Thread 'Boinc does now not work with Smoothwall Gaurdian Installed......'

Author	Message
Tim Send message Joined: 25 Jan 11 Posts: 4	Message 36571 - Posted: 25 Jan 2011, 11:52:39 UTC Hi, We have recently installed Smoothwall Gaurdian on the network. Myself or smoothwall support cannot get it to work. We use NTLM auth. Using Putty I have added all the addresses to pass straight though smoothwall filters (801) I have enabled proxy and http logging It is always telling us that "Message from server: Project has no tasks available" Can anyone please help? Port 1234 (changed) is the port used by the proxy to connect Here are our logs: 25/01/2011 11:30:54 SETI@home update requested by user 25/01/2011 11:30:54 SETI@home Sending scheduler request: Requested by user. 25/01/2011 11:30:54 SETI@home Requesting new tasks 25/01/2011 11:30:54 [http_debug] HTTP_OP::init_post(): http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi 25/01/2011 11:30:54 [http_debug] HTTP_OP::libcurl_exec(): ca-bundle set 25/01/2011 11:30:54 [proxy_debug] HTTP_OP::no_proxy_for_url(): http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi 25/01/2011 11:30:54 [proxy_debug] returning false 25/01/2011 11:30:54 [proxy_debug]: setting up proxy 1.2.3.4:1234 25/01/2011 11:30:54 [http_debug] [ID#1] Info: timeout on name lookup is not supported 25/01/2011 11:30:54 [http_debug] [ID#1] Info: About to connect() to proxy 1.2.3.4 port 1234 (#0) 25/01/2011 11:30:54 [http_debug] [ID#1] Info: Trying 1.2.3.4... 25/01/2011 11:30:54 [http_debug] [ID#1] Info: Connected to 1.2.3.4 (1.2.3.4) port 1234 (#0) 25/01/2011 11:30:54 [http_debug] [ID#1] Info: Proxy auth using Digest with user '' 25/01/2011 11:30:54 [http_debug] [ID#1] Sent header to server: POST http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi HTTP/1.1 25/01/2011 11:30:54 [http_debug] [ID#1] Sent header to server: User-Agent: BOINC client (windows_intelx86 6.10.58) 25/01/2011 11:30:54 [http_debug] [ID#1] Sent header to server: Host: setiboinc.ssl.berkeley.edu 25/01/2011 11:30:54 [http_debug] [ID#1] Sent header to server: Accept: / 25/01/2011 11:30:54 [http_debug] [ID#1] Sent header to server: Accept-Encoding: deflate, gzip 25/01/2011 11:30:54 [http_debug] [ID#1] Sent header to server: Proxy-Connection: Keep-Alive 25/01/2011 11:30:54 [http_debug] [ID#1] Sent header to server: Content-Type: application/x-www-form-urlencoded 25/01/2011 11:30:54 [http_debug] [ID#1] Sent header to server: Content-Length: 0 25/01/2011 11:30:54 [http_debug] [ID#1] Sent header to server: 25/01/2011 11:30:58 [http_debug] [ID#1] Info: HTTP 1.0, assume close after body 25/01/2011 11:30:58 [http_debug] [ID#1] Received header from server: HTTP/1.0 200 OK 25/01/2011 11:30:58 [http_debug] [ID#1] Received header from server: Date: Tue, 25 Jan 2011 11:33:53 GMT 25/01/2011 11:30:58 [http_debug] [ID#1] Received header from server: Server: Apache/2.2.15 (Fedora) 25/01/2011 11:30:58 [http_debug] [ID#1] Received header from server: Content-Type: text/xml 25/01/2011 11:30:58 [http_debug] [ID#1] Received header from server: X-Cache: MISS from sw11.countyinternet.net 25/01/2011 11:30:58 [http_debug] [ID#1] Received header from server: X-Cache: MISS from school.countyinternet.net 25/01/2011 11:30:58 [http_debug] [ID#1] Received header from server: X-Cache-Lookup: MISS from school.countyinternet.net:1234 25/01/2011 11:30:58 [http_debug] [ID#1] Received header from server: X-Cache: MISS from smoothwall.school.local 25/01/2011 11:30:58 [http_debug] [ID#1] Received header from server: Via: 1.1 sw11.countyinternet.net:801 (squid/2.7.STABLE6), 1.0 school.countyinternet.net:1234 (squid/2.6.STABLE21), 1.0 smoothwall.school.local:801 (squid/2.7.STABLE6) 25/01/2011 11:30:58 [http_debug] [ID#1] Received header from server: Connection: Close 25/01/2011 11:30:58 [http_debug] [ID#1] Received header from server: 25/01/2011 11:30:58 [http_debug] [ID#1] Info: Expire cleared 25/01/2011 11:30:58 [http_debug] [ID#1] Info: Closing connection #0 25/01/2011 11:30:59 SETI@home Scheduler request completed: got 0 new tasks 25/01/2011 11:30:59 SETI@home Message from server: Error in request message: fgets() failed 25/01/2011 11:36:04 SETI@home Sending scheduler request: To fetch work. 25/01/2011 11:36:04 SETI@home Requesting new tasks 25/01/2011 11:36:04 [http_debug] HTTP_OP::init_post(): http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi 25/01/2011 11:36:04 [http_debug] HTTP_OP::libcurl_exec(): ca-bundle set 25/01/2011 11:36:04 [proxy_debug] HTTP_OP::no_proxy_for_url(): http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi 25/01/2011 11:36:04 [proxy_debug] returning false 25/01/2011 11:36:04 [proxy_debug]: setting up proxy 1.2.3.4:1234 25/01/2011 11:36:05 [http_debug] [ID#1] Info: timeout on name lookup is not supported 25/01/2011 11:36:05 [http_debug] [ID#1] Info: About to connect() to proxy 1.2.3.4 port 1234 (#0) 25/01/2011 11:36:05 [http_debug] [ID#1] Info: Trying 1.2.3.4... 