WARNING: This is the _old_ Lustre wiki, and it is in the process of being retired. The information found here is all likely to be out of date. Please search the new wiki for more up to date information.

Upgrading to a New Version of Lustre: Difference between revisions

From Obsolete Lustre Wiki
Jump to navigationJump to search
 
(56 intermediate revisions by 2 users not shown)
Line 1: Line 1:
This section provides information about supported upgrades, upgrade paths, and interoperability of nodes with different versions of Lustre installed. It also describes procedures for upgrading your Lustre file system to a new version of Lustre. For information about available releases, see [[Lustre Release Information]].
<small>''(Updated: Dec 2009)''</small>


__TOC__
This section provides information about supported upgrades, upgrade paths, and interoperability of nodes with different versions of Lustre™ installed. It also describes procedures for upgrading your Lustre file system to a new version of Lustre. For information about available releases, see [[Lustre_Release_Information|Lustre Release Information]].


== Supported Upgrades ==
== Supported Upgrades ==
Line 8: Line 10:
* Lustre 1.8.x (any minor version) to Lustre 1.8.x (latest version).
* Lustre 1.8.x (any minor version) to Lustre 1.8.x (latest version).


For Lustre 1.8.0, if you have upgraded from an earlier version, you can downgrade to that version [[or any version in between?]]. For example:
=== Lustre Component Interoperability ===
* If you upgrade from Lustre 1.6.x > 1.8.0, you can downgrade to Lustre 1.6.x [[(latest version?)]].
* If you upgrade from one minor version to the next (for example, Lustre 1.8.0 >
1.8.x), you can downgrade to the earlier minor version (Lustre 1.8.0).


To downgrade from 1.8.x to 1.6.x, see [[#More_Information|More Information]].
Lustre interoperability enables 1.8.x servers ("new" servers) to work with 1.6.x clients ("old" clients), 1.6.x servers ("old" servers) to work with 1.8.x clients ("new" clients), and "mixed" environments with both 1.6.x and 1.8.x servers. For
example, half of each OSS failover pair could be upgraded to enable a quick reversion to 1.6 by powering down the 1.8 servers.


'''''Caution:''''' A new installation of Lustre 1.8.x is not guaranteed to be downgradeable to an earlier Lustre version.
The table below describes the interoperability between clients, OSTs, and MDTs with different versions of Lustre installed.
 
=== Supported Upgrade Paths ===
Two Lustre upgrade paths are supported. Either can be applied to the entire file system or a combination of the two can be used to upgrade individual servers and clients.
 
* ''Rolling upgrade'' - Individual servers (or their failover partners) and clients are upgraded one at a time and restarted, so that the file system never goes down. This type of upgrade limits the ability to change certain parameters.
* ''Entire file system upgrade'' - All servers and clients are shut down and upgraded at the same time.
 
=== Interoperability between Nodes with Different Versions of Lustre Installed ===
 
Interoperability between clients, OSTs, and MDTs with different versions of Lustre installed is described below.


{| border="1" cellspacing=0 cellpadding="10"
{| border="1" cellspacing=0 cellpadding="10"
! Lustre Component !! Interoperability with Other Lustre Components
! Lustre Component !! Interoperability with Other Lustre Components
|-
|-
| rowspan="3" valign="top" | Clients
| rowspan="1" valign="top" | Clients
| Old live clients can continue to communicate with [[old/new/mixed]] servers.
| &#149; Old, live clients can communicate with old, new, or a mix of servers.<br>&#149; Old clients can start up using old, new, or a mix of servers.<br>&#149; New clients can start up using old, new, or a mix of servers.<br>'''''Note:''''' Old clients cannot mount a file system that was created by a new MDT.
|-
| Old clients can start up using old/new/mixed servers.
|-
|-
| New clients can start up using old/new/mixed servers (use old mount format for
| rowspan="1" valign="top" | OSTs
old MDT).
| &#149; Old OSTs can communicate with new clients/MDTs.<br>&#149; New OSTs can only be started after the MGS has been started (typically after the MDT has been upgraded).
|-
|-
| rowspan="2" valign="top" | OSTs
| rowspan="1" valign="top" | MDTs
| New clients/MDTs can continue to communicate with old OSTs.
| &#149; An old MDT can communicate with new clients.<br>&#149; A new co-located MGS/MDT can be started at any point.<br>&#149; A new non-co-located MDT can be started after the MGS starts.
|-
| New OSTs can only be started after the MGS has been started (typically this
means "after the MDT has been upgraded.")
|-
| rowspan="3" valign="top" | MDTs
| New clients can communicate with old MDTs.
|-
| New co-located MGS/MDTs can be started at any point.
|-
| New non-MGS MDTs can be started after the MGS starts.
|}
|}


