• Tracker

    Spaces
    Browse
    Statistics
  • /
  • /Stable Feedback
  • /DZG-2052
Back
1

[Xbox][Nasdara] Repeated KillPlayer/SetHealth crashes and empty building persistence files — missing base walls/doors

BugCrashError
1

State

Open
Issue key
DZG-2052
Access
Public
Space
DayZ
Project
Stable Feedback
Creator
xmgsneakyg
Created
Oct 26, 2026
Views
26
Platform
Xbox
What happened?

PLATFORM AND SCOPE

ABX operates Nitrado-hosted DayZ Xbox community servers: Nasdara (dayzOffline.nasdara) and Chernarus. This report's confirmed crash and storage evidence concerns Nasdara. Earlier Chernarus building losses are reported separately below. The attached Xbox RPT excerpt confirms version 1.30.164200.

SUMMARY AND IMPACT

Repeated interruptions and disappearing player-built structures have made normal play unreliable. Our community has roughly 1,000 Discord members; this is not a claim of 1,000 concurrent server slots. Players have lost substantial building effort and confidence, and people are leaving. We understand occasional crashes, but one confirmed process ran only 4 minutes 31 seconds before another fatal exception. We urgently need engineering investigation and a safe recovery path.

CONFIRMED NASDARA CRASHES

Nine fatal exceptions were identified in retained evidence. These are not every crash since server creation, and ordinary restarts are not counted as crashes. The recurring recorded call path is KillPlayer/SetHealth with access violation. UK local times:

23 October 2026 (BST, UTC+1): 02:22:29; 02:52:57.

25 October 2026 (GMT, UTC): 15:24:08; 15:45:34; 20:21:09; 20:49:22; 20:56:16; 23:41:35.

26 October 2026 (GMT, UTC): 00:21:09.

The 20:51:45 process start to 20:56:16 exception is 4m31s. The 20:49:22 and 20:56:16 exceptions are 6m54s apart. See attached crash index for original filenames.

Players/owners have associated some interruptions with suicide/death actions. The call path warrants investigation, but we have not demonstrated a deterministic reproduction or identified a player as responsible. We do not claim every interruption had this cause.

CONFIRMED BUILDING STORAGE FAILURE SEQUENCE

All following times are UK GMT. Source RPT clock is one hour ahead, so e.g. RPT 00:45 on 26 October corresponds to 23:45 GMT on 25 October. Raw excerpt filenames and line numbers are retained in the attachment.

25 October 23:45:20: building_001.bin restored with 20,480 items; main and fallback generations reported valid.

25 October 23:54:19-21: repeated failures to write building_001.bin, explicitly saying failed to open the file.

25 October 23:58:45-46: building_001.001 and building_001.002 empty and invalid; main restored with 5,120 items.

26 October 00:20:52-54: repeated failure to open building_001.001 for writing.

26 October 00:21:09: confirmed fatal exception.

26 October 00:25:09-10: main building_001.bin and both fallback generations empty/invalid; building_002.bin still reported valid.

These are storage record counts, not an exact count of missing walls. The logs establish write/read failures and empty files, but do not establish whether the underlying cause is engine logic, filesystem access, configuration, hosting or a combination, or whether the crash caused the persistence loss or vice versa.

PLAYER OBSERVATIONS AND OTHER SERVER

Players report the new brick walls, doors and related building elements disappearing after interruptions. Traditional fences/watchtowers/gates appear less affected so far; that is observational and not a guarantee.

Chernarus had earlier reports of disappearing builds. The retained Chernarus logs checked for 18-21 and 23-26 October did not establish the same fatal signature. We are not attributing Nasdara's confirmed signature or counts to Chernarus without evidence.

Digital code locks have also been reported unreliable, including previously working codes becoming unusable. That needs separate reproduction and triage; it is not established as the cause of the fatal crashes or storage loss.

REPRODUCTION STATUS

Observed during ordinary live community-server operation, followed by process restart and persistence loading. Exact deterministic reproduction remains unknown. Please investigate death/suicide processing around KillPlayer/SetHealth and storage writes preceding the failures. We have not deliberately reproduced destructive failures on the live servers.

HOSTING SUPPORT RESPONSE

Nitrado support acknowledged a high number of crashes, said the server had been moved to another machine, and referred access violations to Bohemia. They stated hardware resources were adequate. We have not independently verified their hardware conclusion, and the response did not explain the failed file opens or empty persistence generations. Console hosting restrictions prevent us obtaining all underlying player persistence files directly.

RELATED REPORT

https://report.bistudio.com/issues/DZG-1992 describes another operator's PS4 Nasdara crashes and persistence errors. It is useful for cross-platform triage, but its 27 crashes, fault address and startup-loop behaviour are not our evidence and must not be combined with our nine Xbox crashes. A common root cause is not yet established.

REQUESTED ENGINEERING ACTION

Correlate and symbolicate the Xbox crash evidence; identify the fault behind the repeated KillPlayer/SetHealth access violations.

2. Explain why building storage files fail to open for writing and how all fallback generations can become empty/invalid; determine whether crashes and storage loss share a cause.

3. Coordinate with Nitrado to preserve relevant crash dumps, RPTs and storage generations before rotation, and obtain the restricted persistence files through an appropriate private route.

4. Assess selective recovery of missing building records without rolling back unaffected player inventories, vehicles or traditional building progress.

