We are having 3 domino servers (Clustered). Today one of the server was down(RBDL01).
At the time when user’s try to log-in it comes up with notes client configuration wizard to enter name and server details, when I try to enter other server (RBDL03 or RBDL02, Full Net Address / IP Address / FQHN) it comes up with “The server is not responding. the server may be down or you may be experiencing network problems”
But the user’s who were already logged were not affected as they are able to access the databases on other servers.
Same message even with other server as well. Issue resolved once the server was up (which was down for at least an hour due to OS issues)
When I compared the server’s configuration documents with RBDL01, I was not able to find any difference. Can some one suggest where could be the wrong please.
Do let me know in case if required any further details.
It’s been my experience that no matter which server name you might enter in the Client Configuration Wizard, the Wizard always tries to connect back to the user’s home server. So, the fact that you receive the message “The server is not responding…” while the server is down, I would say, is normal.
Once the user’s client is properly set up, it should first try to connect to the user’s home server, and if that is unavailable, connect to another server in the cluster.
To my knowledge, I’ve never had a client come up with the Configuration Wizard when the home server is down. The only thing that comes to mind is if the user’s client didn’t know how to connect to any other server - are there connection documents to all servers listed in the client’s addresses book?
Subject: Issue was due to default client settings file
Hi Mark and Amit,
Thank you so much for all your information. Issue was due to default client settings file, in which one of the server.name was hard coded. Once after changing the server name it it works with out any issue. Anyway we are testing the settings to check about the impact for not giving the server details.
If you are using notes client build 9.0.x then below is the solution.
The fix for support of Failover on client configuration is in 9.0.x build. To support that you have to do like :
This fix is implemented considering the below things :
Cluster.ncf file with cluster server member details should be present in user’s “Data” Directory while configuring the Client.
names.nsf should be present in user’s “Data” directory. It should contain the connection doc information for all cluster member.
However, with multi-user install, this was not possible as the user data directory gets created during client configuration.
Workaround for multi-user install :
Now, with the solution given through this SPR, we have an alternative way for multi-user install.
So, now, we need to have 2 things as pre-requisite for first time client configuration
cluster.ncf in shared data directory
OS hosts file updated with the IP, DNS & domino name of all cluster members.
Hope this will help.
NOTE : Administrator responsibility to provide pre-populated cluster.ncf and names.nsf file.