OVH Community, your new community space.

( help ) remove binding from fail safe IP'S


avizeke
05-07-2011, 18:30
i tried different kinds of servers to establish the problem here and other servers worked fine, its just source based ones that dont.

Yet, the primary ip already has Source based servers on so i think it is to do with 2 IP's connection to through the apache 80 port even though apache isnt on.

i am doubting that steam is the problem here, i am more to believe of gow its configured its just something somewere stopping this.

could it be that apache isnt installed and that it requires it to be installed in order for two ips to use the same port ?

Myatu
05-07-2011, 18:21
Quote Originally Posted by avizeke
this occured when someone tried connecting to the server. Firstly its on Centos, and apache isnt installed on this system as its not required as i aint running any web services on here.
The problem is with the remote (steam's website, not yours):

Code:
Apache/2.2.14 (Ubuntu) Server at crash.steampowered.com Port 80
And looks that whatever was being used (a proxy?) could not deal with the unexpected crash on the remote side.

avizeke
05-07-2011, 17:23
ok, i tested things directly in SSH, and it seems that the IP is fine and able to connect.

However there is still a issue i think with management of the software on the system i recieved this error.

PreMinidumpCallback: updating dump comment
Uploading dump (in-process) [proxy '']
/tmp/dumps/crash_20110705181450_1.dmp
success = yes
response:

500 Internal Server Error

Internal Server Error


The server encountered an internal error or
misconfiguration and was unable to complete
your request.


Please contact the server administrator,
webmaster@localhost and inform them of the time the error occurred,
and anything you might have done that may have
caused the error.


More information about this error may be available
in the server error log.




Apache/2.2.14 (Ubuntu) Server at crash.steampowered.com Port 80



Add "-debug" to the ./srcds_run command line to generate a debug.log to help with solving this problem
Tue Jul 5 18:14:50 CEST 2011: Server restart in 10 seconds
Running a benchmark to measure system clock frequency...
Tue Jul 5 18:15:02 CEST 2011: Server Quit



this occured when someone tried connecting to the server. Firstly its on Centos, and apache isnt installed on this system as its not required as i aint running any web services on here.

i tried accessing the /tmp/dumps/crash_20110705181450_1.dmp under nano and it came out in a unknown format

Please advise.

Myatu
04-07-2011, 18:26
You need to specify this in the software application specifically. For example, Apache binds to "0.0.0.0" by default, which means "all available IP addresses". So, 127.0.0.1, the main IP address, the failover IP address, etc. You can tell Apache specifically to listen/bind on your failover IP address instead of the "0.0.0.0".

However, this is different for each application and some applications won't support this (as, for example, most web browsers - though they generally do not "listen" to begin with). In that case you may have to "trick" it into using a different IP.

avizeke
04-07-2011, 03:30
thanks for your response.

i want to use both the main IP and a fail safe for both outgoing. are you saying i need to address iptables?

i just want to re address the properties in which the centos dedi system sees the fail safe ip. Once added with to the network adaptor device and its pingable, i just need to change on how the software sees the ip so that it doesnt bind to the main system.

i just want to use the ip as i can the normal one were i can get traffic inbound and outbound from it in the same way i can with the primary ip

again thanks for your responses

Myatu
02-07-2011, 12:55
Both the main IP and failover IP are separate. However, with the default routing, all outgoing traffic will go through the main IP, regardless if you add a failover IP.

Because you cannot (or, better said: SHOULD not) remove the main IP, you need to "trick" outgoing traffic in using the failover IP address. You can do this using iptables (search the forum for "iptables failover" for several solutions) or by modifying your routing table (see "ip route" for the current configuration, looking at "default via").

You use the iptables solution if you need to use the failover IP for accessing specific remote ports or IPs, or the routing table if you wish to use it for anything.

This is not a limitation of the CentOS or OVH's distribution, but default on all operating systems.

avizeke
02-07-2011, 04:33
i am trying to use other services on to a additional ip other than the one situated with the system as the stable one.

so i want the fail safe IP to be able to use it as outgoing traffic. so that the other services i put on it cab sens outgoing packets.


adding the fail safe ip following the OVH IP ALIAS tutorial in order to add a ip to the network adaptor device. But i am trying to remove the bind, that is automaticallt in place.

i am trying to understand if there is a way that i can remove the bind from the fail safe ip.

Razakel
01-07-2011, 17:07
What are you trying to do?

avizeke
01-07-2011, 05:31
is there anyway to stop a failsafe ip from binding on network configuration on a centos 5 template from OVH.

we just want to bind the ips to the network card and not force them to bind. because at the moment, the fail safe is auto binding to to the stable ip and this is not what we want.

The CentOS 5.0 Image installed to the server appears to be a image intended for a VM. The envionment as configured does not adhear to a standard CentOS 5.0 install.