Quantcast
Channel: MySQL Forums - NDB clusters
Viewing all 1560 articles
Browse latest View live
↧

Why does ndb_mgmd consume 9GB RAM? (no replies)

November 9, 2012, 3:16 am
$
0
0
I have a cluster with 12 data nodes with 32 mysql clients and one management node.
The ndb_mgmd.exe consumes 9.5GB, but it's doing nothing. Why is it so high?
↧
Search

Getting Table got full error for "Disk Based Tables" (no replies)

November 9, 2012, 5:47 pm
$
0
0
Description:

I am trying to implement cluster for one of my client who has tables in 30 GB to 150 GB each and the total size of the DB is 1+ TB. They wanted to see how cluster might be suitable for their environment. They wanted mainly disk Based tables for this.

My config file

[ndb_mgmd]
Id=1
HostName=localhost
datadir=/dat/my_cluster/ndb_data
LogDestination=SYSLOG:facility=syslog;FILE:filename=/var/log/mgmd_1

[ndbd default]
NoOfReplicas=2
DataDir=/dat/my_cluster/ndb_data

DataMemory=2080M
IndexMemory=500M
MaxNoOfAttributes=10000
MaxNoOfOrderedIndexes=512

Created undo and tablespace from DB prompt instead of giving it through config file.

UNDO log files

NODEGROUP=0 INITIAL_SIZE=209715200 engine=ndbcluster;

TABLESPACE

USE LOGFILE GROUP ul_log EXTENT_SIZE 16777216 AUTOEXTEND_SIZE 8388608 ENGINE =NDB

DATA FILES

1) ADD DATAFILE 'tb1.dat' INITIAL_SIZE 8388608 ENGINE=ndb
2) ADD DATAFILE 'tb2.dat' INITIAL_SIZE 134217728 ENGINE=ndb;
3) ADD DATAFILE 'tb3.dat' INITIAL_SIZE 32212254720 ENGINE=ndb;

My doubts :

a) I am just loading 35 GB dump of a single partitioned table which is always getting table Full messages.

b) Is there a way to auto_extend a datafile from the initial size ?

c) Please help me out with proper variables that needs to be set for this disk based tables in ndbcluster.

I couldn't find lot of help from any docs so I am in need of some help terribly.

Thanks,
Santhinesh
↧
↧

What operations are delegated to the storage nodes? (no replies)

November 12, 2012, 12:39 am
$
0
0
I find MySQL cluster very interesting but is having a hard time finding information about what type of operations that are delegated to the data nodes. If I for instance perform operations like SUM, MAX etc are they performed on the data nodes or are all the data loaded to the MySQL server and the operations performed there?
↧

Exclude weekends between two dates (no replies)

November 12, 2012, 3:40 am
$
0
0
Good morning, this is my Trigger, i juste want to exclude all the saturdays/sundays between these two dates >> x = cf_802 and cf_860

ex1 : if delay(cf_865) < 3days
cf_802 = 16/11/2012 ( Next Friday )
cf_865 = 2days ( cf_865 )

Our will have this date :
cf_860 = 21/11/2012




ex2 : if Delay(cf_865 between 8day and 12 day )
cf_802 = 16/11/2012
cf_865 = 12days

output without weekends : 17 et 18 Non + 24 et 25 Non + 1 et 2 Non
will be >> 04/12/2012



CREATE TRIGGER Before_Update_TriggerAnalyse BEFORE Update ON vtiger_analysecf
FOR EACH ROW
begin
DECLARE x date;
SET x=( select cf_802 from vtiger_echantilloncf where echantillonid in (select echantillonid from vtiger_echantillon where code in (select cf_925 from vtiger_analysecf where cf_925 = NEW.cf_925)));
if NEW.cf_865 < 3 THEN SET NEW.cf_860=TIMESTAMPADD(DAY,1+NEW.cf_865,x);
ELSEIF NEW.cf_865 between '3' AND '8' THEN SET NEW.cf_860=TIMESTAMPADD(DAY,-1+NEW.cf_865,x);
ELSEIF NEW.cf_865 between '8' AND '12' THEN SET NEW.cf_860=TIMESTAMPADD(DAY,-2+NEW.cf_865,x);
ELSE SET NEW.cf_860=TIMESTAMPADD(DAY,-5+NEW.cf_865,x);
END IF;
END;
↧

NDB crash on mysqld node after an ALTER statement (no replies)

November 12, 2012, 8:36 am
$
0
0
Hi MySQL,

Version 7.2.8

I have had another crash after an ALTER statement.

