Jump to content

Server auto restarting at the wrong times...


dzrealkiller

Recommended Posts

Hi, I am having an Issue on a server. I have set up auto restart In the scheduled tasks to restart the server every 4 hours.

 

The restarts should happen at:

4am, 8am, 12am, 4pm, 8pm, 12pm. (UK Times)

 

The strange thing is the server seems to have a hidden auto restart file some where, the times the server restarts is... (Red means It should not restart)

 

3am, 4am, 7am, 8am, 11am, 12am, 3pm, 4pm, 7pm, 8pm, 11pm, 12pm. (UK Times)

 

The server seems to restart on a 3 hour cycle and a 4 hour cycle.

 

The server is in France so below I will list the France times of the restarts.

 

4am, 5am, 8am, 9am, 12am, 1am, 4pm, 5pm, 8pm, 9pm, 12pm, 1pm. (France Times)

 

The way I see it is Either there is a file somewhere calling the extra restarts or... The server is clashing with time zones and restarting for both of the time zones, There is only 1 hour between UK and France so that's why I think it could be a time zone clash or something like that. I have spoken to the hosters and my server is the only one this is happening on and we have all worked together for the last 48 hours trying our hardest to figure this out but have found no solution. Would anyone happen to know what could cause this?

Link to comment
Share on other sites

"change the restart settings to compensate for the difference" You mean change the restarts to every 3 hours not 4 right? Not sure I want to do that really I would like to be able to have it restart when I like not forced to stay with a 3 hour restart, That's not really a way to fix the problem that's just flowing with the issue.

 

As for changing the GMT setting I'm not sure where to do this I should properly ask my host about that one, I'm not sure if that would mess with anything else.

Link to comment
Share on other sites

I have a scheduler.xml with some things In if somebody could take a look that would be great.

 

scheduler.xml: https://www.dropbox.com/s/s2qymol8x1mfw7b/Scheduler.xml?dl=0

 

These are just restart warnings for the 4 hour restart but guess they don't matter at the moment as there is no 4 hour schedules any more but its restarting on its own every 3 hours its very strange.

Link to comment
Share on other sites

I have checked with them, They have spotted errors in the RPT and that's possibly causing crashes (not restarting) but I think its strange how there at set times all day every day... I still believe its restarting. They are dedicated to get this fixed and they will be jumping on the server around the time it happens to see if it still happens as they may have fixed the errors. Once I know if its fixed or not I will post back here to what the cause was in case somebody finds themselves with a similar issue. If it does not fix it I will need help looking into things more or advice on what other things could do it.

 

They have no auto restart built into there system except the scheduler one that we have disabled on my server. Disabling it made no difference the server crash/restart without the schedule.

 

Never had anything like this before at specific times all day every day its on time and never a minute late or slow.... very weird.

Link to comment
Share on other sites

Based of the scheduler.xml it looks like swiftnode do not automatically restart the server via the BEC scheduler. I assume that they use a scheduled task alongside BEC to do this.

 

    <!-- this job will start at 23:59:00 (11:59 pm) and will only run once a day at set time -->
    <job id='0'>
        <day>1,2,3,4,5,6,7</day>
        <start>000400</start>
        <runtime>000000</runtime>
        <loop>0</loop>
        <cmd>say -1 Testing In Game Warning System!</cmd>
    </job>

This job will only run after 4 minutes of server running time:

<start>HHMMSS</start>

I suggest you use real time like so:

<start>HH:MM:SS</start>

For example this job will run at 1pm everyday:

    <job id='0'>
        <day>1,2,3,4,5,6,7</day>
        <start>13:00:00</start>
        <runtime>000000</runtime>
        <loop>0</loop>
        <cmd>say -1 Testing In Game Warning System!</cmd>
    </job>

this way if you manually restart your server, your restart times are not out of sync with your scheduler.

Link to comment
Share on other sites

Yeah I am already planning this :)

<!-- this job will start at 23:59:00 (11:59 pm) and will only run once a day at set time -->
<job id='0'>
<day>1,2,3,4,5,6,7</day>
<start>000400</start>
<runtime>000000</runtime>
<loop>0</loop>
<cmd>say -1 Testing In Game Warning System!</cmd>
</job>

This code Is literally just to test I had BEC working :) Once we get my server running correctly I can then adjust all of these for the exact reason you mentioned "if you manually restart your server, your restart times are not out of sync with your scheduler

 

Yeah we use the Schedule Tasks to manage restarts, For me this just seems easier :) I may learn about how to run the restart via BEC I guess it would be better that way I can restart at 11:05:20 for an example If I wished :) I just like how fast and easy to setup restart are using the Schedule tasks :)

Link to comment
Share on other sites

Yeah I am already planning this :)

<!-- this job will start at 23:59:00 (11:59 pm) and will only run once a day at set time -->
<job id='0'>
<day>1,2,3,4,5,6,7</day>
<start>000400</start>
<runtime>000000</runtime>
<loop>0</loop>
<cmd>say -1 Testing In Game Warning System!</cmd>
</job>

This code Is literally just to test I had BEC working :) Once we get my server running correctly I can then adjust all of these for the exact reason you mentioned "if you manually restart your server, your restart times are not out of sync with your scheduler

 

Yeah we use the Schedule Tasks to manage restarts, For me this just seems easier :) I may learn about how to run the restart via BEC I guess it would be better that way I can restart at 11:05:20 for an example If I wished :) I just like how fast and easy to setup restart are using the Schedule tasks :)

 

You could use an external batch files to handle the restarts like we do.

