|
|||||||||||
|
Re: Clarification on --initial
From: Jon Stephens <jon(at)mysql.com>
Date: Fri Aug 17 2007 - 12:12:43 EDT
> On Thu, 2007-08-16 at 19:53 +0200, Philipp Taprogge wrote: >> Hi! >> >> Thus spake Al on 08/16/2007 06:58 PM: >>> Say I have two data nodes in my cluster that uses on-disk memory, and >>> I'd like to add two more data nodes to the cluster. When restarting >>> the cluster, I'll need to run ndbd --initial on all the data nodes, >>> not just the new ones, correct? >> I think not. If I get it right, starting a node with --initial will >> effectively wipe it clean, so in your above scenario, you'd have a brand new >> but empty four-node cluster. >> I guess that's not what you want. > > It's the only (current) way to add data nodes. start a fresh cluster and > restore from backup. > >> But I am getting a little confused by the whole tread... >> Perhaps one of the enlightened could explain, what /is/ a scenario where you'd >> use --initial? >> It now seems there isn't one and I wonder why keep the option if it's not >> dangerously misnamed but also superfluous? >> > > There is. > a) some configuration changes require --initial > b) when wanting to force a resync. Thanks to all who've participated in this discussion; I've just overhauled the Manual's description of ndbd --initial and when it ought to be used. Feel free to drop me a note if I've missed anything. cheers jon. -- Jon Stephens - jon@mysql.com Technical Writer - MySQL Documentation Team Mobile (Sweden): +46 (0) 73 677 39 93 (GMT +02.00) MySQL AB: www.mysql.com -- MySQL Cluster Mailing List For list archives: http://lists.mysql.com/cluster To unsubscribe: http://lists.mysql.com/cluster?unsub=lists@pantek.comReceived on Fri Aug 17 12:12:57 2007 This archive was generated by hypermail 2.1.8 : Sun Oct 07 2007 - 10:15:02 EDT |
||||||||||
|
|||||||||||