Interoperability limitations:
== Upgrading Lustre 1.6.x to Lustre 1.8.x ==
Two Lustre upgrade paths are supported to meet the upgrade requirements of different Lustre environments.
 
* ''Entire file system upgrade'' - All servers and clients are shut down and upgraded at the same time. See [[#Performing a Complete File System Upgrade|Performing a Complete File System Upgrade]].
* ''Rolling upgrade'' - Individual servers (or their failover partners) and clients are upgraded one at a time and restarted, so that the file system never goes down.  See [[#Performing a Rolling Upgrade|Performing a Rolling Upgrade]].
 
'''''Note:''''' If you upgrade some Lustre components to 1.8.x but not others (such as running 1.8 clients in a file system with 1.6 OSTs), and run a mixed environment, you may see one or more warnings similar to this:
 
<pre>
LustreError: 3877:0:(socklnd_cb.c:2228:ksocknal_recv_hello())
Unknown protocol version (2.x expected) from 192.168.2.43
</pre>
 
This warning is given when the 1.6 and 1.8 components use different protocols. It
can be safely ignored because the Lustre components negotiate a common protocol.
In this example, the 1.8 clients fall back to use the 1.6 protocol with the 1.6 OSTs.


* Old clients cannot mount a file system which was created by a new MDT.
=== Performing a Complete File System Upgrade ===
This procedure describes a complete file system upgrade in which 1.8.x Lustre packages are installed on multiple 1.6.x servers and clients, requiring a file system shut down. If you want to upgrade one Lustre component at a time and avoid the shutdown, see [[#Performing a Rolling Upgrade|Performing a Rolling Upgrade]].


* If your file system was originally created using Lustre 1.8.x, you will not be able to mount '''the''' '''[[(a?)]]''' file system created using an earlier Lustre version. However, if your system has been upgraded from 1.6.x to 1.8.x, you can mount Lustre clients on both Lustre versions. Lustre 1.8.x clients are interoperable with 1.6.x servers.
'''''Tip:''''' In a Lustre upgrade, the package install and file system unmount steps are reversible; you can do either step first. To minimize downtime, this procedure first performs the 1.8.x package installation, and then unmounts the file system.


== Prerequisites to Upgrading Lustre==
1. ''Make a complete, restorable file system backup before upgrading Lustre.''


This content was in the OM section "Upgrading from Lustre 1.6.x to Lustre 1.8.0"
2. ''Install the 1.8.x packages on the Lustre servers and/or clients.'' Some or all servers can be upgraded. Some or all clients can be upgraded. For help determining where to install a specific package, see [[Lustre Packages]].


[[What are these procedures? Do they need to be completed before the upgrade is started?]]
:a. ''Install the kernel, modules and ldiskfs packages.'' For example:


Remember the following points before upgrading Lustre.
:<pre>
;$ rpm -ivh
;kernel-lustre-smp-<ver> \
;kernel-ib-<ver> \
;lustre-modules-<ver> \
;lustre-ldiskfs-<ver>
</pre>


''The MDT must be upgraded before the OSTs are upgraded.'' [[summarize steps in a sentence]]
:b. ''Upgrade the utilities/userspace packages.'' For example:


1. Shut down lconf failover.
:<pre>
;$ rpm -Uvh lustre-<ver>
</pre>


2. Install the new modules.
:c. ''If a new e2fsprogs package is available, upgrade it.'' For example:


3. Run tunefs.lustre.
:<pre>
;$ rpm -Uvh e2fsprogs-<ver>
</pre>


4. Mount startup.
:There may or may not be a new ''e2fsprogs'' package with a Lustre upgrade. The ''e2fsprogs'' release schedule is independent of Lustre releases.


