Bug #56212 | Random Crashes | ||
---|---|---|---|
Submitted: | 24 Aug 2010 6:32 | Modified: | 20 Oct 2010 12:24 |
Reporter: | Roberto Crisanti | Email Updates: | |
Status: | No Feedback | Impact on me: | |
Category: | MySQL Workbench | Severity: | S1 (Critical) |
Version: | 5.2.26 | OS: | MacOS (10.6.4) |
Assigned to: | Alfredo Kojima | CPU Architecture: | Any |
Tags: | crash |
[24 Aug 2010 6:32]
Roberto Crisanti
[24 Aug 2010 6:35]
Roberto Crisanti
crash log
Attachment: log.txt (text/plain), 64.47 KiB.
[24 Aug 2010 17:19]
Rob Palkowski
Another crash log
Attachment: wb-crash.log (application/octet-stream, text), 47.96 KiB.
[24 Aug 2010 17:24]
Rob Palkowski
I too am having seemingly random crashes of the program. I have attached another log for reference. Sometimes the program will work for hours before crashing; sometimes it crashes before I can even establish a database connection. I have had crashes at all of the following states: - At the Home screen just after having launched the program - While editing a stored routine from the either the ERR or SQL Editor view - While editing or creating tables from either the ERR or SQL Editor view - While selecting a file for use as an SSH authentication key
[24 Aug 2010 18:00]
Alfredo Kojima
In ~/Library/Logs/CrashReporter/ there's probably a bunch of crashlogs for WB, please zip them and attach here so we can make sure they're not from different causes.
[24 Aug 2010 18:20]
Rob Palkowski
Accumulated crash logs
Attachment: wb-crash-logs.tgz (application/x-gzip, text), 683 bytes.
[24 Aug 2010 18:20]
Rob Palkowski
I have attached the relevant crash logs from the computer I am using at the moment.
[25 Aug 2010 12:55]
Alfredo Kojima
Hi Seems Apple changed the location of the crashlogs and is just placing symlinks in the dir, so your tgz contains empty symlinks. Please try again using folder ~/Library/Logs/DiagnosticReports Thanks
[25 Aug 2010 18:10]
Rob Palkowski
Actual logs and not symlinks to them
Attachment: wb-crash-logs.tgz (application/x-gzip, text), 104.07 KiB.
[25 Aug 2010 18:11]
Rob Palkowski
I should have seen the filesize of 600 bytes and thought something was amiss. The actual logs have been uploaded.
[25 Aug 2010 18:31]
Alfredo Kojima
The stack traces you provide indicate 2 different causes for a crash. One of them (which is the one you pasted in this bug report) has been recently fixed. The other one is more mysterious and without some way to repeat it, it will be extremely difficult to address. I will mark the bug as fixed, but if you encounter the same issue in version 5.2.27, please reopen it and attach the new crash report you get.
[25 Aug 2010 18:38]
Rob Palkowski
I certainly shall. Will I have to wait for an official release, or is there a place from which to get the pre-release versions of WorkBench?
[26 Aug 2010 9:30]
Roberto Crisanti
mmm maybe it's just a coincidence but I changed the default collation of the schema I'm working on from latin1 to utf8 and it isn't crashing in a while.
[26 Aug 2010 17:18]
Rob Palkowski
Sorry, that's not it- All of my schemae are UTF-8.
[3 Sep 2010 1:01]
Rob Palkowski
I just had another crash with the new version of Workbench. I will be adding the crash log presently.
[3 Sep 2010 1:03]
Rob Palkowski
Crash report log
Attachment: MySQLWorkbench_2010-09-02-175931_Robs-iMac.crash (application/octet-stream, text), 49.43 KiB.
[18 Sep 2010 1:07]
Alfredo Kojima
Roberto: how did you get the crash? Was there some steps you did? Is it repeatable? The crashlog is different from the others.
[18 Sep 2010 1:12]
Rob Palkowski
The last few times I got the crash, it was while editing a stored procedure. I have also gotten crashes while modifying tables and foreign keys. I haven't been using the program for the past few days; I will pay closer attention to exactly what I've been doing at the time of the crash and upload more logs. I do recall that it most often happens just after clicking; the MouseUp events in the logs that I recall seeing seem to indicate that as well.
[20 Oct 2010 23:00]
Bugs System
No feedback was provided for this bug for over a month, so it is being suspended automatically. If you are able to provide the information that was originally requested, please do so and change the status of the bug back to "Open".