Dedicated server

[suchey 0]

Since 1.42 Ive been having some weird problems with the memory usage of the OFP stand-alone dedicated server...I have kept a rudimentary log of sorts...following is what I recorded:

(all recorded in K under taskmgr)

March 10 (using 1.46)

18:22- 81608 server started

18:50- 149260 server filled

21: 05- 150500

23:59- 168040

March11

0:40- 170156

6:26-Server rebooted after crash

8: 06-145228

9:12-155076

11:40- 158180

12: 00- <<1.47 installed reboot>> 113065

13:24- 136424

14:56- 145156

15:39- 145232

18:50- 149260

19: 02- 149260

21: 05- 150500

23:59- 168040

March 12

8: 09- 194408

12:14- 201308

<<1.48 server patch installed, reboot>> 84852

16:16 138912

18: 00 144692

20: 00 155260

21: 00 <<reboot>>54000

22:22- 139780

23:47- 146188

March 13

7:12- 152436

16:15- 201344

20: 00- 201636

It seems that memory is being eaten somewhere...upon reboot the memory resets but the longer the server runs the more memory seems to be used...these above was recorded using the same maps with the same rotation under a full comliment of players.  I can run a mission right after a reboot and get between 135000-145000K...but that same mission 24 hours later will be at 201000K (including default BIS missions)...so I dont think it is specific to any missions. I have also noticed that using the #monitor command, I get lower FPS scores on the same mission if the server has been up for a while as opposed to running that same mission within a couple hours of a reboot. Is this a problem on my end and if so how can I fix it? I didnt have this with 1.42 and i havent made any hardware or software changes other than the install of the new stand alone server patches. Thanks in advance for your help. On the bright side, the server hasnt crashed since 1.47.

[suma 8]

To make sure I understand your post (some english sentences are not clear enough for me):

The problem was not present in 1.42 and is present in 1.48 on your server.

[Espectro (DayZ) 0]

It wasnt present in 1.42, but was present in 1.46-1.48.

[ALDEGA 0]

I think he means the problem has been there in version 1.42 and newer although his logs only show data since 1.46.

[suma 8]

It is nice to read what you guys think about what Suchey meant, but I think I will rather wait for his answer.

[Espectro (DayZ) 0]

"I didnt have this with 1.42 and i havent made any hardware or software changes other than the install of the new stand alone server patches."

Its pretty obvious what he means.

[suma 8]

</span><table border="0" align="center" width="95%" cellpadding="3" cellspacing="1"><tr><td>Quote (ALDEGA @ Mar. 14 2002,15:59)</td></tr><tr><td id="QUOTE">I think he means the problem has been there in version 1.42 and newer although his logs only show data since 1.46.<span id='postcolor'>

Looks like it is not obvious - at least Algeda understood it differently from me.

[Rage_Frost 0]

oh this is funny LOL but I take it like he did not have a problem with ver 1.42 with memory usage but after that with the newest ver's he did.

At least I know are server didn't before with Ver 1.42.

Also it's just not the server but are side too on are computers with the memory usage after ver 1.42

Just my 2 cents

[suchey 0]

I did not have this problem with 1.42...I have only experienced this with the patches which came AFTER 1.42

Thanks for the response so far.

[suchey 0]

same problem is happening with 1.49 it seems. I will keep an eye on it and repost the usage as above.

[Enron Exec 0]

I think the new patches where done in haste to resolve the crashes. The growing consumption of memory I think is just poor or the absence of "garbage collection." In other words, they simply did not take care of deallocating memory used by the new patches. I would just email your findings to whoever is in charge of the patches and hope they do a better job in the next patch. If only everyone took as much pride in their work as the ppl at MTCP !!!

Tell them to practise the 5 P's of software engineering.