25/01/2011 11:36:05 [http_debug] [ID#1] Info: Connected to 1.2.3.4 (1.2.3.4) port 1234 (#0) 25/01/2011 11:36:05 [http_debug] [ID#1] Info: Proxy auth using Digest with user '' 25/01/2011 11:36:05 [http_debug] [ID#1] Sent header to server: POST http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi HTTP/1.1 25/01/2011 11:36:05 [http_debug] [ID#1] Sent header to server: User-Agent: BOINC client (windows_intelx86 6.10.58) 25/01/2011 11:36:05 [http_debug] [ID#1] Sent header to server: Host: setiboinc.ssl.berkeley.edu 25/01/2011 11:36:05 [http_debug] [ID#1] Sent header to server: Accept: / 25/01/2011 11:36:05 [http_debug] [ID#1] Sent header to server: Accept-Encoding: deflate, gzip 25/01/2011 11:36:05 [http_debug] [ID#1] Sent header to server: Proxy-Connection: Keep-Alive 25/01/2011 11:36:05 [http_debug] [ID#1] Sent header to server: Content-Type: application/x-www-form-urlencoded 25/01/2011 11:36:05 [http_debug] [ID#1] Sent header to server: Content-Length: 0 25/01/2011 11:36:05 [http_debug] [ID#1] Sent header to server: 25/01/2011 11:36:05 [http_debug] [ID#1] Info: Expire cleared 25/01/2011 11:36:14 [http_debug] [ID#1] Info: HTTP 1.0, assume close after body 25/01/2011 11:36:14 [http_debug] [ID#1] Received header from server: HTTP/1.0 200 OK 25/01/2011 11:36:14 [http_debug] [ID#1] Received header from server: Date: Tue, 25 Jan 2011 11:39:09 GMT 25/01/2011 11:36:14 [http_debug] [ID#1] Received header from server: Server: Apache/2.2.15 (Fedora) 25/01/2011 11:36:14 [http_debug] [ID#1] Received header from server: Content-Type: text/xml 25/01/2011 11:36:14 [http_debug] [ID#1] Received header from server: X-Cache: MISS from sw15.countyinternet.net 25/01/2011 11:36:14 [http_debug] [ID#1] Received header from server: X-Cache: MISS from school.countyinternet.net 25/01/2011 11:36:14 [http_debug] [ID#1] Received header from server: X-Cache-Lookup: MISS from school.countyinternet.net:1234 25/01/2011 11:36:14 [http_debug] [ID#1] Received header from server: X-Cache: MISS from smoothwall.school.local 25/01/2011 11:36:14 [http_debug] [ID#1] Received header from server: Via: 1.1 sw15.countyinternet.net:801 (squid/2.7.STABLE6), 1.0 school.countyinternet.net:1234 (squid/2.6.STABLE21), 1.0 smoothwall.school.local:801 (squid/2.7.STABLE6) 25/01/2011 11:36:14 [http_debug] [ID#1] Received header from server: Connection: Close 25/01/2011 11:36:14 [http_debug] [ID#1] Received header from server: 25/01/2011 11:36:14 [http_debug] [ID#1] Info: Expire cleared 25/01/2011 11:36:14 [http_debug] [ID#1] Info: Closing connection #0 25/01/2011 11:36:15 SETI@home Scheduler request completed: got 0 new tasks 25/01/2011 11:36:15 SETI@home Message from server: Error in request message: fgets() failed 25/01/2011 11:36:56 SETI@home Resetting project 25/01/2011 11:37:00 SETI@home Sending scheduler request: To fetch work. 25/01/2011 11:37:00 SETI@home Requesting new tasks 25/01/2011 11:37:00 [http_debug] HTTP_OP::init_post(): http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi 25/01/2011 11:37:00 [http_debug] HTTP_OP::libcurl_exec(): ca-bundle set 25/01/2011 11:37:00 [proxy_debug] HTTP_OP::no_proxy_for_url(): http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi 25/01/2011 11:37:00 [proxy_debug] returning false 25/01/2011 11:37:00 [proxy_debug]: setting up proxy 1.2.3.4:1234 25/01/2011 11:37:00 [http_debug] [ID#1] Info: timeout on name lookup is not supported 25/01/2011 11:37:00 [http_debug] [ID#1] Info: About to connect() to proxy 1.2.3.4 port 1234 (#0) 25/01/2011 11:37:00 [http_debug] [ID#1] Info: Trying 1.2.3.4... 25/01/2011 11:37:00 [http_debug] [ID#1] Info: Connected to 1.2.3.4 (1.2.3.4) port 1234 (#0) 25/01/2011 11:37:00 [http_debug] [ID#1] Info: Proxy auth using Digest with user '' 25/01/2011 11:37:00 [http_debug] [ID#1] Sent header to server: POST http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi HTTP/1.1 25/01/2011 11:37:00 [http_debug] [ID#1] Sent header to server: User-Agent: BOINC client (windows_intelx86 6.10.58) 25/01/2011 11:37:00 [http_debug] [ID#1] Sent header to server: Host: setiboinc.ssl.berkeley.edu 25/01/2011 11:37:00 [http_debug] [ID#1] Sent header to server: Accept: / 25/01/2011 11:37:00 [http_debug] [ID#1] Sent header to server: Accept-Encoding: deflate, gzip 25/01/2011 11:37:00 [http_debug] [ID#1] Sent header to server: Proxy-Connection: Keep-Alive 25/01/2011 11:37:00 [http_debug] [ID#1] Sent header to server: Content-Type: application/x-www-form-urlencoded 25/01/2011 11:37:00 [http_debug] [ID#1] Sent header to server: Content-Length: 0 25/01/2011 11:37:00 [http_debug] [ID#1] Sent header to server: 25/01/2011 11:37:00 [http_debug] [ID#1] Info: Expire cleared 25/01/2011 11:37:11 [http_debug] [ID#1] Info: HTTP 1.0, assume close after body 25/01/2011 11:37:11 [http_debug] [ID#1] Received header from server: HTTP/1.0 200 OK 25/01/2011 11:37:11 [http_debug] [ID#1] Received header from server: Date: Tue, 25 Jan 2011 11:40:06 GMT 25/01/2011 11:37:11 [http_debug] [ID#1] Received header from server: Server: Apache/2.2.15 (Fedora) 25/01/2011 11:37:11 [http_debug] [ID#1] Received header from server: Content-Type: text/xml 25/01/2011 11:37:11 [http_debug] [ID#1] Received header from server: X-Cache: MISS from sw12.countyinternet.net 25/01/2011 11:37:11 [http_debug] [ID#1] Received header from server: X-Cache: MISS from school.countyinternet.net 25/01/2011 11:37:11 [http_debug] [ID#1] Received header from server: X-Cache-Lookup: MISS from school.countyinternet.net:1234 25/01/2011 11:37:11 [http_debug] [ID#1] Received header from server: X-Cache: MISS from smoothwall.school.local 25/01/2011 11:37:11 [http_debug] [ID#1] Received header from server: Via: 1.1 sw12.countyinternet.net:801 (squid/2.7.STABLE6), 1.0 school.countyinternet.net:1234 (squid/2.6.STABLE21), 1.0 smoothwall.school.local:801 (squid/2.7.STABLE6) 25/01/2011 11:37:11 [http_debug] [ID#1] Received header from server: Connection: Close 25/01/2011 11:37:11 [http_debug] [ID#1] Received header from server: 25/01/2011 11:37:11 [http_debug] [ID#1] Info: Expire cleared 25/01/2011 11:37:11 [http_debug] [ID#1] Info: Closing connection #0 25/01/2011 11:37:12 SETI@home Scheduler request completed: got 0 new tasks 25/01/2011 11:37:12 SETI@home Message from server: Error in request message: fgets() failed ID: 36571 ·

