Community
Participate
Working Groups
Hi, When I try to remote debug tomcat 5 (or 4 too) on the same machine I get TimeOutExceptions if the Servlet I debug is a bit complex (see the attatched log). This is true for all 3.X I tried (3.0.1, 3.1M4, 3.1M5a, I20050315-1100). I can connect an I also can debug simple Servlets, but wenn I try to debug some complexer stuff after some steps I get timeouts. If that happens I have do delete the debug session, it can not be recovered (disconnect timeouts). also eclipse freezes for some seconds and is getting slow for a while but this gets away. I tried to set debug timeout to 20000ms, it didn't help. I tried to use the debug plugin attatchen to bug #87144, no change, but I hope it contains some more infos for you. I tried to kill all programs which I think could do something to the network, well my machine is part of a novell-network, I will retry at home as sone I have setup my app not to depend on oracle. Well, I know this may be not reproducable for you, but maybe you can provide a way to give you more information. My System: WinXP, 2GHz , 1GB Ram, 1.4.2_06-b03, tomcat 5.0.30 regards.
Created attachment 18847 [details] Eclipse Log The eclipse log.
Thanks for the report. Your logs show that the timed-out packets eventually get received. I'm not sure what's holding them up. Do you have a sample servlet (that reliably reproduces the problem) that you could attach to this bug report? This could also be a VM issue - ie the target vm really isn't responding fast enough. Have you tried a different VM?
It seems that the debugging gets more stable if I reduce the threads used by tomcat. The problem not completly disappears but happens much more seldom. colleagues of mine still use eclipse 2.1 because of remote debug problems, I know they use the standard configuration of tomcat (which uses much more threads). what was changed between 2.1 and 3.x? regards.
(In reply to comment #2) > Thanks for the report. > Your logs show that the timed-out packets eventually get received. I'm not sure what's holding them I was thinking the TrendMicro OfficeScan firewall, but I turned it off. Maybe some gatway stuff, but I'm using loopback (127.0.0.1), well, I'll talk to our admins to ask what all the netware stuff does and what is otherwise installed on the machine. > up. Do you have a sample servlet (that reliably reproduces the problem) that you could attach to this bug report? This may be hard because of dependecies and copyright reasons, i'll look if I can do something > This could also be a VM issue - ie the target vm really isn't responding fast enough. Have you tried a different VM? I will try JRocket tomorrow. But well with 2.1 everything works with the sun VM (OK, I never tried by myself, but my colleagues did) regards.
made small changes to PacketReceiveManager. I'm still unable to reproduce the timeouts. Added some more debugging info to the error message and made a couple other small changes. Alexander, could you try tomorrow's I-Build when it's available and attach the logs if you get any timeouts while using it?
*** This bug has been marked as a duplicate of 83937 ***
