List:Cluster« Previous MessageNext Message »
From:Jonas Oreland Date:February 24 2011 8:29am
Subject:Re: increasing memory usage after restart of datanodes
View as plain text  
On 02/24/11 09:24, Hendrik Woltersdorf wrote:
> 
> Hi,
> 
> I get _very_ different numbers of the used data memory before and after a
> node restart in a cluster.
> 
> Before:
> 
> Node 10: Data usage is 7%
> Node 10: Index usage is 8%
> Node 11: Data usage is 7%
> Node 11: Index usage is 8%
> 
> After restart of node 11:
> 
> Node 10: Data usage is 6%
> Node 10: Index usage is 6%
> Node 11: Data usage is 20%
> Node 11: Index usage is 6%
> 
> After restart of node 10:
> 
> Node 10: Data usage is 21%
> Node 10: Index usage is 6%
> Node 11: Data usage is 20%
> Node 11: Index usage is 6%
> 
> I am using: Server version: 5.1.51-ndb-7.1.9a-cluster-gpl-log MySQL Cluster
> Server (GPL) (64bit) on CentOS release 5.2.
> The numbers from the management server log are the same as in table
> ndbinfo.memoryusage. A difference of one or two percent downward might be
> caused by cleaning jobs, deleting old data during the night (from 7% to
> 6%).
> 
> The cluster was up and running for 21 days. That was the day when I did an
> upgrade from version 7.0.13 to 7.1.9a.
> 
> The amount of data in the tables (without indexes) is approximately 1,1GB,
> the data memory 10GB.
> So 20% seems more reasonable then 6%.
> 
> My questions:
> 
> - Are the numbers before the restarts wrong?
> - If yes, is this a known bug?
> - If no, how can memory usage increase so much because of a restart?

most likely the numbers before restart were wrong...there has been a couple of bug-fixes
in that area "lately" (last 6 month or so)...but don't have numbers at to of head

> - What can I do to prevent running out of memory without knowing it?

there are various way to monitor memory usage,
e.g in 7.1 the ndbinfo-database is added...
  or ndb_log_events...

maybe look at
http://www.clusterdb.com/mysql-cluster/monitoring-mysql-cluster-with-mysql-enterprise-monitor/

/Jonas

> 
> regards
> 
> Hendrik Woltersdorf
> XCOM AG
> Email: Hendrik.Woltersdorf@stripped
> 
> 
> *** XCOM AG Legal Disclaimer ***
> 
> Diese E-Mail einschliesslich ihrer Anhaenge ist vertraulich und ist allein
> für den Gebrauch durch den vorgesehenen Empfaenger bestimmt. Dritten ist
> das Lesen, Verteilen oder Weiterleiten dieser E-Mail untersagt. Wir bitten,
> eine fehlgeleitete E-Mail unverzueglich vollstaendig zu loeschen und uns
> eine Nachricht zukommen zu lassen.
> 
> This email may contain material that is confidential and for the sole use
> of the intended recipient. Any review, distribution by others or forwarding
> without express permission is strictly prohibited. If you are not the
> intended recipient, please contact the sender and delete all copies.
> 
> Hauptsitz: Bahnstrasse 37, D-47877 Willich, USt-IdNr.: DE 812 885 664
> Kommunikation: Telefon +49 2154 9209-70, Telefax +49 2154 9209-900,
> www.xcom.de
> Handelsregister: Amtsgericht Krefeld, HRB 10340
> Vorstand: Matthias Albrecht, Dirk Franzmeyer, Dr. Rainer Fuchs, Marco
> Marty, Dirk Werner
> Vorsitzender des Aufsichtsrates: Stefan H. Tarach

Thread
increasing memory usage after restart of datanodesHendrik Woltersdorf24 Feb
  • Re: increasing memory usage after restart of datanodesJonas Oreland24 Feb
    • Antwort: Re: increasing memory usage after restart of datanodesHendrik Woltersdorf24 Feb