5. Provide a tested workaround, affected-version information, and a tracked fix or investigation reference. Advise which additional configuration samples and diagnostics are needed through a private upload route.

6. Do not restart, wipe or restore our live servers without owner approval. Both servers must remain active; this technical report is not a cancellation request.

ATTACHMENTS

Nitrado-confirmed-crashes.txt: nine confirmed fatal exception timestamps and original crash filenames.

Nasdara-building-storage-evidence.txt: source RPT excerpts with filenames and line numbers.

Public evidence has been redacted to remove service IDs, host/account details and player identifiers. Supporting evidence and server identification are provided in private attachments, visible only to the report owner and moderators. No claim of a proven cheating exploit is included: unrelated device/security allegations require separate evidence and private security reporting.

ADDITIONAL ATTACHED CRASH EVIDENCE

The redacted crash_2026-10-25_16-26-31.log includes the fatal exception at 16:45:34 server time (15:45:34 UTC) and preceding EffectAreaLoader CreateZones warnings. Those configuration-related warnings are included for completeness; their causal relationship to the fatal exception is not established. Nasdara-crash-25-Oct-2341.txt includes the native stack, RPT version and anonymised activity around the later fatal event. We use customised mission configuration/custom builds; this is an Xbox community server, not a PC Workshop-mod server.

Reproduction steps

No deterministic reproduction is confirmed. Observed sequence: (1) Operate the live Xbox Nasdara community server normally. (2) Player death/suicide actions occur; supplied crash evidence records SurvivorBase, SetHealth and EmoteManager KillPlayer/OnAnimationEvent. (3) A fatal access violation terminates the server. (4) On subsequent startup, inspect CE storage restoration: the supplied RPT sequence shows failed writes followed by empty/invalid building_001.bin and both fallback generations, accompanying reports of missing new walls/doors. These steps describe observations, not a proven trigger. We have not deliberately reproduced destructive failures on the live servers. Please advise a safe isolated reproduction procedure.

Exact build version (optional)
1.30.164200
Where did it happen?
Community
Is it modded?
No
Expected result
  • The dedicated server should remain stable during normal player death/emote processing. Player-built structures should persist through normal saves and restarts, with valid recoverable fallback generations if a write fails.

Public attachments4 files · visible to everyone

Nitrado_confirmed_crashes.txt

924 B

Nasdara_building_storage_evidence.txt

6.6 KB

Nasdara_crash_25_Oct_2341.txt

5.8 KB

crash_2026_10_25_16_26_31.log

51 KB

Reactions

Activity

    about 5 hours ago
  • xmgsneakyg
    created issueabout 5 hours ago
  • xmgsneakyg
    changedabout 4 hours ago
  • imonthefloorcommented
    about 2 hours ago
    imonthefloorcommented
    about 2 hours ago

    Following. ^ needs attention and assignment required.

  • xmgsneakygreplied to a comment
    about 2 hours ago
    xmgsneakygreplied to a comment
    about 2 hours ago

    What do you mean pal

  • imonthefloorcommented
    about 2 hours ago
    imonthefloorcommented
    about 2 hours ago

    They need to assign this report to someone asap hopefully before next patch. As id say more servers will start seeing crashes appearing , mine started roughly 4 days after release, same thing vanillia files. Ive done a wipe of dynamics bin as assuming it was causing corruption, just waiting another 24 hours if it crashes again..

  • xmgsneakygcommented
    about 2 hours ago
    xmgsneakygcommented
    about 2 hours ago

    They just need to start testing properly before releasing also nitrado and host should also test before they release to the public


    It seems that everyone in this process is happy to take money but not happy to fix the problem promptly


    Or do anything to mitigate the problem being a problem in the first place?


    I believe that all hosts should test updates before rolling them out but then again so should Bohemia

    But the fact that hosting providers simply just roll it out without testing it's terrible if they were to test and then return to Bohemia saying we are not selling your product as it is broken I am sure bohemia would do their upmost to fix it then but because they are already making the money, of people buying the servers it's too late and they don't need to do it promptly as they have already been paid


    I can't believe it's 2026 we have Superintelligence AI and these guys can't get a code lock to work I could successfully implement a code lock in a few hours time I think it's time Bohemia introduce AI to the process as the human process of their updates does not have a good track record

  • imonthefloorcommented
    about 2 hours ago
    imonthefloorcommented
    about 2 hours ago

    Honestly even just more access for console , i know its the cliche shit "give us init.c" but like atleast we can run some enforce to try and diagnose the crash, and take the investigation into our own hands . Nitrado just passes off here is all we can do, reset to vanilla files, change machine, or we can rollback you choose.

    - Litterally has access to run diagnostics but doesnt.

  • xmgsneakygcommented
    about 2 hours ago
    xmgsneakygcommented
    about 2 hours ago

    100% correct give us the access we need it's bad enough. We have people attacking us with BYOVD's at kernel level it was hard enough mitigating these people moving entire bases I've spent the last three months looking for a zero day driver to try and create something that could combat these people and now my time is simply consumed with tickets due to a broken update when they hadn't even fixed the kernel level attack issues currently flooding console service they are leaving the consumers to try and fix


    It's almost like they don't know what bring your own vulnerable driver means but then again they are struggling with code locks

You are not signed in. Please sign in to see more details and to reply.

State

Open
Issue key
DZG-2052
Access
Public
Space
DayZ
Project
Stable Feedback
Creator
xmgsneakyg
Created
Oct 26, 2026
Views
26