''A Lustre upgrade can be done across a failover pair.'' - [[In a failover pair, do the backup server first]]
:d. (Optional) ''If you want to add optional packages to your Lustre system, install them now.''


1. On the backup server, install the new modules.
3. ''Shut down the file system.'' Shut down the components in this order: clients, then the MDT, then OSTs. Unmounting a block device causes Lustre to be shut down on that node.


2. Shut down lconf failover.
:a. Unmount the clients. On each client node, run:


3. On the new server, run tunefs.lustre.
:<pre>
;umount <mount point>
</pre>


4. On the new server, mount startup.
:b. Unmount the MDT. On the MDS node, run:


5. On the primary server, install the new modules.
:<pre>
;umount <mount point>
</pre>


== Upgrade procedures ==
:c. Unmount the OSTs (be sure to unmount all OSTs). On each OSS node, run:


=== Upgrading Clients ===
:<pre>
;umount <mount point>
</pre>


* Starting Clients - [[if upgraded client only would need to know how to start a new client with an old MDT - says use old command - relevant depending on which machines you start with  - if you start with clients you need to know this information. ]][[Need instructions on upgrading a client]][[see initial client install ch 3 - intermingled - discuss with Sheila - something missing]]
4. ''Unload the old Lustre modules'' by either:
* Rebooting the node
:- OR -
* Removing the Lustre modules manually. Run ''lustre_rmmod'' several times and use ''lsmod'' to check the currently loaded modules.


You can start a new client with an old MDT by using the old format of the client
5. ''Start the upgraded file system.'' Start the components in this order: OSTs, then the MDT, then clients.
mount command:
:a. ''Mount the OSTs'' (be sure to mount all OSTs). On each OSS node, run:


client# mount -t lustre <mdtnid>:/<mdtname>/client <mountpoint>
:<pre>
;mount -t lustre <block device name> <mount point>
</pre>


You can start a new client with an upgraded MDT by using the new format and
:b. ''Mount the MDT.'' On the MDS node, run:
pointing it at the MGS, not the MDT (for co-located MDT/MGS, this is the same):


client# mount -t lustre <mgsnid>:/<fsname> <mountpoint>
:<pre>
;mount -t lustre <block device name> <mount point>
</pre>


Old clients always use the old format of the mount command, regardless of whether
:c. ''Mount the file system on the clients.'' On each client node, run:
the MDT has been upgraded or not.
 
:<pre>
;mount -t lustre <MGS node>&#58;/<fsname> <mount point>
</pre>
 
If you have a problem upgrading Lustre, contact us by [[Reporting Bugs|submitting a bug]] to our bug tracker [https://bugzilla.lustre.org/query.cgi Bugzilla].


=== Performing a Rolling Upgrade ===
=== Performing a Rolling Upgrade ===
This procedure describes a rolling upgrade in which one Lustre component (server or client) is upgraded and restarted at a time while the file system is running. If you want to upgrade the complete Lustre file system or multiple components at a time, requiring a file system shutdown, see [[#Performing a Complete File System Upgrade|Performing a Complete File System Upgrade]].


=== Upgrading Lustre 1.8.x to the Next Minor Version ===
----
 
'''''Note:''''' If the Lustre component to be upgraded is an OSS in a failover pair, follow these special upgrade steps to minimize downtime:
 
1. ''Fail over the server to its peer server'', so the file system remains available.
 
2. ''Install the Lustre 1.8.x packages on the idle server.''
 
