it looks like project servers for pnw may be temporarily down :(

7/21/2015 8:30:24 PM | climateprediction.net | Temporarily failed upload of hadam3p_pnw_pox0_2013_1_009979930_1_6.zip: transient HTTP error
7/21/2015 8:30:24 PM | climateprediction.net | Backing off 00:02:44 on upload of hadam3p_pnw_pox0_2013_1_009979930_1_6.zip
7/21/2015 8:30:26 PM | | Project communication failed: attempting access to reference site
7/21/2015 8:30:27 PM | | Internet access OK - project servers may be temporarily down.
7/21/2015 8:30:40 PM | climateprediction.net | Temporarily failed upload of hadam3p_pnw_prai_2013_1_009982837_1_9.zip: transient HTTP error
7/21/2015 8:30:40 PM | climateprediction.net | Backing off 00:02:16 on upload of hadam3p_pnw_prai_2013_1_009982837_1_9.zip
7/21/2015 8:30:41 PM | | Project communication failed: attempting access to reference site
7/21/2015 8:30:42 PM | | Internet access OK - project servers may be temporarily down.

same over here:

22.07.2015 05:36:38 | climateprediction.net | [fxd] starting upload, upload_offset -1
22.07.2015 05:36:38 | climateprediction.net | Started upload of hadam3p_pnw_ppwa_2013_1_009981129_0_8.zip
22.07.2015 05:36:38 | climateprediction.net | [file_xfer] URL: http://cpdn-upload2.oerc.ox.ac.uk/cgi-bin/file_upload_handler

22.07.2015 05:41:43 | climateprediction.net | [file_xfer] http op done; retval -184 (transient HTTP error)
22.07.2015 05:41:43 | climateprediction.net | [file_xfer] file transfer status -184 (transient HTTP error)
22.07.2015 05:41:43 | climateprediction.net | Temporarily failed upload of hadam3p_pnw_ppwa_2013_1_009981129_0_8.zip: transient HTTP error
22.07.2015 05:41:43 | climateprediction.net | [file_xfer] project-wide xfer delay for 666.026185 sec
22.07.2015 05:41:43 | climateprediction.net | Backing off 00:27:49 on upload of hadam3p_pnw_ppwa_2013_1_009981129_0_8.zip
22.07.2015 05:41:46 | | Project communication failed: attempting access to reference site
22.07.2015 05:41:47 | | Internet access OK - project servers may be temporarily down.

---

Seems to be running now.

Yes, mine started to upload about an hour ago and the backlog has now cleared :)

I've been getting these "transient HTTP error" messages here - on about 30 hadcm3s uploads pending, each upload failing at a different percentage through. (I think this is unrelated, but in case it matters: currently the pending uploads are consuming the whole quota of disk space.)

7/29/2015 10:32:27 PM | climateprediction.net | Temporarily failed upload of hadcm3s_9tgm_1986_2_009894563_1_2.zip: transient HTTP error
7/29/2015 10:32:27 PM | climateprediction.net | Backing off 03:34:52 on upload of hadcm3s_9tgm_1986_2_009894563_1_2.zip
7/29/2015 10:32:27 PM | climateprediction.net | Started upload of hadcm3s_a2fc_1996_2_009906181_0_1.zip
7/29/2015 10:34:40 PM | climateprediction.net | Temporarily failed upload of hadcm3s_a90p_1996_2_009914726_2_2.zip: transient HTTP error
7/29/2015 10:34:40 PM | climateprediction.net | Backing off 03:31:16 on upload of hadcm3s_a90p_1996_2_009914726_2_2.zip
7/29/2015 10:34:42 PM | | Project communication failed: attempting access to reference site
7/29/2015 10:34:43 PM | | Internet access OK - project servers may be temporarily down.

---

currently the pending uploads are consuming the whole quota of disk space.

I see that your computer has had quite a few pnw tasks, possibly others as well fail. Often when tasks fail they don't delete the task folder in ...projects/clmateprediction.net It is likely that this is what is taking up lots of space rather than the failed uploads.

To reclaim the lost space go into wherever windows keeps your data for BOINC and you can safely delete folders for tasks that don't appear on your BOINC task list. This can be over 1GB/task.

Anyone running linux who hasn't noticed, the short tasks don't clean up even after completing successfully. I just reclaimed 20GB on one of my boxes yesterday!

I see that your computer has had quite a few pnw tasks, possibly others as well fail. Often when tasks fail they don't delete the task folder in ...projects/clmateprediction.net It is likely that this is what is taking up lots of space rather than the failed uploads.

To reclaim the lost space go into wherever windows keeps your data for BOINC and you can safely delete folders for tasks that don't appear on your BOINC task list. This can be over 1GB/task.

Thanks, that was the problem (for the disk space, anyway)!

---

It's been 4 days since my last successful upload of TRIFIDS .zips I constantly get the transient HTTP ERROR......and we are in the weekend...

It appears that there is a problem with the upload of zip files from hadam3p_afr tasks. I now have all 13 zip from a finished stuck in my transfer tab.

---

Don't expect any zips going to oxford to work before Monday now. With luck, the credit script will also be restarted then. Having only linux boxes I haven't kept track of which zips go where but if anyone has tasks left that send them to Oregon or anywhere except Oxford these should still work.

Edit: The credit script seems to have run.

I hadn't checked my Linux box for a few days, and now I discover it has 31 uploads pending, about 1.9GB total. My Windows computer has 4 uploads pending, for 120MB. So I hope it does get cleared up tomorrow, but there may be quite a flood of data considering how many participants are are probably affected by the backlog.

Curiously, downloads of Linux tasks is not impeded. No problem running them, but communication being so unreliable leads me to think there's risk of losing the work that gets done.

The IT people still haven't gotten all of the services running again after the shut down. That will start happening soon after start of Uni working hours, Monday.

And the upload servers where all of this is going may have run out of space.

New message to me, understandable in the circumstances:

09/11/2015 11:11:51 | climateprediction.net | Sending scheduler request: To send trickle-up message.
09/11/2015 11:11:51 | climateprediction.net | Not requesting tasks: too many uploads in progress
09/11/2015 11:11:53 | climateprediction.net | Scheduler request completed

All my Windows work is now finished, so I'll dust off my copy of Virtual Box and get some Linux models.

I asked about the lack of new work, "just in case", and have just been informed that Windows tasks have been waiting on the return to work after the project's recent hospital visit. As I suspected was the case.

At last! The hadam3p_afr zip files are starting to upload.

Mine too. They've finished now :)

Hello Misty

Those messages aren't error codes, just BOINC telling you that it can't find the last few zip files.

I posted an answer to a similar question a few days ago here.

Misty -

I think you are being too concerned.

Sometimes the task tries to upload more zip files than were produced. This not a "failed" task. It is not abnormal.

Generally, if you don't see any "errors" in the Stderr text and the exit code is 0 (zero), you can be almost 100% sure everything is fine.

While I don't think you have any issues, you might want to change some settings to reduce the number of times the model gets interrupted (suspended). The task you pointed to in the previous post shows literally hundreds of suspensions. This is never good.

---

With the continuing problem of uploads getting stuck, I've found this old(ish) thread about uploads.

Please move to using this one, instead of the Credits thread.

Oxford will be starting Trinity term this Sunday, so hopefully the IT people from all over the city will be back from holidays this weekend.
As long as those that have been holding the fort don't suddenly take some leave. :)

Dates of Term

My eu25's are uploading OK to upload3, so it must just be upload2 that has a problem.