Jord Volunteer tester Help desk expert Send message Joined: 29 Aug 05 Posts: 15628	Message 36572 - Posted: 25 Jan 2011, 14:45:09 UTC - in response to Message 36571. Last modified: 25 Jan 2011, 14:46:37 UTC "Message from server: Project has no tasks available" This means you got through. Message from server. That's the Seti server answering your plea. Just as the next one: SETI@home Message from server: Error in request message: fgets() failed SETI@home Message from server. "Project has no tasks available" should be clear: The project had no work available at the time your BOINC made contact and asked for work. "Tasks" == work units. Check at the project, in this case Seti@Home, for more information (News on the home page, their Technical News, their forums if need be). "Error in request message: fgets() failed" does seem to flag a failure of the sched_request file getting in correctly at the other end. This may be a simple corruption on the hard drive (check that), or a communications problem. Now, I do see one thing in your log. By default BOINC assumes to use HTTP 1.1 for communications, even in proxies. You can force BOINC to use HTTP 1.0 only, through use of a core client configuration file (cc_config.xml) with the following entries: <cc_config> <options> <http_1_0>1</http_1_0> </options> </cc_config> Make it with s similar simple text editor(e.g. Notepad in Windows), saving the files as all files, ANSI encoding, called cc_config.xml (make sure it has the .XML extension, not .TXT as well!). Save it in your BOINC DATA directory. If you already have a cc_config.xml file, you can add the part of <options> in. Exit BOINC & restart it to make it read the configuration file. (you can re-read it through the GUI, or command line as well). ID: 36572 ·

