BOINC hasn\'t contacted a server since the 8th of March, and I keep getting this error when I start up:

2006-03-23 10:27:21 [climateprediction.net] Scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi failed with a return value of 500

Followed by:

2006-03-23 10:27:21 [climateprediction.net] No schedulers responded

Can someone help me?

Carey

---

What version of BOINC are you using Carey? Did you upgrade to 5.2.13 on 8th March? If so it has a problem with some proxy servers that cause this error. The problem is fixed in any version from 5.3.17 onwards in the BOINC download archive (the 5.3.* versions are development releases and come with the usual health warning).

---

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

What version of BOINC are you using Carey? Did you upgrade to 5.2.13 on 8th March? If so it has a problem with some proxy servers that cause this error. The problem is fixed in any version from 5.3.17 onwards in the BOINC download archive (the 5.3.* versions are development releases and come with the usual health warning).

I\'ve been on 5.2.13 since 15 Feb, when I started.

The proxy password was changed on around the 8th of March, and I updated the proxy settings with the new password. It hasn\'t helped.

Why would I be getting a 500 (server configuration) error, rather than a 503 (which is what happens when you access http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi with a web browser?

---

There was a bug in the comms.

Scheduler requests in 5.2.13 include an \'Expect: 100-continue\' in the HTTP header and some proxy servers generate a \'100/Continue\' response to that. This isn\'t handled correctly (i.e. it\'s ignored) and the code drops through with the built-in (and totally misleading) default of an HTTP 500 error.

---

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

Duplicate removed :oops:

---

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

I have updated to 5.3.27 and then 5.3.28, and I now get:

2006-03-28 09:19:52 [climateprediction.net] Sending scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi
2006-03-28 09:19:52 [climateprediction.net] Reason: Requested by user
2006-03-28 09:19:52 [climateprediction.net] (not requesting new work or reporting completed tasks)
2006-03-28 09:19:53 [---] web site RPC to http://www.google.com
2006-03-28 09:19:53 [---] HTTP error: failed sending data to the peer
2006-03-28 09:19:53 [---] HTTP error: failed sending data to the peer

The model\'s running fine. I\'m just not updating.

Any suggestions?

Carey

---

The new versions include a test to see if BOINC can access a few known good sites.
I know 2 are Google and Yahoo. Not sure if the 2 failures after Google means that Google succeeded, and the next 2 failed, or if Google also failed.
But it does mean that you\'re not connecting with parts of the world.

Thyme Lawn will know.

It looks like the outgoing traffic is being blocked somewhere Carey. First thing to check is for DNS, firewall or proxy server problems at your end.

If that fails to turn up anything and you can run a network protocol analyzer (e.g. Ethereal) on your system you can capture the communication sequence. If you need any help analysing a capture file I might be able to help.

---

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