c60244@ccfiris.aedc (Kenny McDonald) (02/28/91)
I have a 4D240GTX running as the NIS master server (no secondary servers for now). My clients are a 4D25 PI and another 4D240GTX (which are on the opposite side of a bridge than myself). Today we ran into a problem with our bridge and we lost communications between the clients and the server. The PI's passwd file had included NIS entries for all users but root, and the 240 had all local passwd entries with the exception of a couple of users. Our problem is that when we lost communications, our NIS clients locked up. No one could log into them; we kept getting login timeout (no YP server). We tried to reboot our 240 and it was very,very,very slow in rebooting. We did finally get root logged in on the PI. We created a local entry (on PI) for one of our users that is also on the 240. This user was able to log on to the PI, but it was very slow in doing so, and wsh windows were slow in popping up. Once the 240 rebooted the it was up, this user tried to rlogin to it. But no luck; we got the same message (no YP server). Doesn't these systems use the local ascii files if the NIS server is down, or do you have to kill the ypbind process in order for this to occur? Is there something I should do in order to let it use the local files if there is something screwed up with NIS? I am going to set up the other 240 today as a secondary server to do away with this problem in the future. Any comments/suggestions are welcomed on this topic. Thanks.