Re-IP linux server

[brintaz]

Hi! Does anybody knows how to apply Re-IP rules to linux server? Thank you!

[Vitaliy S.]

Hi Fabrizio, re-IP is only supported for Windows VMs at this time. Thank you!

[tsightler]

You can use a simple in-guest script to do this outside of Veeam itself. This exact process varies based on the specific Linux distribution, but I previously used a simple script that I included in rc.local that would attempt to ping the default gateway, and if it failed, it updated to the secondary IP address. I think there was an example posted to this forum previously.

[Gostev]

Below is Tom's post from internal forums regarding this:

At my old job we used a script in rc.local (for the non-linux versed, it's run during each boot). The script would attempt to ping the default gateway and, if it failed, would swap the network configuration and restart the network services. It was something like this:

Code: Select all

#!/bin/sh
TARGET=`ip route | awk '/default via/ {print $3}'`
RET=`ping -c $PACKETS $TARGET 2> /dev/null | awk '/packets received/ {print $4}'`

if [ "$RET" -ne "$PACKETS" ]; then
     echo "Gateway unreachable...swapping to secondary network."
     mv /etc/sysconfig/network /etc/sysconfig/network.pri
     mv /etc/sysconfig/network.sec /etc/sysconfig/network 
     service network restart
else
      echo "Network is up via $TARGET"
fi

The above is just a concept, our actual script was a quite a bit more sophisticated, but the point is it's pretty easy to deal with the issue.

[brintaz]

Thank you very much !!! Fabrizio

[tsightler]

This particular script was used with Redhat based variants, probably RHEL4 and RHEL5. It would likely need some tweaks and changes for newer versions or other distros like Ubuntu.

[pwjohnston]

Does anyone know of a script that would Re-IP Linux (SLES & RHEL) from the DR side with say PowerCLI or VIX API?

[tsightler]

I haven't tested this with recent RHEL variants, but PowerCLI has a Set-VMGuestNetworkInterface command that certainly worked back in the RHEL5 days and actually was extensible with custom scripts for other Linux. It's use was relatively simple:

Code: Select all

$cred = Get-Credential
$interface = Get-VMGuestNetworkInterface -VM lnx01 -GuestCredential ($cred) | ?{ $_.Ip -like "192.168.1.*"}
Set-VMGuestNetworkInterface -VMGuestNetworkInterface $interface -Ip 192.168.2.10 -Netmask 255.255.255.0 -Gateway 192.168.2.1 -GuestCredential $cred

So this would attempt to find an interface currently in the 192.168.1.0 subnet and would change the IP address of that interface to 192.168.2.10. Obviously you could get more complex to do cooler things like network mapping. I tested this in my lab on a RHEL5 VM (vSphere 5.5 infrastructure) and the IP changed. You must of course have VMware tools installed in the guest VM.

[pwjohnston]

Awesome, thanks tsightler!

[Nir]

Hello everyone,

I'm trying to come up with a Bash-based solution for Re-IP functionality in Linux.
Has anyone here succeeded in creating a working Re-IP functionality for Linux systems, specifically RHEL 6 and 7?

I've been looking around and found https://www.vmware.com/support/develope ... rface.html
Which states it works only on various Windows systems and on RHEL 5.

If anyone has a working solution I would greatly appreciate it if you would be willing to share it with me.


Thanks in advance,
Nir.

[jeffsterv]

Now that I have a requirement to do this and it's been almost five years since this the topic was brought up... does anyone know if-and-or-when the feature will be supported?

[veremin]

We cannot provide any comments regarding future product versions. So, for now the in-guest re-ip script is the only way to go. Thanks.

[joebranca]

Are there any updated best practices or planned functionality for the topic of re-IP on Linux replicas? Someone posted a script above that worked pretty well until we started deploying RHEL8, so I'm just wondering how folks are handling some kind of automated network reconfigs for Linux failover processes.

[joebranca]

Following up on my inquiry last week, have Veeam engineers or any users come up with an updated Linux script solution that works with RHEL 8 commands and resolves the hardware address conflict created by the failed over and power up Veeam process?

In the Linux script under RHEL 8, neither bouncing the network services nor NetworkManager services have any effect on network interface changes performed dynamically using the nmcli command.

In the new (DR) environment, the HWADDR paramenter in the ifcfg-<DEVICE> (e.g. ifcfg-ens192) file is initializzed to the old (PROD) hardware address but in the new (DR) environment, a different hardware address is assigned. This hardware address conflict is preventing the network interface from coming up and allowing the Linux script to perform its re-IP steps.

[joebranca]

bump

[RomanK]

Hello Joe,

I have asked our QA engineer in case they have some scripts for RHEL 8 but unfortunately without luck. No best practices either. However, the topic is interesting, I cannot promise a lot but we'll research.

Thanks