I can expand the working filesystem on the fly. But...
I tested the fs expansion while the SFS benchmark was running on the NFS client
against the NFS shared filesystem sitting on metavolume.
Every time I ran "metaset -s test -a [i]newdevicename[/i]", I've got an error
on the NFS client:
SPEC SFS Benchmark Version 3.0, Creation - 11 July 2001
NFS Protocol Version 3
<...>
sfs37: server not responding: RPC: Port mapper failure - RPC: Timed out
sfs37: portmap/mountd nfsserver:/mnt/8 server not respondingsfs3: caught
unexpected SIGCHLD. Exiting...
SFS NFS Version 3 Benchmark Client Logfile
Client hostname = nfsclient
Prime Client hostname = nfsclient
SPEC SFS Benchmark Version 3.0, Creation - 11 July 2001
NFS Protocol Version 3
<...>
sfs38: server not responding: RPC: Remote system error - Bad file number
sfs38: rpc initialization failed
sfs3: caught unexpected SIGCHLD. Exiting...
The sfs_prime says:
/export/home/spec/SFSV3.0/spec-sfs3.0/benchspec/162.nfsv2/bin/sfs_prime:
Prime Client got too many signals - expected 1 got 1000
The sfs_mgr says:
sfs_mgr: sfs_prime returned an error, exiting
This message posted from opensolaris.org
