I am writing a server that does GSS authentication for an NFS server. For performance reasons, I wish to run multiple servers, all of which are using the same Acceptor Credentials.
I found that the Replay Cache file would get trashed if several of these servers tried to get server credentials at the same time. I fixed that by getting the server credentials before forking off the N service daemons. It seems to work ok now, but I was wondering if anyone knows if there are other Gotchas in the MIT KerberosV libraries when doing this? Thanks in advance for any help, rick ________________________________________________ Kerberos mailing list [EMAIL PROTECTED] https://mailman.mit.edu/mailman/listinfo/kerberos
