unable to create a cluster.

[nwarr...@gmail.com]

Hello,

I am trying PRM for mysql HA and went through the Percona start up guide. I have managed to install pacemaker and corosync on amazon Linux AMI. I have configured the resource on it as well. I was able to set up pacemaker and corosync configurations as well. but as soon as I run crm status I get 

Stack: classic openais (with plugin)

Current DC: ip-172-31-3-18 - partition WITHOUT quorum

Version: 1.1.10-9d39a6b

1 Nodes configured, 2 expected votes

3 Resources configured

Online: [ ip-172-31-3-18 ]

Full list of resources:

 Master/Slave Set: ms_MySQL [my_mysql]

     my_mysql   (ocf::percona:mysql):   FAILED ip-172-31-3-18 (unmanaged)

Failed actions:

    my_mysql_stop_0 on ip-172-31-3-18 'unknown error' (1): call=15, status=complete, last-rc-change='Thu Sep 29 21:34:30 2016', queued=0ms, exec=100ms

my pacemaker configuration is as follows 

node ip-172-31-3-18 \

        attributes my_mysql_mysql_master_IP=172.31.3.18

primitive my_mysql ocf:percona:mysql \

        params pid="/var/run/mysqld/mysqld.pid" replication_user=root replication_passwd="" test_user=root test_passwd="" max_slave_lag=15 evict_outdated_slaves=false async_stop=1 prm_binlog_parser_path="/usr/local/bin/prm_binlog_parser" \

        op monitor interval=5s role=Master timeout=30s OCF_CHECK_LEVEL=1 \

        op monitor interval=2s role=Slave timeout=30s OCF_CHECK_LEVEL=1 \

        op start interval=0 timeout=900s \

        op stop interval=0 timeout=900s

ms ms_MySQL my_mysql \

        meta master-max=1 master-node-max=1 clone-max=3 clone-node-max=1 notify=true globally-unique=false target-role=Master is-managed=true

property cib-bootstrap-options: \

        dc-version=1.1.10-9d39a6b \

        cluster-infrastructure="classic openais (with plugin)" \

        expected-quorum-votes=2 \

        no-quorum-policy=ignore \

        stonith-enabled=false

All the other mysql primitives are where the defaults are , I double checked them. my configuration is as follows 

os : Amazon Linux 2014.03

pacemaker : tested on 1.1.11 and 1.1.15

corosync: 1.4.1 

If any has encountered a similar situation , any help would be much appreciated.

Thank you 

 

[Yves Trudeau]

Hi,

   You need to define mysql passwords for the replication_user and test_user, there's no validation for an empty password and the options are handle like:

 MYSQL_OPTIONS_REPL="$MYSQL_OPTIONS_LOCAL --user=$OCF_RESKEY_replication_user --password=$OCF_RESKEY_replication_passwd"

MYSQL_OPTIONS_TEST="$MYSQL_OPTIONS_LOCAL --user=$OCF_RESKEY_test_user --password=$OCF_RESKEY_test_passwd"

An empty password will cause the mysql client to fail connecting.

Regards,

Yves

--
You received this message because you are subscribed to the Google Groups "PRM-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to prm-discuss+unsubscribe@googlegroups.com.
To post to this group, send email to prm-d...@googlegroups.com.
Visit this group at https://groups.google.com/group/prm-discuss.
For more options, visit https://groups.google.com/d/optout.

[nwarr...@gmail.com]

Hello Yves,

Thank you for taking the time to respond. I tried with proper mysql credentials as well. The same error is popping up when I try that way. I am building pacemaker and was wondering if an improper build is causing this. I'll check the build log and try a different *dummy* resource manager. Ill post an update. 

To unsubscribe from this group and stop receiving emails from it, send an email to prm-discuss...@googlegroups.com.

[Yves Trudeau]

Hi,

   then, the best way to see what is going on is to enable the agent trace with:

mkdir -p /tmp/mysql.ocf.ra.debug

touch /tmp/mysql.ocf.ra.debug/log

Clear the error and look at the log file to see where the error happens.

Regards,

Yves

To unsubscribe from this group and stop receiving emails from it, send an email to prm-discuss+unsubscribe@googlegroups.com.

[nwarr...@gmail.com]

Hello Yves,

I tried with a dummy RA like I said and as it happens the build was incomplete and hence the RA was not being called. I finally got all the dependencies sorted out and got it working.

Thanks for taking the time. 

[Yves Trudeau]

Hi,

   you're welcome!

Regards,

Yves

To unsubscribe from this group and stop receiving emails from it, send an email to prm-discuss+unsubscribe@googlegroups.com.