2008/6/4 Serge Dubrouski <[EMAIL PROTECTED]>: > On Tue, Jun 3, 2008 at 9:53 AM, Xinwei Hu <[EMAIL PROTECTED]> wrote: >> 2008/6/3 Serge Dubrouski <[EMAIL PROTECTED]>: >>> I found it, though it was a commit for nfsexport.sh. One thing isn't >>> clear though. >>> >>> Now they prefer to bind /var/lib/nfs to a subdir under exported >>> filesystem. How do they support multiple NFS resources in this case? >>> How do they support an active/active configuration meaning 2 active >>> NFS severs serving different resources but able to take over each >>> other? >> No, they can't. > > What was the reason to deprecate clurmtabd then? It was able to solve > an active/active problem all right. No, it didn't. >> >> For what you need, an RA for _exportfs_ will depend on an newly >> introduced kernel interface >> via /proc/fs/nfsd/unlock_filesystem. >> And also depends on a scripts passed back to rpc.statd. >> > > How will it help to sync rmtab files between nodes in cluster? From > what I see /proc/fs/nfsd/unlock_ip and > /proc/fs/nfsd/unlock_filesystem will help with clearing locks not to > preserve client mount information. This is critical cause otherwise you have to shutdown whole nfsd to umount the filesystem.
For rmtab, you can do it with rpc.mountd -H if needed. > -- > Serge Dubrouski. > _______________________________________________________ > Linux-HA-Dev: Linux-HA-Dev@lists.linux-ha.org > http://lists.linux-ha.org/mailman/listinfo/linux-ha-dev > Home Page: http://linux-ha.org/ > _______________________________________________________ Linux-HA-Dev: Linux-HA-Dev@lists.linux-ha.org http://lists.linux-ha.org/mailman/listinfo/linux-ha-dev Home Page: http://linux-ha.org/
