Don't hang user processes if Kerberos ticket for nfs4 mount expires
Subsystems (2)
So what are they supposed to do without tickets? Crash?
Jim Rees <[email protected]> wrote:
I'm not arguing that the client machine should hang when tickets expire. I'm arguing that user processes should not hang when tickets expire. Sorry if that wasn't clear.
Myklebust, Trond wrote:
So what are they supposed to do without tickets? Crash?
No, why would they want to do that?
I feel like I've entered the Twilight Zone here, so please continue without me while I dig up the relevant background info.
Quoted 8 lines · Jim Rees
Myklebust, Trond wrote:
So what are they supposed to do without tickets? Crash?
No, why would they want to do that?
I feel like I've entered the Twilight Zone here, so please continue without me while I dig up the relevant background info.
The point is that if the server won't allow them to issue read or write system calls, then there are 2 options available to them:
1. hang until someone renews their ticket
2. get an error which means crash, since most (all?) applications aren't written according to the non-posix assumption that read and write can return EACCES/EKEYEXPIRED errors.
Trond
Signature · 6 lines
Trond Myklebust Linux NFS client maintainer NetApp [email protected] www.netapp.com