3. ''Unload the old Lustre modules on the idle server'' by either:
:* Rebooting the node.
:- OR -
:* Removing the Lustre modules manually by running the ''lustre_rmmod'' command several times and checking the currently loaded modules with the ''lsmod'' command.
 
4. ''Fail back services to the now upgraded server.''
 
5. ''Repeat Steps 1 to 4 on the peer server.'' This limits the outage (per OSS) to a single server for as long as it takes to fail over.
 
----
To perform a rolling upgrade:
 
1. ''Make a complete, restorable file system backup before upgrading Lustre.''
 
2. ''Install the 1.8.x packages on the Lustre component (server or client).'' For help determining where to install a specific package, see [[Lustre Packages]].
 
:a. ''Install the kernel, modules and ''ldiskfs'' packages.'' For example:
 
:<pre>
;$ rpm -ivh
;kernel-lustre-smp-<ver> \
;kernel-ib-<ver> \
;lustre-modules-<ver> \
;lustre-ldiskfs-<ver>
</pre>
 
:b. ''Upgrade the utilities/userspace packages.'' For example:
 
:<pre>
;$ rpm -Uvh lustre-<ver>
</pre>
 
:c. ''If a new ''e2fsprogs'' package is available, upgrade it.'' For example:
 
:<pre>
;$ rpm -Uvh e2fsprogs-<ver>
</pre>
 
There may or may not be a new ''e2fsprogs'' package with a Lustre upgrade. The ''e2fsprogs'' release schedule is independent of Lustre releases.
 
:d. (''Optional) If you want to add optional packages to your Lustre system, install them now.''
 
3. ''Unload the old Lustre modules'' by either:
:* Rebooting the node
:- OR -
:* Removing the Lustre modules manually. Run ''lustre_rmmod'' several times and use ''lsmod'' to check the currently-loaded modules.
 
4. ''If the upgraded component is a server, fail back services to the new server.''
 
If you have a problem upgrading Lustre, contact us via the [https://bugzilla.lustre.org/query.cgi Bugzilla] bug tracker.
 
== Upgrading Lustre 1.8.x to the Next Minor Version ==


To upgrade Lustre 1.8.x to the next minor version, for example, Lustre 1.8.0.1 > 1.8.x,
To upgrade Lustre 1.8.x to the next minor version, for example, Lustre 1.8.0.1 > 1.8.x,
Line 116: Line 197:
* To upgrade one Lustre component (server or client) at a time, while the file system is running, see [[#Performing_a_Rolling_Upgrade|Performing a Rolling Upgrade]]
* To upgrade one Lustre component (server or client) at a time, while the file system is running, see [[#Performing_a_Rolling_Upgrade|Performing a Rolling Upgrade]]


== Downgrading ==
== Downgrading from Lustre 1.8.x ==
See Chapter 13 in the [[Lustre_Documentation|''Lustre Operations Manual'']] for information about downgrading from Lustre 1.8.x to Lustre 1.6.x
For Lustre 1.8.x, the following downgrades are supported:
* If you upgraded from Lustre 1.6.x > 1.8.x, you can downgrade to version 1.6.x.
* If you upgraded from one minor version to the next (for example, Lustre 1.8.0 > 1.8.x), you can downgrade to the earlier minor version.
 
For a procedure to downgrade from Lustre 1.8.x to Lustre 1.6.x, see [http://wiki.lustre.org/manual/LustreManual18_HTML/UpgradingLustre.html#50651255_38760 Section 13.5: ''Downgrading from Lustre 1.8.x to Lustre 1.6.x''] in the [http://wiki.lustre.org/manual/LustreManual18_HTML/index.html ''Lustre Operations Manual''].
 
'''''Caution:''''' A new installation of Lustre 1.8.x is not guaranteed to be downgradable to an earlier Lustre version.

Latest revision as of 10:23, 20 January 2011

(Updated: Dec 2009)

This section provides information about supported upgrades, upgrade paths, and interoperability of nodes with different versions of Lustre™ installed. It also describes procedures for upgrading your Lustre file system to a new version of Lustre. For information about available releases, see Lustre Release Information.

