| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: BOINC Agent Support
Thread: Help Needed - Agent Connectivity Issue |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 7
|
|
| Author |
|
|
Sojyn
Cruncher Joined: Jan 1, 2005 Post Count: 3 Status: Offline Project Badges: 
|
Hi All -
I am having a problem with the BOINC agent (seems to affect UD agents as well). That has happended twice now ... second time is still in progress for over four days now. The last time, it lasted for roughly three days and cleared up on its own. It seems that the agent is unable to get to secure.worldcommunitgrid.org to upload. Right now I have over 20 results trying to upload a couple have made it thought in the last day. When the upload gives up I get the following messages: Sat 30 Sep 2006 04:30:44 PM CDT||Attempting to communicate with [secure.worldcommunitygrid.org] timed out Sat 30 Sep 2006 04:30:47 PM CDT|World Community Grid|Scheduler request to https://secure.worldcommunitygrid.org/boinc/wcg_cgi/fcgi failed with a return value of -182 Sat 30 Sep 2006 04:30:47 PM CDT|World Community Grid|No schedulers responded Running the 5.2 agent on FC4. Connectivity is from a Comcast cable modem and Linksys WRT54G router (same behaviour with a different one as well). This install of the agent was working for quite a while (same with the UD agents). All basic network checks are fine (can resolve the hostname, can ping it, telnet to port 443, etc). No other connectivity issus ... just the agents are having an issue. Ideas? Ive checked other posts and cause.html and have not found anything that seems to explain this behavior. Thanks |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
It sounds like a local network problem.
Anyway, before we try to troubleshoot further, I suggest you upgrade to the latest version of BOINC: 5.4.11. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
It sounds like a local network problem. Anyway, before we try to troubleshoot further, I suggest you upgrade to the latest version of BOINC: 5.4.11. 5.4.9 is the latest client version for Linux. I take it FC4 is Fedora Core 4. [Edit 1 times, last edit by Former Member at Oct 1, 2006 4:56:53 AM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
-182 is the standard timeout ( http://boinc-wiki.ath.cx/index.php?title=Temp...st_Failed_-_Timeout_(-182) )
The servers have not been giving me any problems, so we had best start with complete information about your system, network and security software. Lawrence |
|||
|
||||
|
Sojyn
Cruncher Joined: Jan 1, 2005 Post Count: 3 Status: Offline Project Badges:
|
Agent has been upgraded to v5.4.9. No fix so far.
It sounds like a local network problem. Anyway, before we try to troubleshoot further, I suggest you upgrade to the latest version of BOINC: 5.4.11. |
||
|
|
Sojyn
Cruncher Joined: Jan 1, 2005 Post Count: 3 Status: Offline Project Badges:
|
-182 is the standard timeout ( http://boinc-wiki.ath.cx/index.php?title=Temp...st_Failed_-_Timeout_(-182) ) The servers have not been giving me any problems, so we had best start with complete information about your system, network and security software. Lawrence The system us an AMD 64 3700+ w/1 GB RAM. Running Fedora Core 4. SeLinux is disabled. It is presently connected into that back of a Linsys WRT54G rounter which is connected to a Linksys BEFCMU10 cable modem. Firewall options on the linux firewall are trusting https, ftp, mail, www, and ssh ports as well as eth0. I've upgraded the agent to v5.4.9 since my original post. Interestingly, I am getting different messages now: Sun 01 Oct 2006 02:04:05 AM CDT||Access to reference site succeeded - project servers may be temporarily down. Sun 01 Oct 2006 02:08:34 AM CDT||Project communication failed: attempting access to reference site and occasionally: Sun 01 Oct 2006 02:00:30 AM CDT|World Community Grid|Scheduler request failed: a timeout was reached If I didn't mention it in my original post, I am able to download work units w/o any errors. I just got another one despite the messages I am getting when it attempts an upload. Please let me know what other information would be helpful. Thanks for your help. |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
hmmm......that wiki page is as informative as a blank page, but 182 does pop up on a number of BOINC project fora like here:
----------------------------------------http://boinc.berkeley.edu/dev/forum_thread.php?id=204#993 Message 993 - Posted 22 Nov 2005 11:57:18 UTC - in response to Message ID 992. Last modified: 22 Nov 2005 11:58:57 UTC A) The application.exe file gets a temporary failed to dl with error -182 and is put on retry for X minutes. The wu's dl's fine but then BOINC try's to work on them (despite that the .exe hasn't finished dl'ing and isn't ready to run) and fails of course, since there are no .exe yet. You can look up "-182" in the Wiki, but basically you have some kind of network problem that needs to be addressed. I get the _very_ rare instance where something doesn't download properly, but nothing like this... are you on dialup or cable or DSL? Using a proxy? There are several network gurus around that can probably help you solve that part. As for trying to work on a result without the science app being present, yeah, that's pretty stupid. If it hasn't already been reported as a bug (I think it has), I'll do so later. C) By the time I shut down my computer for the night This isn't a "fix", but it may be a workaround - let BOINC download the work, and as soon as it starts, hit "No new work" in the Projects tab - and then in the Work tab, hit "Suspend" on as many of the results as you can before they start trying to execute. Let it keep trying to get the application (you can retry without the wait if you like, using "Retry now" in the Transfers tab) - leave your computer on overnight, whatever, if at all possible. It sounds like whatever the issue is, it doesn't affect smaller downloads, only the larger ones like the applications. And if you've attached to other projects before successfully, even if you had this problem initially, they do get through sometimes. Also - you probably should upgrade to BOINC V5.x, as there is some "different" networking code in the newer versions. It _might_ solve your problem. Overall, it seems to be an improvement, although some people who were working fine with V4.x have had some difficulties with the changes. the upgrading out of the way, u find no admission, but WCG is somewhere on a grey spot of the internet within the IBM enclosure (and i certainly wont admit i am, because the rest of the w3w races along). What helped me with WCG time-out errors (on the browser) was allowing port 443 & 8080, (yes twice 80) explicitly to come from sites like www.world.....grid.org, secure.world....grid.org and server.world....grid.org in the firewall and thru the router......8080 is risque, thus only WCG should be allowed.
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
|