I suspect it's either RHS or a cup terrains issue. I loaded British Steel up last night. Played for 3 hours straight, player exited, left running all night, came back, played 2 hours on same session this morning. No issues. That mission uses 3CB, AllinArma, CAF aggressors, ACE, ALiVE, ARP objects.

But I load up Thor's Hammer, using a pretty simple setup of RHS,, Chernaus via CUP, ACE, and ALiVE and it desyncs within 60 minutes every time.

So by process of elimination, I think it's either CUP terrains or RHS or both.

It still may be worth a shot PMing those OOM rpts to some of the Devs on BIS forums. They seem eager to receive them; at least they say so publicly. It couldn't hurt.

If Chernerus is AiA compatible it may be useful seeing if a similar mission has the same problems just so you can rule in/out the map pack and its objects.

They only want the performance version RPTs. My server uses stable branch.

If it started happening after the RHS 4.1 update then it's probably that. I doubt it's CUP Terrains since, as far as I know, mostly config work whereas RHS has quite a bit of scripted components.

Yeah that would make sense. @AUTigerGrad are you having issues with missions that use RHS as a dependency but not as the actual faction (I.E. A Bright Morning using @ISC?)?

@Crapachi Had a mission tonight go about 4 hours, and without any sign of desync. But I do recall a time a few versions back when there indeed was desync after just about an hour or so. Same map, mostly the same mods, and similar ALiVE config.

The only major thing I can think of that is different, is that I'm not bothering with a Headless Client this time around.

That might be it. We use Headless Client on our server.

Ok, so I've been experimenting/stress testing on the dedicated server and I have found some interesting results.

I loaded a mission using only the mods listed above on Stratis and ran it....no desync for hours and hours.

So I created a new mission on Al Rayak...only adding CUP Terrains and Al Rayak map to the mod list....and I get this RPT spamming over and over and over and over...causing a desync and a log file of almost 18mb after 10 hours or so.

Error in expression <tive = _profile select 2 select 1;

if!(_active) then {
_position = _profile sel>
Error position: <_active) then {
_position = _profile sel>
Error Undefined variable in expression: _active
Error in expression <active = _profile select 2 select 1;

if!(_active) then {
_position = _profile s>
Error position: <!(_active) then {
_position = _profile s>
Error !: Type Array, expected Bool
Error in expression <tive = _profile select 2 select 1;

if!(_active) then {
_position = _profile sel>
Error position: <_active) then {
_position = _profile sel>
Error Undefined variable in expression: _active
Error in expression <active = _profile select 2 select 1;

if!(_active) then {
_position = _profile s>
Error position: <!(_active) then {
_position = _profile s>
Error !: Type Array, expected Bool
Error in expression <active = _profile select 2 select 1;

if!(_active) then {
_position = _profile s>
Error position: <!(_active) then {
_position = _profile s>
Error !: Type Array, expected Bool
Error in expression <tive = _profile select 2 select 1;

if!(_active) then {
_position = _profile sel>
Error position: <_active) then {
_position = _profile sel>
Error Undefined variable in expression: _active

This looks like an ALiVE error to me. Is that correct? Anyone else seen this? It's the major bugaboo for my issue right now it appears.

If AuTigerGrad is seeing something wonky, there has to be an error dump somewhere causing the problem. He's been able to run his British Steel mission (with AiA) for days on end without any problems at all. And this was recently.

I wonder now that AiA is obsolete and he's likely changed it to CUP if that has changed. It would seem he's implying there's an ALiVE+CUP conflict going on but that would be tough to determine without an ALiVE, CBA and CUP only test.

Ok, so I did the exact same mission, swapping RHS units for Vanilla units.

The error was gone.