One mysqld node replying with this after every query:

ERROR 1296 (HY000): Got error 4547 'RecordSpecification has overlapping offsets' from NDBCLUSTER

Another mysqld node core-dumped with error at end of this email.

This keeps happening to us. I have been trying and failing to make this product stable for several months now. I would really appreciate some feedback on why we keep having crashed, and whether Oracle/MySQL have any plans to try and make this system stable?

Regards,

Ben Clewett

121112 16:04:19 [Note] NDB Binlog: DISCOVER TABLE Event: REPL$ndb/fl_state
121112 16:04:19 [Note] NDB Binlog: logging ./ndb/fl_state (UPDATED,USE_WRITE)
16:04:19 UTC - mysqld got signal 11 ;
This could be because you hit a bug. It is also possible that this binary
or one of the libraries it was linked against is corrupt, improperly built,
or misconfigured. This error can also be caused by malfunctioning hardware.
We will try our best to scrape up some info that will hopefully help
diagnose the problem, but since we have already crashed,
something is definitely wrong and this may fail.

key_buffer_size=8388608
read_buffer_size=5242880
max_used_connections=35
max_threads=200
thread_count=33
connection_count=33
It is possible that mysqld could use up to
key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 2058485 K bytes of memory
Hope that's ok; if not, decrease some variables in the equation.

Thread pointer: 0x7f5a980009b0
Attempting backtrace. You can use the following information to find out
where mysqld died. If you see no messages after this, something went
terribly wrong...
stack_bottom = 7f5ab78dcea8 thread_stack 0x40000
/usr/sbin/mysqld(my_print_stacktrace+0x35)[0x81c7f5]
/usr/sbin/mysqld(handle_fatal_signal+0x403)[0x6e9383]
/lib64/libpthread.so.0(+0xfd00)[0x7f5ab7536d00]
/lib64/libc.so.6(+0x11abdf)[0x7f5ab6811bdf]
/usr/sbin/mysqld[0x84ced7]
/usr/sbin/mysqld(ndb_binlog_thread_func+0x1af4)[0x854ba4]
/lib64/libpthread.so.0(+0x7f05)[0x7f5ab752ef05]
/lib64/libc.so.6(clone+0x6d)[0x7f5ab67d210d]

Trying to get some variables.
Some pointers may be invalid and cause the dump to abort.
Query (0): is an invalid pointer
Connection ID (thread ID): 1
Status: NOT_KILLED

The manual page at http://dev.mysql.com/doc/mysql/en/crashing.html contains
information that should help you find out what is causing the crash.
↧
↧

Stored procedures executed on data nodes? (3 replies)

November 13, 2012, 5:42 am
$
0
0
Is it possible to have some equivalent of "stored procedures" (but written in C) that are executed on the data nodes but "registered on the MySQserver" as if they where normal stored procedure? I realize that one can run applications on the data nodes that use the C API to access the data node but I would like a way to invoke these "stored procedures" through the MySQL server from standard analytical software etc...
↧

Got temporary error 20016 'Query aborted due to node failure' from NDBCLUSTER (no replies)

November 14, 2012, 7:30 am
$
0
0
Strange issue on NDBCLUSTER.

same query runs on all clusters, with same configuration, it is just one cluster having this error

JFI ::
Version :: 5.5.19-ndb-7.2.4-gpl-log
all cluster has same OS / same package / same image / same versions
all data nodes and managmenet node is up, while it has produced this error.

error i have received is

Statement :: select count(*) from dim_time t inner join fact_message m on t.timeId = m.timeId inner join dim_country sc on m.senderCountryId=sc.countryId where t.date between '2012-11-01' and '2012-11-05';
ERROR 1296 (HY000): Got error -1 'Unknown error code' from NDBCLUSTER

