[LUDOC-98] Document new lctl replace_nids feature

• Type: Task
• Resolution: Fixed
• Priority: Blocker
• Labels: LB
• Assignee: Richard Henwood (Inactive)
• Reporter: Andreas Dilger

Description

This is to track documentation for the "lctl replace_nids" feature in LU-1430.

Issue Links

• is related to: LU-1430 Changing network address without --writeconf option

Comments

[Richard Henwood (Inactive)]

Artem's change landed a while ago.

[Artem Blagodarenko (Inactive)]

http://review.whamcloud.com/7459

[Andreas Dilger]

Artem, could you please submit a patch to the Lustre manual with the text that Nathan pasted above. Please see the following pages for details:

http://wiki.hpdd.intel.com/display/PUB/Making+changes+to+the+Lustre+Manual
http://wiki.hpdd.intel.com/display/PUB/Making+changes+to+the+Lustre+Manual+source

[Andreas Dilger]

I don't think there is any confusion about the actual documentation or where it applies. My comments are related to actually getting the text that Nathan wrote into the Lustre Manual. Please see:

http://wiki.whamcloud.com/display/PUB/Making+changes+to+the+Lustre+Manual
http://wiki.whamcloud.com/display/PUB/Making+changes+to+the+Lustre+Manual+source
http://wiki.whamcloud.com/display/PUB/Guidance+for+Manual+reviewers

[Nathan Rutman]

After the landing of LU-1430, the method described here should be used. In previous versions of Lustre, back to 1.6, the old version must be used.

[Artem Blagodarenko (Inactive)]

This patch adds new method to change a NID. But you can use previous with --writeconf that is described here http://wiki.lustre.org/manual/LustreManual20_HTML/LustreMaintenance.html#50438199_31353

[Richard Henwood (Inactive)]

Can you clarify: are there now two methods to change a NID depending on which version of Lustre is being used?

[Andreas Dilger]

Nathan, this would be better off in the form of a patch to the manual (as posted to the list last week), rather than a comment in this ticket. Don't get me wrong, I'm happy to have this text from you, but we're trying to get more contributors to the manual, and if Richard and I are the only ones making the updates, the process and visibility into the manual will never improve for external contributors.

[Nathan Rutman]

Doc:

14.5 Changing a Server NID

If you need to change the NID on the MDT or an OST, a special command must be issued with all Lustre servers and clients down, except for the MGS.
Change a server NID in these situations:
■ New server hardware is added to the file system, and the MDS or an OSS is being moved to the new machine
■ New network card is installed in the server
■ You want to reassign IP addresses

The replace_nids command differs from tunefs.lustre --writeconf in that it does not erase the entire configuration log, precluding the need to "writeconf" all servers and re-specify all permanent parameter settings. However, the "writeconf" method can still be used if desired.

To change a server NID:

1. Update the LNET configuration in the /etc/modprobe.conf file so the list of server NIDs (lctl list_nids) is correct.
   The lctl list_nids command indicates which network(s) are configured to work with Lustre.
   Chapter 14 Lustre Maintenance 14-5
2. Shut down the file system in this order.
   a. Unmount the clients.
   b. Unmount the MDT.
   c. Unmount all OSTs.
3. If the MGS and MDS share a partition, start the MGS only:

   mount -t lustre <MDT partition> -o nosvc <mount point>

4. Run the replace_nids command on the MGS.

    lctl replace_nids <devicename> <nid1>[,nid2,nid3 ...]

   where <devicename> is the Lustre target name, e.g. myfs-OST0013

5. If the MGS and MDS share a partition, stop the MGS:

   umount <mount point>

6. Restart the file system in this order.
   a. Mount the MDT (or the combined MGS/MDT).
   b. Mount the OSTs.
   c. Mount the clients.

Note: The replace_nids command also cleans all old, invalidated records out of the configuration log, while preserving all other current settings.
Note: The previous configuration log is backed up on the MGS disk with the suffix '.bak'.