search

wibbleh / upk-modder

Automatically exported from code.google.com/p/upk-modder
0 stars 0 forks source link

System freezes #13

Closed GoogleCodeExporter closed 8 years ago

GoogleCodeExporter commented 8 years ago 
I've experienced three system freezes while using version 0.65, requiring a 
hard reboot. It's of course hard to be certain upk-modder is the culprit. 

In the most recent, the mouse cursor went a bit haywire -- it looked stuck on 
the resize left-right arrow for a bit before locking up completely.

Original issue reported on code.google.com by johnnyl...@gmail.com on 26 Dec 2013 at 6:53

GoogleCodeExporter commented 8 years ago 
Upkmodder froze up but system didn't. Here's the log:

Dec 27, 2013 6:57:00 PM model.upk.UpkHeader <init>
INFO: Constructed searchable names, took 0ms
Dec 27, 2013 6:57:05 PM model.modtree.ModTree updateDocument
INFO: 0 lines out of 0 re-styled: 2063 total restyling events
Dec 27, 2013 6:57:23 PM ui.ModTab testBeforeAndAfterBlocks
INFO: Block 0 bytecount mismatch. FIND = 443, REPLACE = 435
Exception in thread "D3D Screen Updater" java.lang.OutOfMemoryError: Java heap s
pace
        at java.lang.reflect.Array.newArray(Native Method)
        at java.lang.reflect.Array.newInstance(Unknown Source)
        at java.util.Arrays.copyOf(Unknown Source)
        at java.util.ArrayList.toArray(Unknown Source)
        at sun.java2d.d3d.D3DScreenUpdateManager.run(Unknown Source)
        at java.lang.Thread.run(Unknown Source)
Dec 27, 2013 6:58:35 PM model.modtree.ModTree$ModTreeDocumentListener$1 run
SEVERE: Failure in ModTree event processing threading: java.lang.reflect.Invocat
ionTargetException

Exception: java.lang.OutOfMemoryError thrown from the UncaughtExceptionHandler i
n thread "TimerQueue"

Original comment by johnnyl...@gmail.com on 28 Dec 2013 at 2:01

GoogleCodeExporter commented 8 years ago 
I'm suspecting some of these errors might be related to not properly closing 
files. File handling has been cleaned up, so check to see if the errors persist 
in v0.70.

The error reported in #2 appears related to the currently inefficient way we 
spawn threads for the ModTree to respond to Document event changes (result of 
user typing / cut / paste operations).

Currently one new thread is spawned for each document event, although that 
thread simple updates and consumes the entire document event queue. My attempts 
to only spawn a new thread in the event that one isn't running have been 
unsuccessful.

Original comment by miss.inv...@gmail.com on 29 Dec 2013 at 2:49

GoogleCodeExporter commented 8 years ago 
Using .71, still getting errors. Java showing 36,000 K memory usage on Task 
Manager at startup, and it climbs steadily -- goes up when scrolling through 
modfile or making any changes. Saw crash at around 300,000 K.

(Win 7 64, 8 GB Ram)

Original comment by johnnyl...@gmail.com on 31 Dec 2013 at 8:17

GoogleCodeExporter commented 8 years ago 
Seems to have been resolved.

Original comment by johnnyl...@gmail.com on 9 Feb 2014 at 3:46

GoogleCodeExporter commented 8 years ago 
Issue appears to be resolved separately - closing.

Original comment by miss.inv...@gmail.com on 25 Feb 2014 at 6:29