Community
Participate
Working Groups
I always get "Timeout waiting for JBoss 4.0 to start. Server did not start after 50s." error message. jboss starts on my machine within 2 minutes so I _always_ get that "error message" it should be preference variable so I could set this timeout higher or turn it off at all.
If the specific timeout is unreasonable for an average server please reopen. Otherwise, there already is a server timeout delay preference in the Server preference page. It allows you to set the timeout larger or smaller based on your configuration.
server timeout delay: Normal - I get "Timeout waiting for JBOSS 4.0 to start. Server did not start after 50s." server timeout delay: Shorter - I get "Timeout waiting for JBOSS 4.0 to start. Server did not start after 50s." server timeout delay: Longer - I get "Timeout waiting for JBOSS 4.0 to start. Server did not start after 50s." It doesn't work. I start JBoss in debug mode
Reproduced. The server timeout preference isn't being used correctly.
Fix dropped to HEAD. Will be released to 1.5.1/2.0 shortly.
I have 1.5.1-200607141912 installed. This bug is still present. Also, if the timeout is exceeded, the process terminates, so I am unable to start Tomcat, as it takes more than 45 seconds. (In reply to comment #4) > Fix dropped to HEAD. Will be released to 1.5.1/2.0 shortly.
I have the last successful Junit build M-1.5.1-200607201712 and the server timeout still stays at 45s (for tomcat) even after selecting different timeout values (long, longer etc..) from the server preferences. The server also terminates after timeout. (in 1.0.3, it used to give a message but continue to run)
In build M-1.5.1-200607211907 this bug is still present...
Note the comment "Fix dropped to HEAD. Will be released to 1.5.1/2.0 shortly." I didn't drop code this past week (final 1.0.3 week and working on another conflicting bug) and the bug is still in ASSIGNED state, not RESOLVED/FIXED. I should be able to drop it this week, but in the meantime of course the behaviour of the release builds has not changed.
Sorry, Tim. I thought the continuous builds were pulled out of HEAD. (In reply to comment #8)
Released for the next WTP 1.5.1 & 2.0 builds.
Thanks, that works! It still terminates after timeout, though. What happens if it takes more than 81 seconds to start Tomcat? (In reply to comment #10)
*more* than 80s? :) If you feel the startup is still within 'normal' operating bounds, then please open up a bug to increase the default timeout so that by tweaking the preference you can still get the server to startup. If this really is an edge case, then this should probably be tracked via bug 121593, which asks for fully configurable timeouts instead of a single preference.
Well, I feel it is normal, but perhaps others can comment on this. Many in my team have had trouble with this timeout, even at 80 seconds. Throwing 5 or 6 large webapps in Tomcat's webapps directory will cause it to take about 60 - 100 seconds to start up on a laptop with 1.73GHz and 1GB of RAM. (In reply to comment #12)
Ok, then please open up another bug with the version of Tomcat you are using and that average startup time range, and I'll adjust the default timeouts accordingly. Thanks.
Why not simply change this to an integer and allow to turn it off?
It is an integer under the covers and that's likely the route we'll take, but the remaining issue (timeout is too low for a heavily-loaded Tomcat) is being tracked via bug 151993.
This is part of a "mass update" where bugs showing "fixed" between 7/1/2006 and 9/27/2006 are assumed fixed in WTP 1.5.1. feel free to change the target milestone if you know a more accurate setting.
New Gerrit change created: https://git.eclipse.org/r/107899
