Community
Participate
Working Groups
I am getting an error when I run a program that outputs a really long line of text (a soap message in that case). The error starts happening around 12200 characters.
Created attachment 60327 [details] Code sample that reproduces the problem Run this small program in eclipse to reproduce the bug.
Created attachment 60328 [details] Stack trace Stack trace from the eclipse logs.
I'm getting this problem with the latest 3.3 M5 build. This was working fine with version 3.2.x.
FH the stack indicates that this should be a fixable "string too long" case. I'm not able to reproduce the AIOOB (no error for me), but I'm on win2000 and not quite the same eclipse build.
It doesn't happen all the time for me also but I still get it. Sometimes, I have to resize the window and it happens (sometimes only the first time I try it). It usually happens if you run the test program multiple times quickly.
KB believes the bug is here: org.eclipse.ui.internal.console.ConsoleDocumentAdapter.documentAboutToBeChanged(ConsoleDocumentAdapter.java:298)
*** Bug 198776 has been marked as a duplicate of this bug. ***
*** Bug 201075 has been marked as a duplicate of this bug. ***
As reporter of duplicate bug 201075, I can confirm that my case does seem to be a line length issue. In my case, having the following VM options in use triggers the bug (on Sun JDK1.5): -XX:+UseConcMarkSweepGC -XX:+PrintGCDetails -XX:+PrintTenuringDistribution I note that this is yet to be assigned/targetted, and has been around a while. I'd hope that a regression like this is on the list for 3.3.1...
Created attachment 76944 [details] patch
Released to HEAD
Released to 3.3.1. Please verify, Mike.
Created attachment 76951 [details] replacement plug-in This is a replacement plug-in that the community can use for testing. The patched plug-in addresses a problem with '\r' characters - so I am interested to see if it solves everyone's problem.
Hi Darin, I've just tried the replacement plugin, and I still get the problem, so it looks like there's more than one bug to resolve...
verified in 3.4 and 3.3.1 tried using the vm params mentioned in comment 9 and was unable to produce the problem.
Looks like 201075 needs reopening then, as I still have the bug. I wonder if there are some workspace settings that are triggering it...
I'm seeing something similar, but not sure if its related to this one. If I log a lot of events at invocation time, I never get past the splash screen. The logging code seems to hang in IOConsolePartitioner line 466 (3.3) on the wait. if (fBuffer > 160000) { try { pendingPartitions.wait(); } catch (InterruptedException e) { } } (In reply to comment #16) > Looks like 201075 needs reopening then, as I still have the bug. I wonder if > there are some workspace settings that are triggering it... >
I am seeing this problem in Eclipse 3.3.1.1 under OS X. It was adding the VM argument -XX:+PrintGCDetails that seems to have caused it. My stack trace is very similar to the attachment on comment #2. From the trace it appears that this line is causing an ArrayIndexOutOfBoundsException: org.eclipse.swt.custom.StyledTextRenderer.textChanging(StyledTextRend erer.java:1235) I'm using Java 1.5.0_13.
(In reply to comment #18) > I am seeing this problem in Eclipse 3.3.1.1 under OS X. It was adding the VM > argument -XX:+PrintGCDetails that seems to have caused it. My stack trace is > very similar to the attachment on comment #2. From the trace it appears that > this line is causing an ArrayIndexOutOfBoundsException: > org.eclipse.swt.custom.StyledTextRenderer.textChanging(StyledTextRend > erer.java:1235) > I'm using Java 1.5.0_13. Do you have a sample program that causes the problem? does it happen on 3.3.2?
For me, it's happening when I start up JBoss AS 4.2.2 in debug mode from within Eclipse. The console shows log messages until I get to the first full garbage collection at which point I get a line like: [Full GC [Tenured: 5909K->6561K(121024K), 0.0665359 secs] 8122K->6561K(130176K), [Perm : 12543K->12543K(12544K)], 0.0665 971 secs] Then a dialog pops up in Eclipse stating "IOConsole Updater (Time of error: [TIME])" which appears multiple times and the stack trace appears in my Eclipse log. After the console output above nothing further is output on the console view from the server. I will see if I can get 3.3.2 installed to test later today. Was there a patch provided for this in the 3.3.1 code base when the bug was marked fixed?
I just tested on 3.3.2 and the problem seems to be fixed. It might be helpful to others if you could update the target milestone.
This fix did do into 3.3.1, so the milestone is accurate. There was an additional console fix in 3.3.2 - bug 207743. However, it is good to know the problem is fixed in 3.3.2.
I saw it happened in 3.3.2. The output has a very long line.
I have a Java program that interacts with a remote service, as a result slowly outputs '.' characters to the console on a single line. If I have the 'Fixed width console' option enabled in the Console Preferences, as soon as the number of characters output hits the value for the 'Maximum character width' in the Console Preferences (default 80), this error occurs. If I disable 'Fixed width console', it does not.
I just encountered the same problem. Using: Version: 3.4.0 Build id: I20080617-2000 Under: Windows XP SP2, SUN JDK 1.5.0_10-b03 It happens when several very long lines are outputted to the console and "Fixed width console" is ON. (Doesn't happen when it is OFF.)
Please open a new bug and attach a test case.(In reply to comment #25) > I just encountered the same problem. > Using: Version: 3.4.0 Build id: I20080617-2000 > Under: Windows XP SP2, SUN JDK 1.5.0_10-b03 > It happens when several very long lines are outputted to the console and "Fixed > width console" is ON. (Doesn't happen when it is OFF.) Please open a new bug and attach a test case or steps to reproduce. A verified/fixed bug is not going to get any attention :-)
Done -> bug 243877
