Arma 2: Operation Arrowhead - Linux Standalone Server 1.56 BETA

[Overlord 0]

So how is the current status of Linux 1.56?

How does it perform compared to 1.54? Exept PMC (Lite).

[=-A*C-=NicO 10]

The same as 1.54, even sometimes worst.

Tested both Vanilla and with ACE.

Don't expect anything new about performances. Still stuck on 1 core -> cpu limited.

[Overlord 0]

=NicO;1807490']The same as 1.54' date=' even sometimes worst.

Tested both Vanilla and with ACE.[/quote']

So there's no major benefits to upgrade? Still no fix for MP weapon selection in briefing?

Don't expect anything new about performances. Still stuck on 1 core -> cpu limited.

Do you have 1 core server or is it a setup problem? I run 1.54 on dualcore, server's more than willing to use both.

[=-A*C-=NicO 10]

So there's no major benefits to upgrade? Still no fix for MP weapon selection in briefing?

Don't know what you"re talking about, but is this a client side or server side problem? From what I can tell you, client side performances are better in 1.54 than in 1.55/56. So the overall experience is best on 1.54, but you will probably run into trouble if you use the latest ACE for example (if you're on vanilla, stay with 1.54).

Do you have 1 core server or is it a setup problem? I run 1.54 on dualcore, server's more than willing to use both.

Yep true, we own a corei5 750 (4 cores), and all of them are used. But the main thread of the game uses 100% of 1 core, while the others are quite unused. We ran several test : Windows dedi vs Linux dedi, and the multithreading on Linux is awful compared to Windows. That's a known issue that will probably not be fixed. The point is that we don't want to use an MS box for several reasons, so we're stuck with a bad dedi...

Take a look at an ingame typical load :

[Dwarden 1125]

please read the bold text

http://forums.bistudio.com/showpost.php?p=1797139&postcount=35

ArmA 2 can use up to 31 cores in theory, but experiments have shown that with most scenes the gain above 4 cores is small and above 8 cores unmeasurable.

The explanation is Amdahl's law -

only parts of the application is using all cores.

See Real Virtuality Going Multicore blog.

---------- Post added at 13:58 ---------- Previous post was at 13:54 ----------

In build 76122 and newer the default for dualcores will be changed to -exThreads=3 based on user feedback.

We have also changed the cpu core detection, therefore depending on how many logical cpus are present, default -cpuCount values will be as follows:

1 1

...

6 6

7 7

8 4

9 4

10 5

11 5

12 6

13 6

14 7

....

Some day hopefully we will find a time to provide a proper HT detection, but until then I think the above provides quite reasonable default settings.

so dont worry

if there is possibility or way how to improve engine multicore performance it will be tried and implement if successful

[=-A*C-=NicO 10]

I remember I heard that it was an issue with multithreading librairies, do you confirm? If this is it - it will be hard to solve, right?

[mibf 0]

I think, it is a memory leak: http://goo.gl/ZJqJu

Server build 1.56.76188

Debian 2.6.32-5-amd64

I added a ticket on dev-heaven

Edited December 8, 2010 by MiBF

[Overlord 0]

OK, thanks guys, also to Dwarden/ Suma for a reasonable explanation of how cores is used. Make sense in a way too... :)

[byteslam 10]

I think, it is a memory leak: http://goo.gl/ZJqJu

Server build 1.56.76188

Debian 2.6.32-5-amd64

I added a ticket on dev-heaven

However, the server is unstable, more unstable then the previous releases

kernel: [1453211.193560] server[2640]: segfault at a38dea00 ip 817e75b sp ff9291f0 error 4 in server[8048000+100b000]

This happens if a player with an ver 1.04 tries to connect... even if signature check is on.

There are some Trolls who only connecting to full servers to crash them...

Edited December 10, 2010 by byteslam

[mibf 0]

kernel: [1453211.193560] server[2640]: segfault at a38dea00 ip 817e75b sp ff9291f0 error 4 in server[8048000+100b000]

if it helps, I'll add more segfaults:

