List:Cluster« Previous MessageNext Message »
From:Andrew Morgan Date:October 4 2012 5:33pm
Subject:RE: Problem with memcached
View as plain text  
Connection pooling should happen unless you've included it in your config file for your
mysqlds.

If you didn't start with as many mysqld slots in your config.ini but then increased the
number later then you need to perform a rolling restart. Begin by stopping the ndb_mgmds
and then restart them with the --initial option; you then need to restart each
ndbd/ndbmtd in turn so that they load the new config.

Regards, Andrew.

> -----Original Message-----
> From: Maxime Thébault [mailto:webmaster@stripped]
> Sent: 04 October 2012 18:21
> To: cluster@stripped
> Subject: Re: Problem with memcached
> 
> I flush all iptables :
> 
> iptables -F
> iptables -X
> iptables -t nat -F
> iptables -t nat -X
> iptables -t mangle -F
> iptables -t mangle -X
> iptables -P INPUT ACCEPT
> iptables -P OUTPUT ACCEPT
> iptables -P FORWARD ACCEPT
> 
> and restart all services. And i have the same problem. (with the two
> managers)
> 
> I don't understand...
> 
> Le 04/10/2012 19:00, Andrew Morgan a écrit :
> > It's ndb-cluster-connection-pool but this is a server option rather than a
> cluster-wide one.
> >
> > Try disabling the firewall on the hosts - those running data nodes,
> ndb_mgmd and mysqld/memcached's.
> >
> > Since your last change to config.ini have you restarted (with "--initial") your
> ndb_mgmds and (with "--reload") your data nodes?
> >
> > Andrew.
> >
> >> -----Original Message-----
> >> From: Maxime Thébault [mailto:webmaster@stripped]
> >> Sent: 04 October 2012 17:47
> >> To: Andrew Morgan
> >> Subject: Re: Problem with memcached
> >>
> >> Same problem :
> >>
> >> 2012-10-04 18:45:39 [MgmtSrvr] WARNING  -- Failed to allocate nodeid
> >> for API at XX.XX.XX.XX. Returned eror: 'No free node id found for
> >> mysqld(API).' - Repeated 14 times
> >>
> >> I think it's the option for allow 1 node to have multiple connect at
> >> manager but i can't find his name. Do you know it?
> >>
> >> thank you very much!
> >>
> >> Le 04/10/2012 18:38, Andrew Morgan a écrit :
> >>> Try running another mysqld process from the node that you're running
> >> memcached from. By default, I believe that the memcached plug-in uses
> >> 2 API connections to the cluster and so from your ndb_mgm output you
> >> should have enough.
> >>> Andrew.
> >>>
> >>>> -----Original Message-----
> >>>> From: Maxime Thébault [mailto:webmaster@stripped]
> >>>> Sent: 04 October 2012 17:36
> >>>> To: cluster@stripped
> >>>> Subject: Re: Problem with memcached
> >>>>
> >>>> I have a result with good value.
> >>>>
> >>>> I try to search variable for set limit of use for a node but i cant
> >>>> find it! (i see it on mysql website but impossible to refind!)
> >>>>
> >>>> I'd try increasing the number of connections per node as possible.
> >>>>
> >>>> Thank Andrew.
> >>>>
> >>>> Le 04/10/2012 18:32, Andrew Morgan a écrit :
> >>>>> Hi Maxime,
> >>>>>
> >>>>>     Try running "ndb_mgm -c 37.59.48.15:1186 -e show" from the
> >>>>> same host
> >>>> that you're running memcached from to see if you get the same
> >> response.
> >>>>> Regards, Andrew.
> >>>>>
> >>>>>> -----Original Message-----
> >>>>>> From: Maxime Thébault
> [mailto:webmaster@stripped]
> >>>>>> Sent: 04 October 2012 17:22
> >>>>>> To: cluster@stripped
> >>>>>> Subject: Re: Problem with memcached
> >>>>>>
> >>>>>> Thanks for your answer !
> >>>>>>
> >>>>>> I check in my ndb_1_cluster.log and i have this line :
> >>>>>> 2012-10-04 18:10:39 [MgmtSrvr] WARNING  -- Failed to
> allocate
> >>>>>> nodeid for API at 2.2.2.2. Returned eror: 'No free node id
> found
> >>>>>> for
> >> mysqld(API).'
> >>>>>> 2012-10-04 18:10:42 [MgmtSrvr] WARNING  -- Failed to
> allocate
> >>>>>> nodeid for API at 2.2.2.2. Returned eror: 'No free node id
> found
> >>>>>> for
> >> mysqld(API).'
