HP StorageWorks Scalable File Share - Manual (Page 263)

Troubleshooting file system performance 9–39
2. Rename the new file to the original name using the
mv
command, as shown in the following example:
# mv data.new data.1
mv: overwrite ’data.1’? y
# lfs getstripe data.1
OBDS:
0: ost1_UUID
1: ost2_UUID
2: ost3_UUID
3: ost4_UUID
./data.1
obdidx objid objid group
0 1860 0x744 0
1 1856 0x740 0
2 1887 0x75f 0
3 1887 0x75f 0
#
9.27.3 Checking for unbalanced distribution of OST services
One possible cause of poor file system performance is that one or more of the servers that are serving the
file system may be running in failover mode—that is, a server has failed and its services are being served
from the other server in the server pair.
To determine whether this is the case, enter the following command:
sfs> show filesystem scratch
Name: scratch
OSTs: ost[5-8]
State: started
Mountpoint: /mnt/scratch/
Stripe size: 4194304
Stripe count: 4
Interconnect: gm tcp
MDS mount options: acl,user_xattr
OST mount options: extents
Lustre timeout: 200
Quota options: quotaon=ug
MDS Information:
Name LUN Array Controller Files Used Service State Running on
----- --- ----- ---------- -------- ---- ------------- ----------
mds4 2 1 B 2.3M 20% running south2
OST Information:
Name LUN Array Controller Size(GB) Used Service State Running on
----- --- ----- ---------- -------- ---- ------------- ----------
ost5 13 3 A 290 30% running south4!
ost6 14 3 B 290 31% running south4!
ost7 17 4 A 290 29% running south4
ost8 18 4 B 290 30% running south4
(servers marked ! are not preferred server)
In this example, all four OST services are running on one Object Storage Server (
south4
). The exclamation
mark (
!
) indicates that services
ost5
and
ost6
are not running on their preferred server. Use the
show
server server_name
command to determine the peer server for server
south4
, as follows:
sfs> show server south4
Name: south4
Primary Role: ost
Backup Server: south3
.
.
.