Statement :: select count(*) from dim_time t LEFT join fact_message m on t.timeId = m.timeId LEFT join dim_country sc on m.senderCountryId=sc.countryId where t.date between '2012-11-01' and '2012-11-05';
ERROR 1297 (HY000): Got temporary error 20016 'Query aborted due to node failure' from NDBCLUSTER

if i do == select count(*) from fact_message; and select count(*) from dim_time; works fine.

have following error in API log - [ERROR] Got error 20016 when reading table './fact_message'

Starting thread because "http://forums.mysql.com/read.php?25,515068,515818#msg-515818&quot; is closed for response.

in order to fix issue i have to execute "SET ndb_join_pushdown = 0;" before query and error is gone.

It could be bug, not sure.

Thank you
↧

Disk based tables performance (no replies)

November 14, 2012, 12:28 pm
$
0
0
Hello,

I have the following doubt about disk data based tables in MySQL-Cluster 7.2.8:
I got the following tables:

Table 1: event (all in memory)

id int PK
timestamp date INDEX
...


Table 2: data (some columns in memory, some of them on disk)

event_id int PK
payload varchar ST MEMORY
extra1 varchar ST DISK
extra2 varchar ST DISK
extra3 varchar ST DISK
...


When I launch the following query it lasts 20 seconds and is ok (pushed join@1):

SELECT SQL_CALC_FOUND_ROWS event.*,data.payload FROM event,data WHERE ( event.id=data.event_id AND (((data.payload LIKE '%foo%' OR data.payload LIKE '%666f6f%'))) ) AND ( timestamp >='2012-11-10 00:00:00' ) ORDER BY timestamp DESC LIMIT 0, 11;

If I change the resultset from "data.payload" to "data.*" to retrieve all columns from data table the query never ends, so I guess that before checking the where clause it retrieve all columns from disk from all rows that matches the join. I dont understand this behaviour because all columns involved in the where clause are stored in memory, and maybe these ones should be used before fetching disk columns that are only used in the resultset.

Is this the normal behaviour?

Thanks,
↧
Search

install fails at mysql_install_db (no replies)

November 14, 2012, 1:26 pm
$
0
0
Hey all.
i am following the procedure here http://dev.mysql.com/doc/refman/5.5/en/mysql-cluster-install-linux-binary.html
everything seems to go ok until i get to the line that says scripts/mysql_install_db --user=mysql

i am greeted with this.
FATAL ERROR: Could not find mysqld

The following directories were searched:

/usr/libexec
/usr/sbin
/usr/bin

If you compiled from source, you need to run 'make install' to
copy the software into the correct location ready for operation.

If you are using a binary release, you must either be at the top
level of the extracted archive, or pass the --basedir option
pointing to that location.

this is a data node i am trying to install on.
Thanks for any help
↧
↧

Unexpected row-locks due to wrong index selection (no replies)

November 15, 2012, 5:34 am
$
0
0
Hi,

We observe a strange issue using NDB / Compound UNIQUE indexes.. When creating the following table:

create table t
(
C1 varchar(10) NOT NULL,
C2 VARCHAR(512) NOT NULL,
C3 int NOT NULL
)ENGINE=NDB;


And adding the following indexes:
ALTER TABLE t ADD UNIQUE idx_1 (c3, c2);
ALTER TABLE t ADD INDEX idx_2 (c1, c3);

And add the following data:
INSERT INTO t VALUES ('1', '11', 1);
INSERT INTO t VALUES ('2', '22', 1);
INSERT INTO t VALUES ('3', '33', 1);

And I perform the following from two mysql clients:
Client 1:
START TRANSACTION;
DELETE FROM t WHERE c1 = '1' AND c3 = 1;

Client 2:
START TRANSACTION;
DELETE FROM t WHERE c1 = '2' AND c3 = 1;

I get an "lock timeout exceeded" on the DELETE statement???? I would have expected NDB to use the secondary index (idx_2) to lock the correct rows.