> >>>>>>
> >>>>>> But when i run on my manager ndb_mgm -e show i have this :
> >>>>>>
> >>>>>> ndb_mgm> show
> >>>>>> Connected to Management Server at: localhost:1186 Cluster
> >>>>>> Configuration
> >>>>>> ---------------------
> >>>>>> [ndbd(NDB)]     2 node(s)
> >>>>>> id=3    @1.1.1.1  (mysql-5.5.22 ndb-7.2.6, Nodegroup: 0)
> >>>>>> id=4    @2.2.2.2  (mysql-5.5.22 ndb-7.2.6, Nodegroup: 0,
> Master)
> >>>>>>
> >>>>>> [ndb_mgmd(MGM)] 2 node(s)
> >>>>>> id=1    @3.3.3.3  (mysql-5.5.22 ndb-7.2.6)
> >>>>>> id=2    @4.4.4.4  (mysql-5.5.22 ndb-7.2.6)
> >>>>>>
> >>>>>> [mysqld(API)]   14 node(s)
> >>>>>> id=5    @2.2.2.2  (mysql-5.5.22 ndb-7.2.6)
> >>>>>> id=6    @1.1.1.1  (mysql-5.5.22 ndb-7.2.6)
> >>>>>> id=7 (not connected, accepting connect from any host)
> >>>>>> id=8 (not connected, accepting connect from any host)
> >>>>>> id=9 (not connected, accepting connect from any host)
> >>>>>> id=10 (not connected, accepting connect from any host)
> >>>>>> id=11 (not connected, accepting connect from any host)
> >>>>>> id=12 (not connected, accepting connect from any host)
> >>>>>> id=13 (not connected, accepting connect from any host)
> >>>>>> id=14 (not connected, accepting connect from any host)
> >>>>>> id=15 (not connected, accepting connect from any host)
> >>>>>> id=16 (not connected, accepting connect from any host)
> >>>>>> id=17 (not connected, accepting connect from any host)
> >>>>>> id=18 (not connected, accepting connect from any host)
> >>>>>>
> >>>>>> Twelve free nodes... I have 14 empty [MYSQLD] in my manager
> >>>>>> config
> >> file.
> >>>>>> I must specific option in this sections ?
> >>>>>>
> >>>>>> Maxime
> >>>>>>
> >>>>>> Le 04/10/2012 18:02, Andrew Morgan a écrit :
> >>>>>>> Hi Maxime, could you please check the Cluster log (it
> should be
> >>>>>>> located
> >>>>>> with the ndb_mgmd) to see if there's evidence of the
> memcached
> >>>>>> trying to connect? Otherwise, double check that the connect
> >>>>>> string is for the correct host and port for the ndb_mgmd and
> that
> >>>>>> there are no firewall rules that would block communication
> >>>>>> between these
> >> hosts.
> >>>>>>> Regards, Andrew.
> >>>>>>>
> >>>>>>> Andrew Morgan | MySQL Principal Product Manager
> >>>>>>> Mobile: +447833483595
> >>>>>>> Web: www.clusterdb.com
> >>>>>>> United Kingdom
> >>>>>>>
> >>>>>>> ORACLE Corporation UK Ltd is a company incorporated in
> England &
> >>>>>>> Wales
> >>>>>>> | Company Reg. No. 1782505 | Reg. office: Oracle
> Parkway, Thames
> >>>>>>> Valley Park, Reading RG6 1RA
> >>>>>>>
> >>>>>>>> -----Original Message-----
> >>>>>>>> From: Maxime Thébault
> [mailto:webmaster@stripped]
> >>>>>>>> Sent: 04 October 2012 09:56
> >>>>>>>> To: cluster@stripped
> >>>>>>>> Subject: Problem with memcached
> >>>>>>>>
> >>>>>>>> Hello,
> >>>>>>>>
> >>>>>>>> I try to configure memcached on my mysql cluster but
> i have an
> >> error.
