Anonymous | Login | Signup for a new account | 2024-03-28 19:55 UTC |
My View | View Issues | Change Log | Roadmap | Zandronum Issue Support Ranking | Rules | My Account |
View Issue Details [ Jump to Notes ] | [ Issue History ] [ Print ] | ||||||||
ID | Project | Category | View Status | Date Submitted | Last Update | ||||
0003110 | Zandronum | [All Projects] Bug | public | 2017-04-30 11:13 | 2018-09-30 21:45 | ||||
Reporter | Korshun | ||||||||
Assigned To | Torr Samaho | ||||||||
Priority | normal | Severity | minor | Reproducibility | always | ||||
Status | closed | Resolution | fixed | ||||||
Platform | Microsoft | OS | Windows | OS Version | XP/Vista/7 | ||||
Product Version | 3.0-beta | ||||||||
Target Version | 3.0 | Fixed in Version | 3.0 | ||||||
Summary | 0003110: Using anything other than console commands to exit to a nonexistent map makes the server enter an infinite loop | ||||||||
Description | "map" and "changemap" check that destination exists before executing. Other ways don't. While switching to a nonexistent map in singleplayer simply stops the game, doing so on a server in 3.0 makes it enter an infinite map loading loop with 100% cpu. But 2.1.2 just goes back to the same map. The server should just stop or go back to previous valid map or not allow going to nonexistent maps in the first place. The attached wad is a map with two switches that lead to nonexistent maps using different methods. Host it on a real server. | ||||||||
Attached Files | exittest.wad [^] (1,893 bytes) 2017-04-30 11:13 | ||||||||
Notes | |
(0017448) Torr Samaho (administrator) 2017-04-30 15:42 |
The offline behavior of 3.0 seems to be consistent with GZDoom 1.8.6. Can you confirm that? If so, is GZDoom 2.4.0 behaving differently? The new behavior of GZDoom 1.8.6, i.e. to error out, seems correct to me. |
(0017449) Korshun (reporter) 2017-04-30 16:08 |
Yes, it is consistent with GZDoom 1.8.6. And GZDoom 1.8.6 is the same as GZDoom 2.4.0. |
(0017481) Torr Samaho (administrator) 2017-05-01 17:25 |
The server should now restart itself on map01/e1m1 when a mod instructs it to switch to a non-existing map. |
(0017486) WaTaKiD (updater) 2017-05-01 18:00 |
this build contains the fix: 'https://www.dropbox.com/s/hayd6havowr3my5/zandronum-3.0-r170501-1721-6265097-windows.zip?dl=0 [^]' |
(0017487) Korshun (reporter) 2017-05-01 18:08 |
It works! But what if the iwad has neither MAP01 nor E1M1? From the code of the fix and CalcMapName it seems it would get stuck in the same endless loop of loading nonexistent MAP01 or E1M1. |
(0017493) Torr Samaho (administrator) 2017-05-01 18:41 |
Yeah, that's right. Are there iwads that have neither? |
(0017495) Korshun (reporter) 2017-05-01 18:50 edited on: 2017-05-01 18:51 |
I am not aware of any. But it's not too far-fetched to imagine an iwad that only uses maps like MYMOD01 or SOMENAME. Actually, a commercial game that uses GZDoom GPL does just that. Even if there are no zandronum wads like that yet, they might appear. Going to the previous map is more reliable, as the map is guaranteed to exist. |
(0017499) WaTaKiD (updater) 2017-05-01 19:06 |
i decided to check out 'https://zdoom.org/wiki/IWAD [^]' and it seems the adventures of square is the only iwad here that doesnt use map01 or e1m1, but instead e1a1, however zan 3.0 cant load it, but zan 4.0 can |
(0017500) Torr Samaho (administrator) 2017-05-01 19:08 |
The problem is that by the time the server has errored out, it's not so clear anymore what the previous map was. Thus, I followed what the ZDoom code used in 3.0 does to initialize "startmap". A different possibility would be to use AllEpisodes[0].mEpisodeMap. |
(0017569) Torr Samaho (administrator) 2017-05-07 10:35 |
SERVER_ErrorCleanup now more carefully looks for a valid map to restart the game and exits the server if no map can be found. I think this should be sufficiently robust now. |
(0017575) WaTaKiD (updater) 2017-05-07 12:58 |
this build contains the new fix: 'https://www.dropbox.com/s/6x8dmuya1y7qm8r/zandronum-3.0-r170507-1032-c5ac7b4-windows.zip?dl=0 [^]' |
(0017681) Ru5tK1ng (updater) 2017-05-14 16:32 |
Tested this with r170513 and the server successfully restarted to map01, e1m1 and e1a1 after pushing the switches in the testwad online. |
This issue is already marked as resolved. If you feel that is not the case, please reopen it and explain why. |
|
Supporters: | Korshun |
Opponents: | No one explicitly opposes this issue yet. |
Issue History | |||
Date Modified | Username | Field | Change |
2017-04-30 11:13 | Korshun | New Issue | |
2017-04-30 11:13 | Korshun | File Added: exittest.wad | |
2017-04-30 15:42 | Torr Samaho | Note Added: 0017448 | |
2017-04-30 15:42 | Torr Samaho | Assigned To | => Torr Samaho |
2017-04-30 15:42 | Torr Samaho | Status | new => feedback |
2017-04-30 16:08 | Korshun | Note Added: 0017449 | |
2017-04-30 16:08 | Korshun | Status | feedback => assigned |
2017-05-01 17:25 | Torr Samaho | Note Added: 0017481 | |
2017-05-01 17:25 | Torr Samaho | Status | assigned => needs testing |
2017-05-01 17:25 | Torr Samaho | Target Version | => 3.0 |
2017-05-01 18:00 | WaTaKiD | Note Added: 0017486 | |
2017-05-01 18:08 | Korshun | Note Added: 0017487 | |
2017-05-01 18:41 | Torr Samaho | Note Added: 0017493 | |
2017-05-01 18:50 | Korshun | Note Added: 0017495 | |
2017-05-01 18:51 | Korshun | Note Edited: 0017495 | View Revisions |
2017-05-01 19:06 | WaTaKiD | Note Added: 0017499 | |
2017-05-01 19:08 | Torr Samaho | Note Added: 0017500 | |
2017-05-07 10:35 | Torr Samaho | Note Added: 0017569 | |
2017-05-07 12:58 | WaTaKiD | Note Added: 0017575 | |
2017-05-14 16:32 | Ru5tK1ng | Note Added: 0017681 | |
2017-05-14 16:32 | Ru5tK1ng | Status | needs testing => resolved |
2017-05-14 16:32 | Ru5tK1ng | Resolution | open => fixed |
2017-05-14 16:32 | Ru5tK1ng | Fixed in Version | => 3.0 |
2018-09-30 21:45 | Blzut3 | Status | resolved => closed |
Copyright © 2000 - 2024 MantisBT Team |