Supported Upgrades

For Lustre 1.8.x, the following upgrades are supported:

  • Lustre 1.6.x (latest version) to Lustre 1.8.x (latest version).
  • Lustre 1.8.x (any minor version) to Lustre 1.8.x (latest version).

Lustre Component Interoperability

Lustre interoperability enables 1.8.x servers ("new" servers) to work with 1.6.x clients ("old" clients), 1.6.x servers ("old" servers) to work with 1.8.x clients ("new" clients), and "mixed" environments with both 1.6.x and 1.8.x servers. For example, half of each OSS failover pair could be upgraded to enable a quick reversion to 1.6 by powering down the 1.8 servers.

The table below describes the interoperability between clients, OSTs, and MDTs with different versions of Lustre installed.

Lustre Component Interoperability with Other Lustre Components
Clients &#149; Old, live clients can communicate with old, new, or a mix of servers.
&#149; Old clients can start up using old, new, or a mix of servers.
&#149; New clients can start up using old, new, or a mix of servers.
Note: Old clients cannot mount a file system that was created by a new MDT.
OSTs &#149; Old OSTs can communicate with new clients/MDTs.
&#149; New OSTs can only be started after the MGS has been started (typically after the MDT has been upgraded).
MDTs &#149; An old MDT can communicate with new clients.
&#149; A new co-located MGS/MDT can be started at any point.
&#149; A new non-co-located MDT can be started after the MGS starts.

Upgrading Lustre 1.6.x to Lustre 1.8.x

Two Lustre upgrade paths are supported to meet the upgrade requirements of different Lustre environments.

Note: If you upgrade some Lustre components to 1.8.x but not others (such as running 1.8 clients in a file system with 1.6 OSTs), and run a mixed environment, you may see one or more warnings similar to this: 

LustreError: 3877:0:(socklnd_cb.c:2228:ksocknal_recv_hello())
Unknown protocol version (2.x expected) from 192.168.2.43

This warning is given when the 1.6 and 1.8 components use different protocols. It can be safely ignored because the Lustre components negotiate a common protocol. In this example, the 1.8 clients fall back to use the 1.6 protocol with the 1.6 OSTs.

Performing a Complete File System Upgrade

This procedure describes a complete file system upgrade in which 1.8.x Lustre packages are installed on multiple 1.6.x servers and clients, requiring a file system shut down. If you want to upgrade one Lustre component at a time and avoid the shutdown, see Performing a Rolling Upgrade.

Tip: In a Lustre upgrade, the package install and file system unmount steps are reversible; you can do either step first. To minimize downtime, this procedure first performs the 1.8.x package installation, and then unmounts the file system.

1. Make a complete, restorable file system backup before upgrading Lustre.

2. Install the 1.8.x packages on the Lustre servers and/or clients. Some or all servers can be upgraded. Some or all clients can be upgraded. For help determining where to install a specific package, see Lustre Packages.

a. Install the kernel, modules and ldiskfs packages. For example:


$ rpm -ivh


kernel-lustre-smp-<ver> \


kernel-ib-<ver> \


lustre-modules-<ver> \


lustre-ldiskfs-<ver>



b. Upgrade the utilities/userspace packages. For example:




$ rpm -Uvh lustre-<ver>



c. If a new e2fsprogs package is available, upgrade it. For example:




$ rpm -Uvh e2fsprogs-<ver>



There may or may not be a new e2fsprogs package with a Lustre upgrade. The e2fsprogs release schedule is independent of Lustre releases.


d. (Optional) If you want to add optional packages to your Lustre system, install them now.


3. Shut down the file system. Shut down the components in this order: clients, then the MDT, then OSTs. Unmounting a block device causes Lustre to be shut down on that node.



a. Unmount the clients. On each client node, run:




umount <mount point>



b. Unmount the MDT. On the MDS node, run:




umount <mount point>



c. Unmount the OSTs (be sure to unmount all OSTs). On each OSS node, run:




umount <mount point>



4. Unload the old Lustre modules by either:



  • Rebooting the node


- OR -


  • Removing the Lustre modules manually. Run lustre_rmmod several times and use lsmod to check the currently loaded modules.


