MySQL Forums
Forum List  »  NDB clusters

Re: Forcing Nodegroups
Posted by: Chris Studt
Date: March 15, 2005 03:21PM

The memory figures were from top, but after checking some caching discrepancies, we determined that the same load was on each storage node. Sorry about that...

As for your math, that all works out the way I would expect, but as you can see in my original post, both storage nodes were showing up in the management console as:

id=1 @192.168.0.117 (Version: 4.1.10, Nodegroup: 0, Master)
id=2 @192.168.0.118 (Version: 4.1.10, Nodegroup: 0)

(2 nodes within 1 nodegroup) as opposed to what I would expect:

id=1 @192.168.0.117 (Version: 4.1.10, Nodegroup: 0, Master)
id=2 @192.168.0.118 (Version: 4.1.10, Nodegroup: 1, Master)

(2 node groups of one node each)

We've done some further testing and found that if we bring down one node, the cluster continues to function as normal (as it should), so it appears that "1 node per replica" rule does work out. My concern is, is the management console displaying incorrect status of the nodegroup or is there something I'm missing?

Options: ReplyQuote


Subject
Views
Written By
Posted
3247
March 09, 2005 06:34PM
1889
March 11, 2005 08:55AM
Re: Forcing Nodegroups
2054
March 15, 2005 03:21PM
2014
March 15, 2005 06:46PM


Sorry, you can't reply to this topic. It has been closed.

Content reproduced on this site is the property of the respective copyright holders. It is not reviewed in advance by Oracle and does not necessarily represent the opinion of Oracle or any other party.