23.07.2014 Views

Lustre 1.6 Operations Manual

Lustre 1.6 Operations Manual

Lustre 1.6 Operations Manual

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

4.2.2 More Complex Configurations<br />

In case of NID/node specification (a node is a server box), it may have multiple<br />

NIDs if it has multiple network interfaces. When a node is specified, generally all of<br />

its NIDs must be listed, delimited by commas (,) so other nodes can choose the NID<br />

appropriate to their own network interfaces. When multiple nodes are specified,<br />

they are delimited by a colon (:) or by repeating a keyword (--mgsnode= or --<br />

failnode=). To obtain all NIDs from a node (while LNET is running), run:<br />

lctl list_nids<br />

4.2.2.1 Failover<br />

This example has a combined MGS/MDT failover pair on uml1 and uml2, and a<br />

OST failover pair on uml3 and uml4. There are corresponding Elan addresses on<br />

uml1 and uml2.<br />

uml1> mkfs.lustre --fsname=testfs --mdt --mgs \<br />

--failnode=uml2,2@elan /dev/sda1<br />

uml1> mount -t lustre /dev/sda1 /mnt/test/mdt<br />

uml3> mkfs.lustre --fsname=testfs --ost --failnode=uml4 \<br />

--mgsnode=uml1,1@elan --mgsnode=uml2,2@elan /dev/sdb<br />

uml3> mount -t lustre /dev/sdb /mnt/test/ost0<br />

client> mount -t lustre uml1,1@elan:uml2,2@elan:/testfs /mnt/testfs<br />

uml1> umount /mnt/mdt<br />

uml2> mount -t lustre /dev/sda1 /mnt/test/mdt<br />

uml2> cat /proc/fs/lustre/mds/testfs-MDT0000/recovery_status<br />

Where multiple NIDs are specified, comma-separation (for example, uml2,2@elan)<br />

means that the two NIDs refer to the same host, and that <strong>Lustre</strong> needs to choose the<br />

"best" one for communication. Colon-separation (for example, uml1:uml2) means<br />

that the two NIDs refer to two different hosts, and should be treated as failover<br />

locations (<strong>Lustre</strong> tries the first one, and if that fails, it tries the second one.)<br />

Note – If you have an MGS or MDT configured for failover, perform these steps:<br />

1. On the OST, list the NIDs of all MGS nodes at mkfs time.<br />

OST# mkfs.lustre --fsname sunfs --ost --mgsnode=10.0.0.1<br />

--mgsnode=10.0.0.2 /dev/{device}<br />

2. On the client, mount the filesystem.<br />

client# mount -t lustre 10.0.0.1:10.0.0.2:/sunfs /cfs/client/<br />

4-10 <strong>Lustre</strong> <strong>1.6</strong> <strong>Operations</strong> <strong>Manual</strong> • September 2008

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!