Unsolved
This post is more than 5 years old
1 Rookie
•
9 Posts
0
924
October 23rd, 2009 15:00
Failing to backup Solaris 10 server with zones
A couple of our client machines are Solaris 10 servers with multiple data zones.
Let's see if I can describe this clearly. Our Networker server has been failing to backup a certain host represented within a particular Solaris machine, and that host is configured for one of the four physical interfaces on the global zone that is that Solaris server. The Solaris server and it's Solaris Zones are being backed up from the global zone when Networker follows the Zone storage path at /zones. The local zones are not backed up by Networker through direct connections to the zones.
Two other Solaris servers have similar setups without the failing backups. In this case a particular host errors out during attempted backup because the Networker client is complaining that x.x.x.x IP address is not properly configured on the Networker server (the IP address of one of the other four network interfaces on the server which the host should *not* be trying to route traffic through. Somehow just the Networker client is traversing these other physical interfaces that have other FQDN's associated with them.
Supposing the hostname in question is "so2" and the hostname associated with another interface is called "so3", an examples of the errors are:
so2:/zones/k5 39078:save: SYSTEM error: client `so3' is not properly configured on the NetWorker Server
Or it may reference an IP address of another interface instead.
I've tried trickery on the server end like adding entries to the /etc/hosts file for the other interfaces' IP addresses and such, adding aliases in the client config but nothing has seemed to work.
Is there a known issue with Networker and Solaris zones, or is there a simple fix or wrong configuration I'm missing? I am not familiar with Solaris 10 and zones, and don't have direct access to the Solaris server in question, so it may be something easy I am not aware of to suggest to the administrator of the Solaris server.
Our Networker server and the clients are all at 7.5.1. The Solaris 10 machine is not at the latest Solaris update, but neither are the other Solaris machines that are getting backed up successfully.
thanks in advance,
joe
Let's see if I can describe this clearly. Our Networker server has been failing to backup a certain host represented within a particular Solaris machine, and that host is configured for one of the four physical interfaces on the global zone that is that Solaris server. The Solaris server and it's Solaris Zones are being backed up from the global zone when Networker follows the Zone storage path at /zones. The local zones are not backed up by Networker through direct connections to the zones.
Two other Solaris servers have similar setups without the failing backups. In this case a particular host errors out during attempted backup because the Networker client is complaining that x.x.x.x IP address is not properly configured on the Networker server (the IP address of one of the other four network interfaces on the server which the host should *not* be trying to route traffic through. Somehow just the Networker client is traversing these other physical interfaces that have other FQDN's associated with them.
Supposing the hostname in question is "so2" and the hostname associated with another interface is called "so3", an examples of the errors are:
so2:/zones/k5 39078:save: SYSTEM error: client `so3' is not properly configured on the NetWorker Server
Or it may reference an IP address of another interface instead.
I've tried trickery on the server end like adding entries to the /etc/hosts file for the other interfaces' IP addresses and such, adding aliases in the client config but nothing has seemed to work.
Is there a known issue with Networker and Solaris zones, or is there a simple fix or wrong configuration I'm missing? I am not familiar with Solaris 10 and zones, and don't have direct access to the Solaris server in question, so it may be something easy I am not aware of to suggest to the administrator of the Solaris server.
Our Networker server and the clients are all at 7.5.1. The Solaris 10 machine is not at the latest Solaris update, but neither are the other Solaris machines that are getting backed up successfully.
thanks in advance,
joe
No Events found!