22/12/2005 10:28:54|climateprediction.net|Started upload of 40xy_200209730_0_3.zip
22/12/2005 10:34:06||Attempting to communicate with [uploader1.atm.ox.ac.uk] timed out
22/12/2005 10:34:06|climateprediction.net|Temporarily failed upload of 40xy_200209730_0_3.zip: timeout
22/12/2005 10:34:06|climateprediction.net|Backing off 2 hours, 3 minutes, and 50 seconds on upload of file 40xy_200209730_0_3.zip

---

For other reasons I needed to perform XP restore to 4/12/05, after which I received BOINC message errors. After undoing the restore, my BOINC work has disappeared (100% was waiting to upload and 60% was in progress). This just seems to be getting worse with the following messages now being displayed, the last two highlighted in red.

23/12/2005 06:45:34|climateprediction.net|Sending scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi
23/12/2005 06:45:34|climateprediction.net|Reason: Requested by user
23/12/2005 06:45:34|climateprediction.net|Requesting 8640 seconds of new work
23/12/2005 06:50:42||Attempting to communicate with [climateapps2.oucs.ox.ac.uk] timed out
23/12/2005 06:50:44|climateprediction.net|Scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi failed with a return value of -182
23/12/2005 06:50:44|climateprediction.net|No schedulers responded

---

Hi, Andrew, I doubt I can be of help because you\'ve been at this for a long time but please allow me to throw my two cents on the pile.

I don\'t understand how M$ handles its backup/restore. I\'m convinced it takes a bit of time and that user activity isn\'t paused. (I hope there will be quick correction if that isn\'t so.)

On the other hand, I do understand that CPDN backups *MUST* be done with boinc suspended because there are so many files that things will otherwise be out of sync.

For that reason, I have Windoze backups quashed for the partition holding CPDN. I do CPDN backups manually, and I think that is common practice among CPDN participants.

HTH.

---

"We have met the enemy and he is us." -- Pogo
Greetings from coastal Washington state, the scenic US Pacific Northwest.

Thanks for the advise on backups. It is something that I will certainly do in the future. In the meantime, however, I seem to have no data and my application will not communicate with the climate server.

---

Still with error -182? (From a list captured from boinc site: #define ERR_TIMEOUT -182)

Could a Proxy or Firewall setting have been changed in the fall-back? Or in some routine maintenance? It looks suspiciously like the Scheduler Request isn\'t getting out of your machine, so boinc eventually sets the timeout error.

The Server Stats page indicates everything is currently running.

---

"We have met the enemy and he is us." -- Pogo
Greetings from coastal Washington state, the scenic US Pacific Northwest.

I have not had any work listed for the past two days and assume I have lost 1½ runs. Unfortunately I am not recieving a new piece of work due to the \'-182\' error. Disabling the firewall has no effect.
I can only guess that I need to uninstall and re-install BOINC.

---

My BOINC project is now active and has started a new application. I wonder if a change to the server following the holiday period corrected it. Unfortunately the last fully completed application did not automatically upload. Is there any way I can manually upload it from my hard drive or backup? Alternatively, can I re-apply it for automatic upload?

---

Since the 9th January, my BOINC conection has been running correctly with the new application. Imediately prior to that I was getting error 500. It must have been a server problem as I didn\'t change anything else at my end.

Unfortunately my previous 1½ runs seem to have been lost.

---

Since the 9th January, my BOINC conection has been running correctly with the new application. Imediately prior to that I was getting error 500. It must have been a server problem as I didn\'t change anything else at my end.

So, am I to understand that upgrade/reinstallation of BOINC will fix the \'attempting to communicate with [uploader1.ox.ac.uk] timed out\' problem?

As you\'ve probably guessed, I\'m experiencing exactly the same issues as Andrew. Ordinary trickles seem to get through OK, it\'s just the phase 1 sulphur cycle upload that it\'s baulking at. I\'m pretty sure the problem is at my end, but I\'ve tried tweaking firewall and antivirus software settings, to no effect as yet. Any clues appreciated.

---

I would suggest that the problem relates to the BOINC installation or setup on your PC, as BOINC re-installation was the only change I made to enable successful operation. I had previously tried disabling Antivirus and Firewall\'s without correcting the error.

Incidentally, I never did recover the work completed to that point.

---