StoreVirtual Storage
1748129 Members
3654 Online
108758 Solutions
New Discussion юеВ

Re: Ad FOM as qourum withnes betwen to p4300

 
gortex78
Occasional Contributor

Ad FOM as qourum withnes betwen to p4300

Hi,

We have 1 P4300 starter kits.
We learnd the hard way that we need a FOM server so that if 1 node fails the other takes over.

We have instaleld the FOM as virtual machine that dos not use the P4300 storage.

Now for the milion dollar question is how do we configre the nodes to use the FOM as a witnes. Al guides we found are on installing the FOM virtual server. We canot find any documention on hov we sholud configure the nodes to use the FOM as a withnes in the CMC (9.0)

We are not so happy with the way HP are handeling the custemors that buy P4300.

We bought 24/7
Installations service
knoledge transfer

yet the nodes are nod clusterd. HP phone supports some times takes 2-3 days to call back and when they call back they say the only work with "break fix"
5 REPLIES 5
Jitun
HPE Pro

Re: Ad FOM as qourum withnes betwen to p4300

The Failover Manager is used in two-system and in Multi-Site SAN configurations to support automatic quorum management in the SAN.

Configuring a Failover Manager in the management group enables the SAN to have automated failover without requiring a regular manager running on a storage system.

A Failover Manager runs as a virtual machine on a VMware Server or on ESX and must be installed
on network hardware other than the storage systems in the SAN.

You can run FOM on VMware Player on any windows unit which has access to the SAN (same subnet)

Change the setting in the FOM Console with the IP Address of the SAN Subnet as shown with the Attachment.

Once done search the IP Address provided to the FOM in CMC.

Add it to the Management group which needs FOM.

Done.


I work for HPE
--------------------------------------------------------------
How to assign points? Click the KUDOS! star!

Accept or Kudo

Dominik Dienst
Visitor

Re: Ad FOM as qourum withnes betwen to p4300

Hi,

After you add the FOM to the management group, you can easily check the status of your management group and the managers with the CMC.
Click on your management group and check the count and the kind of managers. You should have 3 managers running, 2 normal manager (1 per node) and 1 failover manager.

Best regards,
Dominik


Andong Liu
New Member

Re: Ad FOM as qourum withnes betwen to p4300

Hi all,

I just want to append anther question, we had 2 P4300 cross 2 site and 1 FOM. The failover for each site is ok, but with 1 FOM and 1 vip there is tricky scenario like link broken between 2 sites, if vip can float to another site the esx server will lost storage.

I heard some people mentioned we can have 2 FOM and set up 2 VIP, then enable multiple pathing in ESX server so link broken scenario can be handled. Not sure the detail procedure for implementing this, any experts can help us out this issue?

Also if this is true story after link recovery how the data would be synchronized between two site?

Much appreciate for your help!

Cheers,
Andong
Paul Hutchings
Super Advisor

Re: Ad FOM as qourum withnes betwen to p4300

Andong, as I understand it you really have two options:

1) Put the FOM in the site you consider "primary" - if the link dies, this site keeps quorum.

2) Look at a 3rd physical site and put the FOM there and have links to sites 1 and 2 - doing this you also gain link redundancy.
Andong Liu
New Member

Re: Ad FOM as qourum withnes betwen to p4300

Hi Paul,

Thanks a lot for your solution. We considered these options, however:

1) if put FOM to primary site, when primary site down FOM will be down as well, in this case backup site will get trouble as well, this "both site down" scenario is the one we try to avoid.

2) unfortunately we have no third site or other servers to keep FOM running ~>_<~

The solution I knew from another post is:

"Enter the multi-site (also referred to as CAMPUS SAN): Each SITE must have a unique subnet, and you assign a unique VIP for each subnet. Then inside your servers or (ESX) hosts you assign both VIP addresses to the discovery tab. This will now allow your initiator to find multiple unique paths to your storage volumes, 1 at site A, and one at site B. If one of the paths fails the initiator immediately tries the other path. Since there is already an active responding VIP (you don't have to wait for the it to fail-over), you recover connectivity to your storage immediately, and your critical volumes stay online."

It sounds the right thing we should to do and HP should provide. Otherwise the dependance of FOM will be a disaster for lefthand solution. Just wondering anyone had tried above solution out and could you please give us some guide?

Much appreciate for help!

Cheers,
Andong