Instead when I change the indexes as such:
ALTER TABLE t DROP INDEX idx_1;
ALTER TABLE t ADD INDEX idx_1 (c3, c2);

Or even as such:
ALTER TABLE t DROP INDEX idx_1;
ALTER TABLE t ADD UNIQUE idx_1 (c2, c3);

It all works as expected....

Is this a bug or expected behaviour? Does anyone know? To me it looks like a bug in the query-plan for the delete statement, incorrecly promoting the UNIQUE index to be used, even when the other index has all the values used in the WHERE clause present.

Hope you can provide some insight.

Regards,

//Ton
↧

CPU binding on a Hexacore CPU (no replies)

November 15, 2012, 9:41 am
$
0
0
I'm playing around with [ndbd]'s Threadconfig parameter for a hyperthreaded hexacore Intel CPU.
This is what I tried:

ThreadConfig=ldm={count=8,cpubind=4,5,6,7,8,9,10,11},main={cpubind=0},rep={cpubind=1},recv={cpubind=2},io={cpubind=3}

but when starting the data nodes, they complained about a bad config file.


Then I just tried this:
MaxNoOfExecutionThreads=8

which seems to work. But this way, I didn't bind specific threads to a CPU.

So what's the best configuration in my case?
↧

I need some insight about a new Cluster setup (no replies)

November 15, 2012, 2:08 pm
$
0
0
I am in the process of deploying a Load Balancer and I would like some advice. It is the first I will be deploying a MySQL Cluster so please tell me if I say anything that make no sense.

The traffic is not very high, but the requests are pretty slow, mostly on reads.
Databases have about 1GB of data, but not all databases are used completely.

I will be deploying on http://www.linode.com/

My plan:

• 1 Load Balancer
• 1 Manager on 512MB: Puppet + NDB Master
• 2 Web nodes on 512MB
• 2 NDB SQL nodes on 512MB, each connected to their Web node
• 2 NDB Data nodes on 1024MB

The idea is that I can later upgrade any of those, depending on their usage.

Do I need the SQL nodes on another server or they can be on the Web nodes? Do I need more RAM on the SQL nodes ?

I mean, I could do this:
• 2 Web + NDB SQL nodes on 1024MB
• 2 NDB Data nodes on 1024MB

Thank you very much for your time
↧

ndb_size does not match reality (no replies)

November 17, 2012, 5:21 am
$
0
0
Hi,

I'm using ndb_size.pl to assess how much memory would I need for one particular table. This table has 250MB in MyISAM. The result was:

Parameter Default 4.1 5.0 5.1
DataMemory (KB) 81920 1012256* 1012256* 989312*
NoOfOrderedIndexes 128 5 5 5
NoOfTables 128 6 6 6
IndexMemory (KB) 18432 172408* 26880* 26880*
NoOfUniqueHashIndexes 64 4 4 4
NoOfAttributes 1000 24 24 24
NoOfTriggers 768 41 41 41

Which seemed too high for me (datamemory = myisam * 4?). So I imported the table into ndbcluster and I got

Node 3: Data usage is 25%(9848 32K pages of total 38400)
Node 3: Index usage is 1%(1070 8K pages of total 64032)

So approximately 300MB in one node. 1GB is not 300MB ...

I'm using mysql-cluster 7.2.8.

So? :) Thanks.

EDIT: I see that I have one BLOB_msg in table so each row has 2000bytes reserved for this column. Evidently ndbcluster is not reserving so much for every row.. so what value should I trust?

EDIT2: Ok I see ndb_size value is the MAX datamemory size if every row has all the fields pimped to max.. so that's why it takes 2000bytes for this field for every row. Right ?
↧
↧

Bug in ndb_mgm ? (no replies)

November 18, 2012, 7:05 am
$
0
0
Ok, this is rather funny ...

ndb_mgm> all report memory
Node 2: Data usage is 223679%(-3208 32K pages of total 19200)

Ndb node is still in the starting phase.

