Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I heard rumors at first and later saw it once that the sparc lab at my university occasionally had to be shut down and turned on in a particular order to get the whole thing to spool back up after a server glitch. I think the problem got really nasty once you had NFS mounts from multiple places.


sync with tcp on NFS 2 had the hang problem. async on NFS 3+ was better but could still hang if configured using sync.

NFS 4.1 introduced pNFS scalability and 4.2 has even more optimizations.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: