View Issue Details
| ID | Project | Category | Date Submitted | Last Update | |
|---|---|---|---|---|---|
| 0004456 | Valley 1 | Crash/Exception | Oct 1, 2011 1:14 pm | Oct 11, 2011 1:25 pm | |
| Reporter | mrhanman | Assigned To | keith.lamothe | ||
| Severity | crash | ||||
| Status | resolved | Resolution | fixed | ||
| Product Version | 0.511 | ||||
| Fixed in Version | 0.512 | ||||
| Summary | 0004456: Exception Exploring Evil Outpost | ||||
| Description | I got an exception while exploring an evil outpost. Not really doing anything special. Maybe moving from room to room faster than normal. | ||||
| Tags | No tags attached. | ||||
| Internal Weight | Game Crash | ||||
| related to | 0004406 | resolved | Chris_McElligottPark | 0.510 Possibly Causing Freezing & Forcing Reboot on Mac |
|
|
Thanks, will take a look :) |
|
|
That's... really odd. From what I'm seeing in the code the only way it could have failed to release the buffer in another part of the code is if there had been an exception before that other part completed its operation. And your UnhandledErrors.txt file makes no mention of any other exceptions. Though perhaps the root cause was pushed out by the sheer spam of this thing. Well, I can stop it from repeatedly hitting that exception, at least. |
|
|
If something odd is going to happen, it'll be while I'm playing it. XD |
|
|
Ok, I really don't think I'll be able to track down what happened from the info available, but for 0.512 I did put in a number of changes that should make it a lot easier for the error log to actually have room for the original error (assuming it was getting bumped out, dunno) if the buffers get messed with, and also made it basically impossible for the buffers to not get properly released if there is an exception inside those branches. None of that fixes the original bug, but it should help us actually be able to find it if it happens again :) Though it might be one of those that doesn't happen again for a long time. So if it happens again in 0.512 (once it hits) or later, please let us know :) |
|
|
Sure thing! |
|
|
I'm wondering if this is at all related to 0004406. |
|
|
It doesn't sound the same; the error log here indicates that frames were still completing (albeit after failing each time; the errors could only occur once each frame since they weren't being caught except at our top level handling), so it wouldn't be freezing. |
|
|
Yes, but he's on windows. I'm suspecting that if there's a crash under certain circumstances on OSX, particularly during saving a world file, that it might be locking up instead of writing the errors as it's doing on windows. I don't think there's an OSX-specific hang, I think there's a crash that is on both OSes that is causing only OSX to hang while windows reports the error properly. So I'd like to make sure we get rid of any crashes that we're seeing on windows (which we'd do anyway, of course), and hope that that helps the hang on OSX. And then I plan to review the code to look for potential OSX hangs anyway. |
|
|
Ok, yea, that's actually pretty likely then. The errors in this report could really only happen from an exception occurring during the serialization process, and it certainly fits with our previous experience that the Mac responds differently to some forms of exception. I should note, though, that the errors that could cause what we saw here would occur during the creation of the serialized string, but strictly before any attempt to write that string to disk. By the time it gets to the writing, the temporary buffer has already been released. |
|
|
Makes sense. |
|
|
Optimistically calling this one resolved ;) |
|
|
I haven't encountered it recently. In hindsight, I was experimenting with using dropbox and symlinks to keep my saves updated between Mac OS and Windows. Could that have had anything to do with this? |
|
|
If dropbox was locking the files... maybe that could have been related. But I don't think dropbox does that, so I wouldn't have expected this from that, honestly. We made a ton of other changes to help try to fix these general sorts of things, though, so hopefully we just fixed it without really directly finding the cause for this specific case. |
| Date Modified | Username | Field | Change |
|---|---|---|---|
| Oct 1, 2011 1:14 pm | mrhanman | New Issue | |
| Oct 1, 2011 1:14 pm | mrhanman | File Added: UnhandledErrors.txt | |
| Oct 1, 2011 1:21 pm | keith.lamothe | Internal Weight | => Game Crash |
| Oct 1, 2011 1:21 pm | keith.lamothe | Note Added: 0015193 | |
| Oct 1, 2011 1:21 pm | keith.lamothe | Assigned To | => keith.lamothe |
| Oct 1, 2011 1:21 pm | keith.lamothe | Status | new => assigned |
| Oct 1, 2011 1:37 pm | keith.lamothe | Note Added: 0015196 | |
| Oct 1, 2011 1:40 pm | mrhanman | Note Added: 0015197 | |
| Oct 1, 2011 2:03 pm | keith.lamothe | Note Added: 0015203 | |
| Oct 1, 2011 2:03 pm | keith.lamothe | Status | assigned => feedback |
| Oct 1, 2011 2:05 pm | mrhanman | Note Added: 0015204 | |
| Oct 1, 2011 2:05 pm | mrhanman | Status | feedback => assigned |
| Oct 1, 2011 2:06 pm | keith.lamothe | Status | assigned => feedback |
| Oct 2, 2011 7:27 pm | Chris_McElligottPark | Relationship added | related to 0004406 |
| Oct 2, 2011 7:27 pm | Chris_McElligottPark | Note Added: 0015350 | |
| Oct 2, 2011 7:39 pm | keith.lamothe | Note Added: 0015354 | |
| Oct 2, 2011 9:23 pm | Chris_McElligottPark | Note Added: 0015360 | |
| Oct 2, 2011 10:27 pm | keith.lamothe | Note Added: 0015370 | |
| Oct 2, 2011 10:45 pm | Chris_McElligottPark | Note Added: 0015371 | |
| Oct 10, 2011 2:24 pm | keith.lamothe | Note Added: 0016375 | |
| Oct 10, 2011 2:24 pm | keith.lamothe | Status | feedback => resolved |
| Oct 10, 2011 2:24 pm | keith.lamothe | Fixed in Version | => 0.512 |
| Oct 10, 2011 2:24 pm | keith.lamothe | Resolution | open => fixed |
| Oct 10, 2011 4:54 pm | mrhanman | Note Added: 0016379 | |
| Oct 11, 2011 1:25 pm | Chris_McElligottPark | Note Added: 0016412 | |
| Apr 14, 2014 9:28 am | Chris_McElligottPark | Category | Bug - Crash or Exception => Crash/Exception |