2012-11-18 16:02:40 [ndbd] WARNING -- Watchdog: Warning overslept 263 ms, expected 100 ms.
2012-11-18 16:02:40 [ndbd] INFO -- timerHandlingLab now: 133124833 sent: 133124612 diff: 221
2012-11-18 16:02:41 [ndbd] INFO -- Watchdog: User time: 234 System time: 78386
2012-11-18 16:02:41 [ndbd] WARNING -- Watchdog: Warning overslept 226 ms, expected 100 ms.
2012-11-18 16:02:43 [ndbd] INFO -- Watchdog: User time: 234 System time: 78578
2012-11-18 16:02:43 [ndbd] WARNING -- Watchdog: Warning overslept 450 ms, expected 100 ms.
2012-11-18 16:02:43 [ndbd] INFO -- timerHandlingLab now: 133127097 sent: 133126695 diff: 402
2012-11-18 16:03:13 [ndbd] WARNING -- Ndb kernel thread 0 is stuck in: Polling for Receive elapsed=169
2012-11-18 16:03:13 [ndbd] INFO -- Watchdog: User time: 241 System time: 81665
2012-11-18 16:03:13 [ndbd] INFO -- timerHandlingLab now: 133157974 sent: 133157720 diff: 254

How should I proceed now?

EDIT: Mysql cluster 7.2.8 debian 6.0 64bit

EDIT2:

2012-11-18 16:26:08 [ndbd] WARNING -- Ndb kernel thread 0 is stuck in: Allocating memory elapsed=12009
2012-11-18 16:26:08 [ndbd] INFO -- Watchdog: User time: 26 System time: 2429
2012-11-18 16:26:14 [ndbd] WARNING -- Ndb kernel thread 0 is stuck in: Allocating memory elapsed=18020
2012-11-18 16:26:14 [ndbd] INFO -- Watchdog: User time: 27 System time: 3029

1,3GB free, datamemory 600M, indexmemory 100M
↧

ndbinfo counters details (no replies)

November 19, 2012, 1:37 am
$
0
0
Hi
Please help me to get the following details of ndbinfo counters,

What is threshold value of TRANS counter in ndb stat ?
What is threshold value of SCANS counter in ndb stat ?
What is threshold value of WRITE counter in ndb stat ?
What is threshold value of READ counter in ndb stat ?
What is threshold value of ATTRINFO counter in ndb stat ?

Thanks in Advance
Ahafan C
↧
Search

Memcached Error 29024: Out of Ndb instances, growing freelist (1 reply)

November 22, 2012, 9:24 pm
$
0
0
Hi all

I was doing some testing/benchmarking of a MySQL NDB Cluster through memcached interface, using brutis (http://code.google.com/p/brutis/) to generate memcached transaction and I've seen that when pushing writes to the node upper limit (it reaches aprox 18-20k sets/sec), after a few minutes, it starts throwing

NDB Temporary Error 410: REDO log files overloaded (decrease TimeBetweenLocalCheckpoints or increase NoOfFragmentLogFiles)

I guess this is due to the fact that it cannot commit the REDO log (on disk) to the checkpoint (on disk), due to the massive amount of writes. And if I get this right, this is in spite of the REDO log size; I mean, the REDO log size just moves forward the error point, or it can help you manage a write burst, but not such a high sustained write traffic.

So, I slowed down the writes to about 2k/sec (I'm running on a RAID-1 SSD system capable of 350MB/sec according to hdparm) hoping to reach the optimal write/read point for this hardware...now the test doesn't fail after a couple of minutes but again after like 15 minutes fails with this new, strange error

Memcached Error 29024: Out of Ndb instances, growing freelist

I googled it and I cannot find anything about it, a part from the bzr commit of the error message in Mysql Cluster code, so I'm asking: what's this error?