Tim Send message Joined: 25 Jan 11 Posts: 4	Message 36698 - Posted: 2 Feb 2011, 14:43:13 UTC Sorry I should have added before.... I have 2 machines with BOINC installed, one is my machine and this bypasses the smoothwall proxy. The other is set to go through the proxy. Mine DOES get work through. Where as the machine that students/staff used does not. "No Work Available". I obviously cannot bypass the proxy on the staff machine as they need to have the internet filtered. I have tried both 1.1 and 1.0.. Both do not work. Thanks for the reply. Any further ideas anyone please? ID: 36698 ·

Gundolf Jahn Send message Joined: 20 Dec 07 Posts: 1069	Message 36699 - Posted: 2 Feb 2011, 15:16:22 UTC - in response to Message 36698. Again: "Message from server: Project has no tasks available" is nothing to do with your machine or the firewall or anything local. Gruß, Gundolf ID: 36699 ·

Tim Send message Joined: 25 Jan 11 Posts: 4	Message 36709 - Posted: 3 Feb 2011, 12:11:12 UTC - in response to Message 36699. Again: "Message from server: Project has no tasks available" is nothing to do with your machine or the firewall or anything local. Gruß, Gundolf Hi, I have 2 machines, both identical bionc setup, both connected to the same project. When the proxy is bypassed on EITHER machine, it downloads and processes tasks. As soon as the client is changed to run through the smoothwall proxy the messages posts: "Message from server: Project has no tasks available" If I then change it back to bypass the proxy it then processes tasks once more. I can have one machine processing tasks at the same time whilst the other says "no tasks" and then also swap them over so the other says "no tasks" and the other works. Many thanks for the help but I think you are incorrect saying its nothing to do with firewall or PC... Something is somewhere gettting blocked ID: 36709 ·

