Regular crashes on Mac OS

Hello,

since IntelliJ 8.1.x (I used all versions up to 8.1.2 so far) IntelliJ crashes regularly when invoking tasks like running, building or sometimes even formatting. It is a kind of random behaviour and incredible annoying. This is far away from the stability, I used to have with 8.0 or even the 8 prerelase versions. The crashes happens about 2 to 4 times per day.

I attached the mac crash information and the idea.log of the last crash.

Java Version is
java version "1.6.0_13"
Java(TM) SE Runtime Environment (build 1.6.0_13-b03-196)
Java HotSpot(TM) 64-Bit Server VM (build 11.3-b02-77, mixed mode)

Best Regards

Sebastian



Attachment(s):
mac.log
idea.log
19 comments
Comment actions Permalink

Hello Sebastian,

Note that the officially supported JDK on MacOS X is still only JDK 1.5,
not JDK 1.6.

The crashes happen inside the Apple AWT code, so it definitely looks like
the bug is in the latest Apple Java update, not in IntelliJ IDEA.

Here's a log for a crash of a completely different app in the same location:
https://jira.jboss.org/jira/secure/attachment/12321203/JavaNativeCrash_pid4311.crash.log

since IntelliJ 8.1.x (I used all versions up to 8.1.2 so far) IntelliJ
crashes regularly when invoking tasks like running, building or
sometimes even formatting. It is a kind of random behaviour and
incredible annoying. This is far away from the stability, I used to
have with 8.0 or even the 8 prere

lase versions. The crashes happens about 2 to 4 times per day.

I attached the mac crash information and the idea.log of the last
crash.

Java Version is
java version "1.6.0_13"
Java(TM) SE Runtime Environment (build 1.6.0_13-b03-196)
Java HotSpot(TM) 64-Bit Server VM (build 11.3-b02-77, mixed mode)
Best Regards



--
Dmitry Jemerov
Development Lead
JetBrains, Inc.
http://www.jetbrains.com/
"Develop with Pleasure!"


0
Comment actions Permalink

Hello Dmitry,

I upgraded from 1.5 to 1.6 as well so this really could be the reasonb. I will run IntelliJ with 1.5 and provide you with a feedback if stability is improved.

Best Regards

Sebastian Hennebrueder

0
Comment actions Permalink

I can say that since I'm using Java 6, IDEA takes more memory but it's faster. And any crash at all. Just various EAP build issues.

0
Comment actions Permalink

Now a crash using Java 1.5. I used the shortcut for (find in path). The idea.log had no entries during crash time.

Though there is no indication that this is plugin related. Could this still be a plugin. I am not using Ruby in the current project but the plugin is installed.

Best Regards

Sebastian Hennebrueder



Attachment(s):
mac2.log
0
Comment actions Permalink

I am seeing the same type of issue. The "find in path" window stops coming up when envoked (after 1st or 2nd use), doesn't crash, just doesn't work until I restart.

I have seen the same issue running under Java 1.6 and 1.5 (IDEA 8.1.2 and 8.1.3).

0
Comment actions Permalink

Good morning.

Thought I would add my 2cents of data to this thread.

I'm on my second crash this morning (and maybe a dozen crashes over the last few weeks), and have lost a bit of work both times, as if the editor is not autosaving in the run-up to the crash.

Intellij version:  8.1.3  (build 9886)

with

$ java -version
java version "1.6.0_13"
Java(TM) SE Runtime Environment (build 1.6.0_13-b03-211)
Java HotSpot(TM) 64-Bit Server VM (build 11.3-b02-83, mixed mode)

Under OS X 10.5.7

0
Comment actions Permalink

If it's a VM crash you are experiencing, there is no "run-up to the crash".
The whole Java process will die completely without warning or the chance
for IDEA to do anything about it. Other than when you manually trigger a
'Save All', IDEA only saves at certain times like frame deactivation, so
yes you will lose some of your recent changes if the VM rug gets pulled out
from under you. Perhaps try enabling Local History, AFAIK that checkpoints
quite frequently and might reduce the amount of work you lose. The real problem
of course is the poor state of Java on OSX and that doesn't seem to be something
that Apple is going out of its way to address.

Good morning.

Thought I would add my 2cents of data to this thread.

I'm on my second crash this morning (and maybe a dozen crashes over
the last few weeks), and have lost a bit of work both times, as if the
editor is not autosaving in the run-up to the crash.

Intellij version: 8.1.3 (build 9886)

with

$ java -version
java version "1.6.0_13"
Java(TM) SE Runtime Environment (build 1.6.0_13-b03-211)
Java HotSpot(TM) 64-Bit Server VM (build 11.3-b02-83, mixed mode)
Under OS X 10.5.7

---
Original message URL:
http://www.jetbrains.net/devnet/message/5240990#5240990



0
Comment actions Permalink

IDEA is crashing a few times a day for me now. Please don't make me use Eclipse, I'm hoping to be productive today

I haven't come up with a anything to get the repeated crashing to abate at all. Anyone have any ideas?

0
Comment actions Permalink

What do you think Jetbrains can do if Apple releases buggy Java versions and
does not provide a way to install a different one?

0
Comment actions Permalink

They could try and implement a work around, much like they do for clipboard problems in Linux. Thanks for the unhelpful follow up though.

0
Comment actions Permalink

I'm sorry about your problems, but please don't blame Jetbrains but rather
Apple. Not just for the bugs, but also making it nearly impossible to
reinstall a previous, less-buggy Java.

0
Comment actions Permalink

Ever since I upgraded to 8.1.3 IntelliJ keeps crashing on me a couple of times a day when I make changes in an HTML file or TML file (Tapestry) while HTML preview is visible. I don't think this is related to the other complaints above though and I never had that problem in the 8.1.2 release.

0
Comment actions Permalink

I don't know how frequently the IDE automatically saves edit buffers (maybe essentially instantly for every key-up), but I have noticed that at some point prior to the crash, the edit buffer stops being saved to disk.  So, e.g., changes I made to the edit buffer 30s or 1m ago may not appear in the file that was being edited when I restart the IDE after the crash.  That's what I meant by run-up:  a time period greater than what I perceive to be the time it takes for the edit buffer to be saved to disk under normal conditions.  I've seen this lost-work effect a half dozen times since the crashes started, although I do not always necessarily lose work on crash.  And I don't know if this information is helpful or not; I offer it in case it is.

0
Comment actions Permalink

Hi,

While I understand this is not the fault of JetBrains, can you suggest a work around for users on OS X?  I'm having serveral crashes a day using JDK 1.5 and it's becoming unbearable.

Cheers,

Ross

0
Comment actions Permalink

There should be a developer preview of Java available which fixes a couple
of bugs:

http://lists.apple.com/archives/Java-dev/2009/Jul/msg00349.html

Tom

0
Comment actions Permalink

Installing it now. Having Idea crash a minimum of 4 times a day and losing a little bit of work every time is getting kinda old.
I'll report back...

0
Comment actions Permalink

This may be superstitious, but I seem to have reduced the number of crashes/day by assiduously performing manual saves - Command-S.  Now I get perhaps a crash per day or so vs. half a dozen.  Hard to say whether this really helps, but it's worth a try.

0
Comment actions Permalink

Thanks Tom. I'll give it a go.  Though since I posted I found this: http://www.jetbrains.net/devnet/docs/DOC-1155
and IntelliJ has not crashed since (OSX JDK 5). So maybe I will not rock the boat unless it crashes again.

Cheers

0
Comment actions Permalink

Installed that Mac OS java update this morning and so far so good. No crashes today.

0

Please sign in to leave a comment.