5. Start the upgraded file system. Start the components in this order: OSTs, then the MDT, then clients.



a. Mount the OSTs (be sure to mount all OSTs). On each OSS node, run:




mount -t lustre <block device name> <mount point>



b. Mount the MDT. On the MDS node, run:




mount -t lustre <block device name> <mount point>



c. Mount the file system on the clients. On each client node, run:




mount -t lustre <MGS node>:/<fsname> <mount point>



If you have a problem upgrading Lustre, contact us by submitting a bug to our bug tracker Bugzilla.



Performing a Rolling Upgrade


This procedure describes a rolling upgrade in which one Lustre component (server or client) is upgraded and restarted at a time while the file system is running. If you want to upgrade the complete Lustre file system or multiple components at a time, requiring a file system shutdown, see Performing a Complete File System Upgrade.





Note: If the Lustre component to be upgraded is an OSS in a failover pair, follow these special upgrade steps to minimize downtime:

1. Fail over the server to its peer server, so the file system remains available.

2. Install the Lustre 1.8.x packages on the idle server.

3. Unload the old Lustre modules on the idle server by either:



  • Rebooting the node.


- OR -

  • Removing the Lustre modules manually by running the lustre_rmmod command several times and checking the currently loaded modules with the lsmod command.


4. Fail back services to the now upgraded server.

5. Repeat Steps 1 to 4 on the peer server. This limits the outage (per OSS) to a single server for as long as it takes to fail over.





To perform a rolling upgrade:

1. Make a complete, restorable file system backup before upgrading Lustre.

2. Install the 1.8.x packages on the Lustre component (server or client). For help determining where to install a specific package, see Lustre Packages.



a. Install the kernel, modules and ldiskfs packages. For example:




$ rpm -ivh


kernel-lustre-smp-<ver> \


kernel-ib-<ver> \


lustre-modules-<ver> \


lustre-ldiskfs-<ver>



b. Upgrade the utilities/userspace packages. For example:




$ rpm -Uvh lustre-<ver>



c. If a new e2fsprogs package is available, upgrade it. For example:




$ rpm -Uvh e2fsprogs-<ver>



There may or may not be a new e2fsprogs package with a Lustre upgrade. The e2fsprogs release schedule is independent of Lustre releases.



d. (Optional) If you want to add optional packages to your Lustre system, install them now.


3. Unload the old Lustre modules by either:



  • Rebooting the node


- OR -

  • Removing the Lustre modules manually. Run lustre_rmmod several times and use lsmod to check the currently-loaded modules.


4. If the upgraded component is a server, fail back services to the new server.

If you have a problem upgrading Lustre, contact us via the Bugzilla bug tracker.



Upgrading Lustre 1.8.x to the Next Minor Version


To upgrade Lustre 1.8.x to the next minor version, for example, Lustre 1.8.0.1 > 1.8.x,
follow these procedures:




Downgrading from Lustre 1.8.x


For Lustre 1.8.x, the following downgrades are supported:



  • If you upgraded from Lustre 1.6.x > 1.8.x, you can downgrade to version 1.6.x.
    • 

  • If you upgraded from one minor version to the next (for example, Lustre 1.8.0 > 1.8.x), you can downgrade to the earlier minor version.


For a procedure to downgrade from Lustre 1.8.x to Lustre 1.6.x, see Section 13.5: Downgrading from Lustre 1.8.x to Lustre 1.6.x in the Lustre Operations Manual.

Caution: A new installation of Lustre 1.8.x is not guaranteed to be downgradable to an earlier Lustre version.



Retrieved from "http://wiki.old.lustre.org/index.php?title=Upgrading_to_a_New_Version_of_Lustre&oldid=12219"