>> If you have application servers in the back end (Windows, *nix desktops, ssh 
>> and screen, etc), you would just reboot the box you want to patch into 
>> >>single user mode, and the units will connect to the other server. The 
>> outage would be less than a minute, and they would resume where they left 
>> off.

No windows backends, just two Solaris servers


>>If you're serving desktops from the SRS or have different type of setup.. 
>>Then you would put the server in offline mode so new sessions can be created 
>>on that server. Then you would have to >>get the current sessions on that 
>>server off, by either telling the users to logout (or forcing them?)

That's what I was afraid of.  I think I'll have to plan my outages days in 
advance so I can take one of the servers offline so no new connections can be 
made to it.



Carl Holzhauer wrote:
I had asked this question before, but never really received many responses.
We're using two SRSS servers to serve a manufacturering environment that runs 
three shifts. This means that there is no good time to schedule a server for an 
outage to patch it. My question is this. Is there a way to move sessions from 
one SRSS server to another without loss of data or losing the spot users were 
working in? How does everyone else deal with patching in a 24/7 environment? 
Thanks



_______________________________________________
SunRay-Users mailing list
[email protected]
http://node1.filibeto.org/mailman/listinfo/sunray-users

Reply via email to