My config.ini:

[ndbd default]
# Options affecting ndbd processes on all data nodes:
NoOfReplicas=3 # Number of replicas
DataMemory=4000M # How much memory to allocate for data storage
IndexMemory=2000M
MaxNoOfTables=256
MaxNoOfOrderedIndexes=256
MaxNoOfUniqueHashIndexes=128
MaxNoOfConcurrentOperations=200000
MaxNoOfExecutionThreads=13
TimeBetweenLocalCheckpoints=6
NoOfFragmentLogFiles=8

[ndb_mgmd]
# Management process options:
hostname=test-01.backend # Hostname or IP address of MGM node
datadir=/var/lib/mysql-cluster # Directory for MGM node log files
NodeId=1

[ndbd]
hostname=test-01.backend # Hostname or IP address
datadir=/var/lib/mysql/data # Directory for this data node's data files
NodeId=2

[ndbd]
hostname=test-02.backend # Hostname or IP address
datadir=/var/lib/mysql/data # Directory for this data node's data files
NodeId=3

[ndbd]
hostname=test-03.backend # Hostname or IP address
datadir=/var/lib/mysql/data # Directory for this data node's data files
NodeId=4

[mysqld]
NodeId=50

[mysqld]
NodeId=51

[mysqld]
NodeId=52

[mysqld]
NodeId=53

[mysqld]
NodeId=54

[mysqld]
NodeId=55

[mysqld]
NodeId=56

[mysqld]
NodeId=57

[mysqld]
NodeId=58

[mysqld]
NodeId=59

[mysqld]
NodeId=60

[mysqld]
NodeId=61

[mysqld]
NodeId=62

[mysqld]
NodeId=63

[mysqld]
NodeId=64

The emcached command line

# memcached -E /var/lib/mysql/lib/ndb_engine.so -e "connectstring=test-kyoto-01.backend:1186;role=db-only" -u mysql -t 50

The Brutis command line:

# ./brutis -x test-02.backend:11211 -v -P -f 50 -r 20:100 -s 600 -p "test4-" -n 10000000 -z 40000 -a S:R



Thanks!
↧

mysql cluster- fails to create Data node (no replies)

November 21, 2012, 7:38 am
$
0
0
Hi all,
I am using MySQL Cluster 7, Linux Red Hat 5 platform and VMware.
I have 6 virtual machines with the following IP's:
-Managment node 1 IP: 192.168.4.36 node ID:1
-Managment node 2 IP: 192.168.4.41 node ID:2

-Data node 1 IP: 192.168.4.50 Node ID:10
-Data node 2 IP: 192.168.4.40 Node ID:6

-SQL node 1 IP: 192.168.4.37 Node ID:3
-SQL node 2 IP: 192.168.4.42 Node ID:4

for Data node 1 & 2 and SQL node 1 & 2 I applied the following my.cnf file:

[mysqld]

datadir=/var/lib/mysql-cluster
NodeId= "number" //I applies a uniqe number for each node


ndbcluster
ndb_connectstring = 192.168.4.41
ndb_connectstring = 192.168.4.36


[ndb_mgm]

Connect-string=192.168.4.41

[ndb_mgm]

Connect-string=192.168.4.36

[ndb_mgmd]

Config-file=/var/lib/mysql-cluster/config.ini
connect-string=192.168.4.41

[ndb_mgmd]

Config-file=/var/lib/mysql-cluster/config.ini
connect-string=192.168.4.36

[mysql_cluster]

ndb-connectstring=192.168.4.41
ndb-connectstring=192.168.4.36


For the management nodes I applied this config.ini file

[ndbd DEFAULT]
NoOfReplicas = 2
DataMemory = 80M
IndexMemory = 18M
Datadir=/usr/local/mysql/data

LockPagesInMainMemory = 1



[NDB_MGMD]
HostName = 192.168.4.41

Datadir=/var/lib/mysql-cluster

NodeId= 2

[NDB_MGMD]
HostName = 192.168.4.36

Datadir=/var/lib/mysql-cluster

NodeId= 1


[ndbd]
HostName = 192.168.4.40

NodeId= 6

[ndbd]
HostName = 192.168.4.50

NodeId= 10


[MYSQLD]
HostName = 192.168.4.42
NodeId= 4


[MYSQLD]
HostName = 192.168.4.37
NodeId= 3


Then on each Management machine I started the Managment nodes and they started correctly.

then I typed "shell> ndbd" command on the data nodes machines, the "Data node 2" was created successfully, but "Data node 1" showed this error:

ERROR --Failed to allocate nodeid, error:'Failed to alloc node id at 192.168.4.36 port 1186: connection done form the wrong host ip 192.168.4.50.'

