[Bug 289711] NFS4 access is very slow - "Request couldn't be completed in time"
| Newsgroups | gmane.os.freebsd.devel.file-systems |
|---|---|
| Message-ID | <[email protected]/bugzilla/> |
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=289711 --- Comment #11 from Rick Macklem <[email protected]> --- (In reply to Peter Much from comment #10) Well, expired (lost locks) indicates that the client has been network partitioned from the server for some reason (assuming the server did not reboot). or Two or more clients end up with the same hostid. This typically happens when someone clones root disk images with a /etc/hostid set in it. or Something else is somehow causing the NFS4ERR_BAD_STATEID error return (that's the 10025 replies). For NFSv4.1/4.2, which is what you should be getting, I cannot think how this can ever happen. Just in case, check this with # nfsstat -m on the client while mounted. It should not show a minor version of 0. Another thing that can cause serious confusion is dismounting a file system on the server while it is exported. (This causes all sorts of weirdness, because it uncovers the parent fs and that file system might not be exported or will have a different inode# for the mount point or..) I will take a closer look at the packet dump one of these days and see if I can figure out more w.r.t. what is going on. And this is happening when you do not have delegations enabled? (vfs.nfsd.issue_delegations=0) -- You are receiving this mail because: You are the assignee for the bug.