Here's the RPT of the mission with RHS https://www.dropbox.com/s/awpvlbwgmbzqhe7/arma3server_2016-06-01_22-54-47%20%281%29.rpt?dl=0

Here's the RPT of the mission without RHS https://www.dropbox.com/s/dz8e6ytkpfpks6p/arma3server_2016-06-02_14-56-43%20%283%29.rpt?dl=0

That one error spammed over 18mb of rpt spam. Every thing else was exactly the same other than I swapped the Vanilla units for RHS units.

Now when I tested with RHS on Stratis (using no CUP Terrains) I didn't get an error. And when I used CUP Terrains without using RHS on this latest mission...I didn't get the error.

So....from what I can gather...it's some kind of conflict between using all 3 together (ALiVE + CUP Terrains + RHS) that is causing that error to spam the RPT.

I feel like Sherlock Holmes right now.

So a quick update:

I deleted all ALiVE Modules from my missions creating the error, then re placed each module, saved...and loaded on the server.

It appears to have fixed the error.

Apparently, the latest ALiVE release was causing an issue on my older build missions if I didn't re place the modules.

So far so good which makes me very happy.

@AUTigerGrad, for what it's worth, those errors are related to the spawning of inactive groups when you get close enough, originating from sys_profile/profileSpawner_v2_1.fsm line 220.

Can't really figure out what's happening without a mission that reproduces those errors, so if you can provide one with the smallest mod set possible (CUP + RHS is fine for me). :)

@marceldev89 @AUTigerGrad, for what it's worth, those errors are related to the spawning of inactive groups when you get close enough, originating from sys_profile/profileSpawner_v2_1.fsm line 220.

Can't really figure out what's happening without a mission that reproduces those errors, so if you can provide one with the smallest mod set possible (CUP + RHS is fine for me). :)

Just got an Out of Memory/"No message received"/Desync crash after 90 minutes of gameplay. The error was gone, but it still desynced/crashed. RPT was 2.6 mb after 90 minutes.

Rather than post another RPT, I'm gonna post the mission PBO.

Marcel, or anyone else, who would like try it, feel free, to see if you get the same crash/desync.

https://www.dropbox.com/s/cybonqb2tig52i2/Operation_Gathering_Storm.pja310.pbo?dl=0

Marcel is this what you are referring to?

ALiVE SYS PROFILE Warning: ALIVE_fnc_waypointsToProfileWaypoints has wrong inputs!
Error in expression <"_value") then { nil } else { _value },
_hash select 3] call (uiNamespace getVar>
Error position: <_hash select 3] call (uiNamespace getVar>
Error Undefined variable in expression: _hash
File x\alive\addons\x_lib\functions\data\fnc_hashSet.sqf, line 42
Error in expression <l ALIVE_fnc_hashGet;

if(_profileID in (_profilesActiveSide select 1)) then {
[_>
Error position: <_profilesActiveSide select 1)) then {
[_>
Error Undefined variable in expression: _profilesactiveside
File x\alive\addons\sys_profile\fnc_profileHandler.sqf, line 770
Error in expression <then {
_profileType = _profile select 2 select 5;
_profileID = _profile select 2>
Error position: <select 5;
_profileID = _profile select 2>

That's in the latest RPT as well.

I was refering to these errors:

if!(_active) then {
_position = _profile s>
Error position: <!(_active) then {
_position = _profile s>
Error !: Type Array, expected Bool
Error in expression <active = _profile select 2 select 1;

You do seem to get a lot of errors, judging by the RPTs you've been posting.

I'll check out the mission in a bit.

Sorry, not going to test that mission, it has way too many dependencies; Al Rayak, ACE, CUP, 3den objects and RHS. Could be anything causing the crash with that setup.

@marceldev89 Sorry, not going to test that mission, it has way too many dependencies; Al Rayak, ACE, CUP, 3den objects and RHS. Could be anything causing the crash with that setup.

That's a pretty standard modset. I've already tested vanilla and there was no issue. But being able to play Vanilla units on a vanilla map isn't my goal.

No worries.

Yeah, I got all of my issues fixed. It was my server that was the culprit. Got a new box and now I'm good to go.