Dec  3 13:06:37 g13 kernel: [587359.773295] server[4847]: segfault at 11 ip 00000000088b6103 sp 00000000ff81f380 error 4 in server[8048000+100b000]
Dec  3 16:07:29 g13 kernel: [598207.449113] server[7599]: segfault at 3c ip 00000000085045c4 sp 00000000ffd41a90 error 4 in server[8048000+100b000]
Dec  4 23:06:26 g13 kernel: [709677.923239] server[7530]: segfault at c ip 0000000008f666cd sp 00000000f7392370 error 4 in server[8048000+100b000]
Dec  4 23:07:27 g13 kernel: [709738.957386] server[10472]: segfault at c ip 0000000008f666cd sp 00000000f73c3370 error 4 in server[8048000+100b000]
Dec  4 23:07:41 g13 kernel: [709752.101370] server[10546]: segfault at 0 ip (null) sp 00000000f76ea1dc error 14 in server[8048000+100b000]
Dec  4 23:08:09 g13 kernel: [709780.863954] server[10577]: segfault at c ip 0000000008f666cd sp 00000000f73f6370 error 4 in server[8048000+100b000]
Dec  4 23:29:57 g13 kernel: [711087.718161] server[10620]: segfault at c ip 0000000008f666cd sp 00000000f7410370 error 4 in server[8048000+100b000]
Dec  5 07:33:47 g13 kernel: [740087.862639] server[20107]: segfault at 0 ip 000000000835af02 sp 00000000ffff4360 error 6 in server[8048000+100b000]
Dec  5 14:48:05 g13 kernel: [766134.053571] server[11979]: segfault at 0 ip 0000000008a02804 sp 00000000ffc57d10 error 4 in server[8048000+100b000]
Dec  5 15:41:52 g13 kernel: [769359.644545] server[20070]: segfault at 3c ip 00000000085045c4 sp 00000000ffee8580 error 4 in server[8048000+100b000]
Dec  6 00:00:24 g13 kernel: [799258.763394] server[20162]: segfault at 0 ip 0000000008a6b2e3 sp 00000000ffaccd10 error 6 in server[8048000+100b000]
Dec  8 11:56:36 g13 kernel: [1014883.413278] server[9998]: segfault at 3c ip 00000000085045c4 sp 00000000ffd10700 error 4 in server[8048000+100b000]
Dec  9 02:04:26 g13 kernel: [1065730.663182] server[13989]: segfault at c ip 0000000008f666cd sp 00000000f737e370 error 4 in server[8048000+100b000]
Dec  9 20:37:02 g13 kernel: [1132440.498626] server[14338]: segfault at 9 ip 000000000862d93b sp 00000000fffdc8c0 error 6 in server[8048000+100b000]
Dec 11 04:34:32 g13 kernel: [1247421.819694] server[5547]: segfault at 0 ip 00000000081c0690 sp 00000000ffe00680 error 6 in server[8048000+100b000]
Dec 11 20:35:22 g13 kernel: [1305029.112209] server[5621]: segfault at 6c ip 00000000082f4de0 sp 00000000ffd20cc4 error 4 in server[8048000+100b000]

and I have a few big dumps (about 700mb) - I can upload it in some place if it needs.

Edited December 12, 2010 by MiBF

[=-A*C-=NicO 10]

please read the bold text

http://forums.bistudio.com/showpost.php?p=1797139&postcount=35

so dont worry

if there is possibility or way how to improve engine multicore performance it will be tried and implement if successful

How can you explain that multithreading is really better on Windows?

Amdahl's law is the same on both platform...

[Sira 10]

Does anyone know how to fix this error?

Warning Message: No challenge value was received from the master server.
GameSpy QR2 error: 5, No challenge value was received from the master server.

This posts mentioned the same error:

http://forums.bistudio.com/showthread.php?t=74044

But has no solution.

reportingIP = "arma2oapc.master.gamespy.com";

EDIT: I got it working twice but the server dies as soon as I connect, and above problem still reoccurs randomly.

Any help is appreciated, thx.

Edited December 18, 2010 by Sira

[Overlord 0]

Anyone know if there is any difference (performance/ stability etc.) between 32 and 64 bits Linux OS or should I assume 32 runs better (as in M$)? RAM ain't an issue here.

[Sira 10]

Anyone know if there is any difference (performance/ stability etc.) between 32 and 64 bits Linux OS or should I assume 32 runs better (as in M$)? RAM ain't an issue here.

64bit is always better but also more prone to cause problems eventually with such a poorly supported software like the arma2 server.

Edited December 18, 2010 by Sira

[visceralsyn 10]

64bit OS's use ram more efficiently than 32bit OSs, as well as can address more, thats not OS specific. Which makes 64bit OS's perform and scale better than 32bit OS's.

ArmA² is not poorly supported, its poorly coded.

...Syn...

Edited December 18, 2010 by VisceralSyn
corrections