Re: Impossible to recover from datanode down
Posted by:
Adam D
Date: August 29, 2006 05:38PM
Always when I restart my nodes I keep an eye on IO of both network and disk, generally they will be doing something if the node is starting up correctly.
Is this the case for you? Do you have graphs of Network IO and whats it up to?
Can you in your managment client run 'all status' and see what Phase the starting node is in.
On a normal startup of my nodes, with a DataMemory of 5700MB, It takes around 40 minutes. Previously I have experianced the 'stuck in phase X' issue, where another kill of the ndbd process and starting it again has fixed it.
Adam
Subject
Views
Written By
Posted
1775
August 16, 2006 04:10AM
1023
August 23, 2006 08:57PM
1125
August 29, 2006 03:30AM
Re: Impossible to recover from datanode down
1026
August 29, 2006 05:38PM
1031
August 30, 2006 04:48AM
992
August 30, 2006 05:57AM
949
August 31, 2006 05:41AM
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.