- Community Home
- >
- Storage
- >
- Midrange and Enterprise Storage
- >
- StoreVirtual Storage
- >
- Re: Node failure in P4000 SAN cluster
Categories
Company
Local Language
Forums
Discussions
Forums
- Data Protection and Retention
- Entry Storage Systems
- Legacy
- Midrange and Enterprise Storage
- Storage Networking
- HPE Nimble Storage
Discussions
Discussions
Discussions
Forums
Forums
Discussions
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
- BladeSystem Infrastructure and Application Solutions
- Appliance Servers
- Alpha Servers
- BackOffice Products
- Internet Products
- HPE 9000 and HPE e3000 Servers
- Networking
- Netservers
- Secure OS Software for Linux
- Server Management (Insight Manager 7)
- Windows Server 2003
- Operating System - Tru64 Unix
- ProLiant Deployment and Provisioning
- Linux-Based Community / Regional
- Microsoft System Center Integration
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Community
Resources
Forums
Blogs
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО11-15-2009 11:32 AM
тАО11-15-2009 11:32 AM
Node failure in P4000 SAN cluster
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО11-15-2009 05:00 PM
тАО11-15-2009 05:00 PM
Re: Node failure in P4000 SAN cluster
You can set the # of copies to more then 2, but it cannot be more than the # of nodes in the cluster.
Example, 4 nodes in the cluster, you can have 4 copies, and it will keep 1 copy on each node.
Tom
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО11-15-2009 05:12 PM
тАО11-15-2009 05:12 PM
Re: Node failure in P4000 SAN cluster
What I was looking for is how many nodes can fail. And it turns out that even with Raid 2 you can lose 2 non concurrent nodes in a 4 node cluster and with Raid 3 and 4 you can lose concurrent nodes.
Jamie
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО11-15-2009 09:40 PM
тАО11-15-2009 09:40 PM
Re: Node failure in P4000 SAN cluster
By using Network Raid 2 in a two node cluster it will depend on which nodes fails for the volume to be available.
Another thing to take into consideration is the Quorum - it's based on majority node set - so if you loose 2 nodes in a 4 node cluster, you won't have quorum, and the whole cluster is down.
This can be solved by implementing failover managers.
Regards
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО11-16-2009 06:21 AM
тАО11-16-2009 06:21 AM
Re: Node failure in P4000 SAN cluster
If network raid is set to 3 or 4, not 2, shouldn't you be able to lose 2 nodes and still be able to access the raided volumes?
Jamie
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО11-17-2009 09:14 AM
тАО11-17-2009 09:14 AM
Re: Node failure in P4000 SAN cluster
If you have a 4 node cluster and your volumes are 3 way replicated then you should be able to lose two nodes assuming that you have a failover manager running. This will enable you to run five managers and when the second node fails you will still have three managers running thus maintaining quorum and the volumes will still be accessible.
Mike.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО11-17-2009 07:44 PM
тАО11-17-2009 07:44 PM
Re: Node failure in P4000 SAN cluster
Also remember that if there is a 4 node multi-site cluster, then you can lose a whole site, or two nodes. And you only have to have 2-Way replication for that.
Then there is the case where you lose non-consecutive nodes in a multinode cluster and should be safe. Again with only 2-Way replication. Take a look again at how the blocks are written across the nodes and that should show you how you can lose multiple nodes as long as they are not consecutive.
Hope this helps show that the real answer is it depends. Also like others have stated make sure that you maintain quorum as well. Losing quorum doesn't cause data loss, it just causes the cluster to shut down and access to volumes is stopped.