> >>>>>>>> Result of ndb_mgm -e show :
> >>>>>>>> ndb_mgm -e show
> >>>>>>>> Connected to Management Server at: 3.3.3.3:1186
> Cluster
> >>>>>>>> Configuration
> >>>>>>>> ---------------------
> >>>>>>>> [ndbd(NDB)]     2 node(s)
> >>>>>>>> id=3    @1.1.1.1  (mysql-5.5.22 ndb-7.2.6,
> Nodegroup: 0)
> >>>>>>>> id=4    @2.2.2.2  (mysql-5.5.22 ndb-7.2.6,
> Nodegroup: 0, Master)
> >>>>>>>>
> >>>>>>>> [ndb_mgmd(MGM)] 2 node(s)
> >>>>>>>> id=1    @3.3.3.3  (mysql-5.5.22 ndb-7.2.6)
> >>>>>>>> id=2    @4.4.4.4  (mysql-5.5.22 ndb-7.2.6)
> >>>>>>>>
> >>>>>>>> [mysqld(API)]   14 node(s)
> >>>>>>>> id=5    @2.2.2.2  (mysql-5.5.22 ndb-7.2.6)
> >>>>>>>> id=6    @1.1.1.1  (mysql-5.5.22 ndb-7.2.6)
> >>>>>>>> id=7 (not connected, accepting connect from any
> host)
> >>>>>>>> id=8 (not connected, accepting connect from any
> host)
> >>>>>>>> id=9 (not connected, accepting connect from any
> host)
> >>>>>>>> id=10 (not connected, accepting connect from any
> host)
> >>>>>>>> id=11 (not connected, accepting connect from any
> host)
> >>>>>>>> id=12 (not connected, accepting connect from any
> host)
> >>>>>>>> id=13 (not connected, accepting connect from any
> host)
> >>>>>>>> id=14 (not connected, accepting connect from any
> host)
> >>>>>>>> id=15 (not connected, accepting connect from any
> host)
> >>>>>>>> id=16 (not connected, accepting connect from any
> host)
> >>>>>>>> id=17 (not connected, accepting connect from any
> host)
> >>>>>>>> id=18 (not connected, accepting connect from any
> host)
> >>>>>>>>
> >>>>>>>> But when i execute the command from 1.1.1.1 (api
> node) :
> >>>>>>>> memcached -E /usr/local/mysql/lib/ndb_engine.so -u
> mysql -e
> >>>>>>>> "connectstring=3.3.3.3:1186;role=db-only;debug=true"
> -vv -c 20
> >>>>>>>> (3.3.3.3 is the manager)
> >>>>>>>>
> >>>>>>>> The result is :
> >>>>>>>>
> >>>>>>>> main --> ndb_initialize()
> >>>>>>>> main --> connect_to_primary_cluster()
> >>>>>>>> 04-Oct-2012 10:54:42 CEST NDB Memcache
> 5.5.22-ndb-7.2.6
> started
> >>>>>>>> [NDB 7.2.6; MySQL 5.5.22] Contacting primary
> management server
> >>>>>>>> (37.59.48.15:1186) ...
> >>>>>>>> main --> ClusterConnectionPool::connect()
> FAILED.
> >>>>>>>> Could not connect to NDB.  Shutting down.
> >>>>>>>> main --> ndb_destroy()
> >>>>>>>> Failed to initialize instance. Error code: 255
> >>>>>>>>
> >>>>>>>> I try with localhost, without connectstring...
> >>>>>>>>
> >>>>>>>> Can you help me ?
> >>>>>>>>
> >>>>>>>> Many thanks !
> >>>>>>>>
> >>>>>>>>
> >>>>>>>>
> >>>>>>>>
> >>>>>>>> --
> >>>>>>>> MySQL Cluster Mailing List
> >>>>>>>> For list archives: http://lists.mysql.com/cluster
> >>>>>>>> To unsubscribe:    http://lists.mysql.com/cluster
> >>>>>>>>
> >>>>>> --
> >>>>>> MySQL Cluster Mailing List
> >>>>>> For list archives: http://lists.mysql.com/cluster
> >>>>>> To unsubscribe:    http://lists.mysql.com/cluster
> >>>>>>
> >>>>>
> >>>> --
> >>>> MySQL Cluster Mailing List
> >>>> For list archives: http://lists.mysql.com/cluster
> >>>> To unsubscribe:    http://lists.mysql.com/cluster
> >>>>
> >>>
> >>>
> >
> >
> >
> 
> 
> --
> MySQL Cluster Mailing List
> For list archives: http://lists.mysql.com/cluster
> To unsubscribe:    http://lists.mysql.com/cluster
> 
Thread
Problem with memcachedMaxime Thébault4 Oct
  • RE: Problem with memcachedAndrew Morgan4 Oct
    • Re: Problem with memcachedMaxime Thébault4 Oct
      • RE: Problem with memcachedAndrew Morgan4 Oct
        • Re: Problem with memcachedMaxime Thébault4 Oct
          • RE: Problem with memcachedAndrew Morgan4 Oct
            • Re: Problem with memcachedMaxime Thébault4 Oct
RE: Problem with memcachedAndrew Morgan4 Oct
  • Re: Problem with memcachedMaxime Thébault4 Oct
    • RE: Problem with memcachedAndrew Morgan4 Oct