Jord Volunteer tester Help desk expert Send message Joined: 29 Aug 05 Posts: 15628	Message 36710 - Posted: 3 Feb 2011, 12:49:21 UTC - in response to Message 36709. Many thanks for the help but I think you are incorrect saying its nothing to do with firewall or PC... As near as I can figure, it's the proxy server that does this. It caches pages and contacts, and as such it will use the last message it got from the project to tell you it can't make contact whether that's the correct message for this moment or not. Normally anything saying "Message from server" means that you got through to the project but that the project has the problem with your request, it will tell you what kind of problem after the message from server. Ask the people maintaining the proxy server to refresh the Seti page on that server. Now, the Seti project did have some contact problems the past couple of days, I'm not sure if they have fixed those as I no longer voluntarily get work from them. ID: 36710 ·

Richard Haselgrove Volunteer tester Help desk expert Send message Joined: 5 Oct 06 Posts: 5149	Message 36711 - Posted: 3 Feb 2011, 14:25:00 UTC - in response to Message 36710. SETI hasn't had any communication problems that I've seen since the end of this week's maintenance window, around 16:00 Pacific time Tuesday 01 Feb, 00:00 UTC Weds 02 Feb. Even the proposed lab-wide outage on Wednesday didn't cause any comms loss, so far as I could see. But as always, congestion after maintenance may have meant that individual works requests failed for a while. Jord, surely the 'message from server' and the subsequent reason should arrive in the same sched_reply_...xml from the server: and that must have arrived in its entirety, otherwise the message log would contain 'parse error' or 'incomplete' remarks. And surely, BOINC scheduler RPC generators - both client and server - should automatically put "no cache" directives, or whatever the proxy language equivalent is, in front of scheduler RPCs - otherwise the sequence numbers might get scrambled, and we'd be seeing random CPID changes.... ....oh, sorry, I forgot for a moment. We are seeing random CPID changes, already, aren't we? ID: 36711 ·

Gundolf Jahn Send message Joined: 20 Dec 07 Posts: 1069	Message 36712 - Posted: 3 Feb 2011, 15:43:34 UTC - in response to Message 36709. When the proxy is bypassed on EITHER machine, it downloads and processes tasks. As soon as the client is changed to run through the smoothwall proxy... Is the proxy set up correctly in BOINC manager? Gruß, Gundolf ID: 36712 ·

Tim Send message Joined: 25 Jan 11 Posts: 4	Message 37345 - Posted: 31 Mar 2011, 13:17:23 UTC Hi, Yes the proxy settings have been enabled in the GUI Thanks Still no progress here :( Smoothwall are currently doing a TCPIP dump ID: 37345 ·

Pepo Send message Joined: 3 Apr 06 Posts: 547	Message 37355 - Posted: 31 Mar 2011, 20:51:32 UTC Last modified: 31 Mar 2011, 21:31:34 UTC A few years ago I've been 'living' behind a NTLM proxy too. From the time I call to remembrance that different NTLM connection modes have been implemented, for various nuances. Just where to find the related settings... [edit]From what I've found out in the dust... Maybe some tips will work for you: From BOINC core client : Internet access thru proxy thread: o BOINC 5.8.16 o http://ntlmaps.sourceforge.net/ - NTLM Authorization Proxy Server (se an example for setup) o BOINC 5.10.40 / 5.10.41 From BOINC core client : No comms through Proxy Sever thread: o Set up cc_config.xml with <force_auth>xx</force_auth>, where 'xx' is any of the available flags: basic \| digest \| gss-negotiate \| ntlm I remember some other long-lasting thread, devoted to NTLM problems, but can not find any trace of it. (Maybe it was on the mailing lists?) But I believe that <force_auth> was the last BOINC revolution on this field. Hope it helps... Peter ID: 37355 ·

Copyright © 2025 University of California.
Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.