 

	<!-- Restart Server -->
	<job id='5'>
		<day>1,2,3,4,5,6,7</day>
		<start>020000</start>
		<runtime>000000</runtime>
		<loop>0</loop>
		<cmd>D:\TCAFiles\Users\test\100\RESTART_SERVER.bat</cmd>
	</job>

This way your restarts can never be out of sync with the restart messages. Ask swiftnode if they can do something like that for you. If not let me know and I'll supply you with a batch file

Link to comment
Share on other sites

This is all I have, Its strange tho It crashes every...

3am, 7am, 11am, 3pm, 7pm11pm, (UK Times)

4am8am, 12am, 4pm, 8pm, 12pm,(France Times)

Without fail.. The weird thing is we have reinstalled the whole server fresh.. and it still do it. Its only my server tho..

 

The below crash is 4pm France Time.. My server Is In France... just thought this little comment could help.. I dunno :(

 

Mods: @EpochHive
Distribution: 0
Version 1.36.128579
Fault time: 2014/12/10 04:00:14
Fault address:  65798C1F 01:00017C1F C:\TCAFiles\Users\dzrealkiller\71\@EpochHive\EpochServer.dll
file:     epoch (__cur_mp)
world:    Altis
Prev. code bytes: 83 C4 04 85 C0 74 09 50 E8 CF 75 02 00 83 C4 04
Fault code bytes: 83 3F 06 75 35 C6 03 00 8B 57 14 80 3A 00 75 0D

Registers:
EAX:00000000 EBX:00B6C22C
ECX:3DBBB2CC EDX:3DBBB2E8
ESI:00B6C230 EDI:00000000
CS:EIP:0023:65798C1F
SS:ESP:002B:00B6C104  EBP:00B6C210
DS:002B  ES:002B  FS:0053  GS:002B
Flags:00210246
=======================================================
note: Minidump has been generated into the file SC\arma3server_2014-12-10_01-45-22.mdmp

Link to comment
Share on other sites

This is all I have, Its strange tho It crashes every...

3am, 7am, 11am, 3pm, 7pm11pm, (UK Times)

4am8am, 12am, 4pm, 8pm, 12pm,(France Times)

Without fail.. The weird thing is we have reinstalled the whole server fresh.. and it still do it. Its only my server tho..

 

The below crash is 4pm France Time.. My server Is In France... just thought this little comment could help.. I dunno :(

 

Mods: @EpochHive

Distribution: 0

Version 1.36.128579

Fault time: 2014/12/10 04:00:14

Fault address:  65798C1F 01:00017C1F C:\TCAFiles\Users\dzrealkiller\71\@EpochHive\EpochServer.dll

file:     epoch (__cur_mp)

world:    Altis

Prev. code bytes: 83 C4 04 85 C0 74 09 50 E8 CF 75 02 00 83 C4 04

Fault code bytes: 83 3F 06 75 35 C6 03 00 8B 57 14 80 3A 00 75 0D

Registers:

EAX:00000000 EBX:00B6C22C

ECX:3DBBB2CC EDX:3DBBB2E8

ESI:00B6C230 EDI:00000000

CS:EIP:0023:65798C1F

SS:ESP:002B:00B6C104  EBP:00B6C210

DS:002B  ES:002B  FS:0053  GS:002B

Flags:00210246

=======================================================

note: Minidump has been generated into the file SC\arma3server_2014-12-10_01-45-22.mdmp

 

Never seen the EpochServer.dll be the fault before.

Can you send me the contents of your epochserver.ini (Remember to remove and passwords before posting).

Also if you have access, your redis.conf

Link to comment
Share on other sites

Is it possible that there is another server on their network with the instance ID: Developer Test.

 

Also by looking at the error you sent me: 

Fault address:  65798C1F 01:00017C1F C:\TCAFiles\Users\dzrealkiller\71\@EpochHive\EpochServer.dll

And the BattlEye path in your EpochServer.ini:

BattlEyePath = C:\TCAFiles\Users\dzrealkiller\77\Battleye

One has a service ID of 71 and the other of 77, is this correct?

Link to comment
Share on other sites

No, that's his new server we just reinstalled after a quite some time fighting the server faults.

 

We don't use any type of auto restarting by default. (clients normally configure the tcadmin scheduler to do restarts)

 

I'm waiting on an update from vbawol about the address of that fault, as it seems to be the only lead we have.

Link to comment
Share on other sites

Is it possible that there is another server on their network with the instance ID: Developer Test.

 

Also by looking at the error you sent me: 

Fault address:  65798C1F 01:00017C1F C:\TCAFiles\Users\dzrealkiller\71\@EpochHive\EpochServer.dll

And the BattlEye path in your EpochServer.ini:

BattlEyePath = C:\TCAFiles\Users\dzrealkiller\77\Battleye

One has a service ID of 71 and the other of 77, is this correct?

 

Hi,

 

 

Service 71 was the old service that produced the Fault address: "65798C1F 01:00017C1F C:\TCAFiles\Users\dzrealkiller\71\@EpochHive\EpochServer.dll" and the server that was producing these unexpected crashes at certain times of the day which is very odd.

 

 

We have spent most of the week looking through errors and logs and we even did a file comparing process comparing each config file with a full working server for syntax errors, none was found. Service number 77 is the new service we have provided him.

 

 

 

 

Regards,

 

Liam Benfield

Swifnode.net

Link to comment
Share on other sites

From looking at your redis.conf file and your EpochServer.ini, I see the database port is

23423

But in your redis.conf you're not specifying the port therefor it will run on the default port of

6379

Try changing the redis.conf file to specify the port to match what the server is trying to connect to:

bind 37.187.158.76
port 23423
maxmemory 4gb
save 900 1
save 300 10
save 60 1000
requirepass PASSWORDHERE
Link to comment
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
  • Advertisement
×
×
  • Create New...