Since the WU is finished, I\'ve got for the 5 zip files the message :
Temporarly failed upload of 14II_100073165_0_1.zip: system I/O

The file is from Bern, but I thought the problem was solved. Any idea of what happens ?

---

I think Bern is down again. I have a workunit for that server that has been retrying for the past few days.

---

Bern seems to be back up.

---

I confirm, every thing is ok now.

Thanks for your messages.

---

I\'ve finished the WU (0z2s_100065920_0) a few weeks ago, but I cann\'t upload the results and I cann\'t download another WU (3dun_000179508_0). Can you help me? Thanks

---

I\'ve finished the WU (0z2s_100065920_0) a few weeks ago, but I cann\'t upload the results and I cann\'t download another WU (3dun_000179508_0). Can you help me? Thanks

Can you cut and paste what is in the Message tab of the BOINC GUI so we can see the errors you are getting?

I have the same problems - no more uploads since weeks - and trying since nearly 10days - i already have have thrown an other complete result away. i am missing a server status page that gets regularly updated automatically. there should be some backup servers available or an alternate way to upload.

28.09.2005 16:52:04|climateprediction.net|Temporarily failed upload of 1qqq_100102146_1_1.zip: -103
28.09.2005 16:52:04|climateprediction.net|Backing off 32 minutes and 10 seconds on upload of file 1qqq_100102146_1_1.zip

---

Horst
There are two upload servers.
Have a look in client_state.xml for the model in question.
Not far from the top, there will be 5 appearances, one for each of the zips.
A few lines before each should be one with \"uploader\".
This line will contain the destination, phkup21.unibe.ch for Bern, uploader1.atm.ox.ac.uk for Oxford.
If it\'s Bern, it may be down again. In which case let us know straight away, so something can be done about it.

NOTE: BOINC will abort the upload attempts after 14 days, and delete the zips.

This was mentioned extensively a few weeks ago when Bern had a hd failure, and then got affected by the floods.
A lot of people lost models because they don\'t check them or these boards.

There is a temporary fix for long delays posted by Thyme Lawn <a href=\"http://www.climateprediction.net/board/viewtopic.php?t=3152&postdays=0&postorder=asc&start=15\"> here.</a>

This line will contain the destination, phkup21.unibe.ch for Bern, uploader1.atm.ox.ac.uk for Oxford.

Hi Les,

i have two different targets with the same problem:

\"1qqq_100102146_1_{1-4}.zip
url=http://cpdn-upload1.comlab.ox.ac.uk/cpdn_cgi/file_upload_handler

and:

469w_200295028_1_1.zip
url=http://phkup21.unibe.ch/boinc/file_upload_handler

with the same problem:
29.09.2005 07:54:56|climateprediction.net|Temporarily failed upload of 469w_200295028_1_1.zip: -103


is it possible to change these upload directions ?

Thanks

Horst

---

29.09.2005 07:54:56|climateprediction.net|Temporarily failed upload of 469w_200295028_1_1.zip: -103

Error -103 indicates a failure to write to the upload server socket. That could point toward a firewall or proxy problem (the upload servers are not the same as the one trickles are sent to).

---

"The ultimate test of a moral society is the kind of world that it leaves to its children." - Dietrich Bonhoeffer

Hi Thyme,

i do not think, there might be a fw or proxy problem (on my side) as i do not have any problems with seti-boinc on the same machine

---

It looks as if the destination address (see below) is not correct - is it possible to change this address ?

Thanks

Horst

---

I haven\'t seen the comlab address before, but it IS shown on the Project Stats page.
I\'m not sure about changing it.
Does anyone know what this server is for?

I thought there had always been 2 at oxford

Carl wrote on the upload sever problem thread on 25 Aug http://www.climateprediction.net/board/viewtopic.php?t=3152&start=0

\"The uploader1.atm and cpdn-upload1.comlab sites at Oxford are OK, the Bern, Switzerland one is still down (hopefully not underwater -- bad flooding there!)

The main server was down a bit due to problems with adding the sulphur cycle experiment workunits (more about this later...)\"

Classic stats are on that server and since ~26th September the stats have been posted a little earlier than usual. I wonder if this means some routine needed for uploads has crashed so less diversion of time away from stats generation. I think I\'ll send a PM to Carl.

---

Visit BOINC WIKI for help

And join BOINC Synergy for all the news in one place.

I think I\'ll send a PM to Carl.

any news ?

---

Hmm. It is still uncollected in my outbox. So maybe Carl is ignoring these until the spinups are ready so he can then trawl for participants for that.

(I also thought it might have been fixed on the 4th but we are back to early stats on the 5th.)

For the moment, can you follow Thyme Lawn\'s solution to the 2 week limit
detailed <a href=\"http://www.climateprediction.net/board/viewtopic.php?t=3152&postdays=0&postorder=asc&start=15\">here</a>.

---

Visit BOINC WIKI for help

And join BOINC Synergy for all the news in one place.

On top of everything else I imagine Carl and Tolu are tied up making the server software changes enforced by the imminent release of BOINC 5.2 ...

---

"The ultimate test of a moral society is the kind of world that it leaves to its children." - Dietrich Bonhoeffer

sorry, I\'ve been busy with the cm3 spinups, has the upload gone through? I freed up space on our upload servers. Bern seems to be up so the \"real\" Bern server may be used again in a week. In other news, we\'re interviewing sysadmin/programmers next week to help out here (amen)!