Discussions
Categories
Groups
Community Home
Categories
INTERNAL ENABLEMENT
POPULAR
PUBLIC CLOUD
PRIVATE CLOUD
Quick Links
MY LINKS
HELPFUL TIPS
Back to website
Home
Web CMS (TeamSite)
Deploying to a cluster
mike_jaixen
We are converting our OpenDeploy receivers (HP-UX) from an environment where one server is always the primary and another is a failover server to an environment where the primary and failover servers will switch back and forth. In the old environment, we only had OpenDeploy installed on the primary server, and if that server failed, we would be unable to deploy until the primary was back in service, and we failed back to the primary.
Now, when we have a failure, we will switch over to the other server, and stay there until it encounters a failure. Since that could take days/weeks/months, we will need to have OpenDeploy running on both servers. The failure will be managed through the DNS, so we will use an alias that will resolve to the IP address of whichever server is designated as the active server.
Here is how I plan to configure:
In odnodes.xml on my TeamSite server, I'll use the DNS alias that will point to the currently designated primary server instead of the machine name of the primary server.
In odrcvr.xml on the receivers, I'll use <localNode host="0.0.0.0">.
The hope is that not only will everything work with this configuration, but also that if a failover occurs and the DNS is updated, my TeamSite server will send the deployments out, and it will be received by the newly activated OpenDeploy server, and nobody will notice the difference. (Except someone that might have been in the middle of a deployment when the failure occurred, who might have to redeploy...)
Does this make sense? Am I on the right track? Or is this plan headed for trouble?
Find more posts tagged with
Comments
Migrateduser
So if I understand correctly, the sending host's odnodes.xml will use the DNS name for the target, which resolves to a different IP depending upon which target host is considered primary. Each of the target hosts (primary and backup) has a receiver running.
It seems you should be able to configure each receiver normally, since the sender will direct deployments to one or the other, depending upon which IP the target host name resolves to. Each would have to have the sending host in its allowedHosts element.
Todd Scallan
Director of Product Management
Interwoven
t: 408-530-7167
e:
tscallan@interwoven.com
mike_jaixen
That's it exactly. Sounds like this should be easy, then.