[Ocfs2-users] Question about increasing node slots

[Tim Lank]

No problem.  :-)

The main points of having a RAC cluster as I understand it are
availability and scalability on low-cost systems.  Shouldn't ocfs2 have
the ability to perform online expansion like this?  I know that Red Hat's
GFS can add journals to accomodate new nodes while the filesystems are
online (gfs_jadd -j Number MountPoint).

What is the downside of installing a starter two-node cluster with ocfs2
filesystems configured for say 8 node slots?  The documentation seems to
suggest that you not do that.


> You are absolutely right. Sorry about that. I think I didn't have enough
> caffeine this morning ;-).
>
> The procedure is when you have available slots and just want to add a
> node to the cluster.
>
> Any change to the superblock needs to have the partition offline.
>
> You don't need to umount to upgrade database/clusterware.
>
> In some situations, you need to upgrade ocfs2 in sync, meaning, all
> nodes at once. Other times, you can run different versions in the same
> cluster until you get all nodes upgraded. That is documented in the FAQ.
>
> If you plan to add more nodes later, it would be a good idea to have
> additional slots defined to reduce cluster outage.
>
> About having it done online (with partitions mounted), I don't think we
> will have it.
>
>
> Regards,
>
> Marcos Eduardo Matsunaga
>
> Oracle USA
> Linux Engineering
>
>
>
>
>
> Tim Lank wrote:
>> So since tunefs.ocfs2 is the tool that actually does all of the work to
>> add more slots and the tunefs.ocfs2 man page states:
>>
>> DESCRIPTION
>>       tunefs.ocfs2  is  used to adjust OCFS2 file system parameters on
>> disk.
>>       In order to prevent data  loss,  tunefs.ocfs2  will  not  perform
>> any
>>       action  on  the  specified  device if it is mounted on any node in
>> the
>>       cluster.  This tool requires the O2CB cluster to be online.
>>
>> I return to my original question:
>> "Is it true that we will need to umount these filesystems for the
>> upgrade
>> (i.e. Database and Clusterware also)?"
>>
>> Since our cluster is running entirely on ocfs2 filesystems, this will
>> cause a cluster outage.
>>
>> Is there a way to do this while the filesystems are mounted?  If not,
>> are
>> there plans for allowing a node count increase while the fileystems are
>> mounted, and if so, when?
>>
>>
>>
>>> The console does that (You can use the console to add a new node).
>>> tunefs.ocfs2 is actually the tool that will change the superblock to
>>> add
>>> more slots (see the man pages) and it is called by the console (a more
>>> user friendly interface) to perform the action.
>>>
>>> o2cb_ctl only defines the new node(s) on the existing nodes that are
>>> running.
>>>
>>> Regards,
>>>
>>> Marcos Eduardo Matsunaga
>>>
>>> Oracle USA
>>> Linux Engineering
>>>
>>>
>>>
>>>
>>>
>>> Tim Lank wrote:
>>>
>>>> So does the o2cb_ctl command touch the ocfs2 filesystem superblock and
>>>> increase the node slot value in this example?
>>>>
>>>>
>>>>
>>>>
>>>>>> From
>>>>>> http://oss.oracle.com/projects/ocfs2/dist/documentation/ocfs2_faq.html
>>>>>>
>>>>>>
>>>>> 19 - How do I add a new node to an online cluster?
>>>>> You can use the console to add a new node. However, you will need to
>>>>> explicitly add the new node on all the online nodes. That is, adding
>>>>> on
>>>>> one node and propagating to the other nodes is not sufficient. If the
>>>>> operation fails, it will most likely be due to bug#741
>>>>> <http://oss.oracle.com/bugzilla/show_bug.cgi?id=741>. In that case,
>>>>> you
>>>>> can use the o2cb_ctl utility on all online nodes as follows:
>>>>>
>>>>> 	# o2cb_ctl -C -i -n NODENAME -t node -a number=NODENUM -a
>>>>> ip_address=IPADDR -a ip_port=IPPORT -a cluster=CLUSTERNAME
>>>>>
>>>>> Ensure the node is added both in /etc/ocfs2/cluster.conf and in
>>>>> /config/cluster/CLUSTERNAME/node on all online nodes. You can then
>>>>> simply copy the cluster.conf to the new (still offline) node as well
>>>>> as
>>>>> other offline nodes. At the end, ensure that cluster.conf is
>>>>> consistent
>>>>> on all the nodes.
>>>>>
>>>>> 20 - How do I add a new node to an offline cluster?
>>>>>
>>>>> You can either use the console or use o2cb_ctl or simply hand edit
>>>>> cluster.conf. Then either use the console to propagate it to all
>>>>> nodes
>>>>> or hand copy using scp or any other tool. The o2cb_ctl command to do
>>>>> the
>>>>> same is:
>>>>>
>>>>>         # o2cb_ctl -C -n NODENAME -t node -a number=NODENUM -a
>>>>> ip_address=IPADDR -a ip_port=IPPORT -a cluster=CLUSTERNAME
>>>>>
>>>>> Notice the "-i" argument is not required as the cluster is not
>>>>> online.
>>>>>
>>>>>
>>>>> Regards,
>>>>>
>>>>> Marcos Eduardo Matsunaga
>>>>>
>>>>> Oracle USA
>>>>> Linux Engineering
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> Tim Lank wrote:
>>>>>
>>>>>
>>>>>> We have a test 10gR2 RAC cluster using ocfs2 filesystems for the
>>>>>> Clusterware files and the Database files.
>>>>>>
>>>>>> We need to increase the node slots to accomodate new RAC nodes.  Is
>>>>>> it
>>>>>> true that we will need to umount these filesystems for the upgrade
>>>>>> (i.e.
>>>>>> Database and Clusterware also)?
>>>>>>
>>>>>> We are planning to use the following command format to perform the
>>>>>> node
>>>>>> slot increase:
>>>>>>
>>>>>> # tunefs.ocfs2 �N 3 /dev/mapper/mpath1p1
>>>>>>
>>>>>> _______________________________________________
>>>>>> Ocfs2-users mailing list
>>>>>> Ocfs2-users at oss.oracle.com
>>>>>> http://oss.oracle.com/mailman/listinfo/ocfs2-users
>>>>>>
>>>>>>
>>>>>>
>>>>
>>
>>
>