Althogh I always make sure that the nodeid in the "my.cnf" file always match the nodeid in the "config.ini"

The second problem is when I run the SHOW command on the Management nodes I recive this outputs:

Management node 1:-

Cluster Configuration
---------------------
[ndbd(NDB)] 1 node(s)
id=2 (not connected, accepting connect from 192.168.4.38)

[ndb_mgmd(MGM)] 1 node(s)
id=1 @192.168.4.36 (mysql-5.5.27 ndb-7.2.8)

[mysqld(API)] 2 node(s)
id=3 (not connected, accepting connect from any host)
id=4 (not connected, accepting connect from any host)
-----------------------------------------------------------------------
So the problems are:-
- it is not showing two Managment node it is only showing one only.
-the datanode it is showing is not present in the configurations at all, BTW, the Data node it is showing is the old IP, 192.168.4.38, for the Data node 1, the new IP is 192.168.4.50

Management Node 2:-

Cluster Configuration
---------------------
[ndbd(NDB)] 1 node(s)
id=3 (not connected, accepting connect from 192.168.4.40)

[ndb_mgmd(MGM)] 1 node(s)
id=1 @192.168.4.41 (mysql-5.5.27 ndb-7.2.8)

[mysqld(API)] 1 node(s)
id=2 (not connected, accepting connect from 192.168.4.42)
-------------------------------------------------------------------

The Problems are:
-It is only showing one management node
-It is only accepting connection form one SQL node (SQL node 2)

summarizing my questions:-
1- how to solve the issue with the Data node 1?
2- how to reconfigure the Managment nodes to connect to each other?
3- how to reconfigure Management node 2 to connect to 2 SQL nodes not only one?

Best Regards,
Omar Elbakary
↧
↧

how to enable ndbcluster storage engine (no replies)

November 23, 2012, 6:56 am
$
0
0
Hi everbody,
I am using MySQL cluster 7.2 and I have some questions about the installation. I am using Red Hat 5 (32 bit), and I installed the files in RPM formate.
I have 6 Nodes, 2x Managment, 2x ndbd, 2x SQL

My my.cnf file looks like that:-


[mysqld]

datadir=/var/lib/mysql
socket=/var/lib/mysql/mysql.sock
NodeId= 4


ndbcluster
ndb_connectstring = 192.168.4.41
ndb_connectstring = 192.168.4.36

[ndb_mgm]

Connect-string=192.168.4.41

[ndb_mgm]

Connect-string=192.168.4.36


[ndb_mgmd]

Config-file=/var/lib/mysql-cluster/config.ini
connect-string=192.168.4.41

[ndb_mgmd]

Config-file=/var/lib/mysql-cluster/config.ini
connect-string=192.168.4.36

[mysql_cluster]

ndb-connectstring=192.168.4.41

ndb-connectstring=192.168.4.36

the problem is, when I type "SHOW ENGINES" in command line of the SQL nodes, I find the NDBcluster support = "NO"

so please can anybody help with that issue and tell me how to change the status to YES

thank you in advanced
↧

question about READ COMMITTED isolation level (no replies)

November 24, 2012, 1:16 am
$
0
0
Hello folks,

I am new in MySQL cluster although I have some experiences with InnoDB. What is bugging me now is NDB engine doesn't support MVCC. So that means it doesn't support consistent nonlocking read, right? So does it acquire a lock before reading a row? I can't find anything in the docs, it seems everything relevant to READ COMMITTED isolation level is about InnoDB. Could anyone please point me to the right material?

Jim
↧

NDB API indexscan for async (no replies)

November 25, 2012, 1:03 am
$
0
0
Hi,

How to use ndb c++ native api for reading or any operation using part(COL1) of a primary key(CLO1+COL2) with "asynchronous" calls.

Thanks,
SYED
↧
Viewing all 1560 articles
Browse latest View live
Search
<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>