MySQL Group Replication stuck on RECOVERING

MySQLUbuntu

I'm testing MySQL Group Replication.

Environment:
Microsoft Azure
Ubuntu 16.0.4

I've followed this guide:
https://www.digitalocean.com/community/tutorials/how-to-configure-mysql-group-replication-on-ubuntu-16-04

I ALSO, however, added a second NIC to the two machines I'm testing on, in order to have the replication take place on a separate LAN so the primary one is only for connection with the applications.

As far as I can tell, the nodes can see each other:

mysql> SELECT * FROM performance_schema.replication_group_members;
+---------------------------+--------------------------------------+-------------+-------------+--------------+
| CHANNEL_NAME              | MEMBER_ID                            | MEMBER_HOST | MEMBER_PORT | MEMBER_STATE |
+---------------------------+--------------------------------------+-------------+-------------+--------------+
| group_replication_applier | f0bcfc98-4255-11e8-b39f-000d3a1db637 | 10.3.1.4    |        3306 | ONLINE       |
| group_replication_applier | f8867ce8-4255-11e8-b03c-000d3a133573 | 10.3.1.5    |        3306 | RECOVERING   |
+---------------------------+--------------------------------------+-------------+-------------+--------------+

10.3.1.0/24 is my replication LAN. 10.3.0.0/24 is the application/primary LAN.

The problem is that when I add the second node, it stays at RECOVERING. After a while, it ends up removing the second node.

2018-05-01T15:01:18.067031Z 0 [Note] Plugin group_replication reported: 'getstart group_id e459a5fc'
2018-05-01T15:01:19.067227Z 11 [Note] Plugin group_replication reported: 'Only one server alive. Declaring this server as online within the replication group'
2018-05-01T15:01:19.067238Z 0 [Note] Plugin group_replication reported: 'Group membership changed to 10.3.1.4:3306 on view 15251868790666823:1.'
2018-05-01T15:01:19.074671Z 0 [Note] Plugin group_replication reported: 'This server was declared online within the replication group'
2018-05-01T15:01:47.201473Z 0 [Note] Plugin group_replication reported: 'getstart group_id e459a5fc'
2018-05-01T15:01:52.684929Z 0 [Note] Plugin group_replication reported: 'Members joined the group: 10.3.1.5:3306'
2018-05-01T15:01:52.685016Z 0 [Note] Plugin group_replication reported: 'Group membership changed to 10.3.1.4:3306, 10.3.1.5:3306 on view 15251868790666823:2.'

    2018-05-01T15:10:53.575405Z 0 [Note] Plugin group_replication reported: 'getstart group_id e459a5fc'
    2018-05-01T15:10:53.913923Z 0 [Warning] Plugin group_replication reported: 'Members removed from the group: 10.3.1.5:3306'
    2018-05-01T15:10:53.914018Z 0 [Note] Plugin group_replication reported: 'Group membership changed to 10.3.1.4:3306 on view 15251868790666823:3.'

Here's the contents of /etc/mysql/mysql.conf.d/mysqld.cnf

[mysqld_safe]
socket          = /var/run/mysqld/mysqld.sock
nice            = 0

[mysqld]
user            = mysql
pid-file        = /var/run/mysqld/mysqld.pid
socket          = /var/run/mysqld/mysqld.sock
basedir         = /usr
datadir         = /var/lib/mysql
tmpdir          = /tmp
log-error       = /var/log/mysql/error.log

gtid_mode = ON
enforce_gtid_consistency = ON
master_info_repository = TABLE
relay_log_info_repository = TABLE
binlog_checksum = NONE
log_slave_updates = ON
log_bin = binlog
binlog_format = ROW
transaction_write_set_extraction = XXHASH64
loose-group_replication_bootstrap_group = OFF
loose-group_replication_start_on_boot = OFF
loose-group_replication_ssl_mode = REQUIRED
loose-group_replication_recovery_use_ssl = 1

# Shared replication group configuration
loose-group_replication_group_name = "fe2d1be8-1ba8-403b-83df-f75711b2f3d1"
loose-group_replication_ip_whitelist = "10.3.1.0/24"
loose-group_replication_group_seeds = "10.3.1.4:33061,10.3.1.5:33061"

# Single or Multi-primary mode? Uncomment these two lines
# for multi-primary mode, where any host can accept writes
loose-group_replication_single_primary_mode = OFF
loose-group_replication_enforce_update_everywhere_checks = ON

# Host specific replication configuration
server_id = 1
bind-address = "10.3.0.23"
report_host = "10.3.1.4"
loose-group_replication_local_address = "10.3.1.4:33061"



lc-messages-dir = /usr/share/mysql
skip-external-locking
max_connections = 1000
table_open_cache=2048
sql_mode = "STRICT_TRANS_TABLES,ERROR_FOR_DIVISION_BY_ZERO,NO_AUTO_CREATE_USER,NO_ENGINE_SUBSTITUTION"

# By default we only accept connections from localhost
# bind-address  = 127.0.0.1
# Disabling symbolic-links is recommended to prevent assorted security risks
# symbolic-links=0

# * Fine Tuning
#
key_buffer_size = 64M
max_allowed_packet      = 768M
thread_stack            = 256K
thread_cache_size       = 64
join_buffer_size = 16M
sort_buffer_size = 8M

long_query_time = 2
log-queries-not-using-indexes
#
# The following can be used as easy to replay backup logs or for replication.
# note: if you are setting up a replication slave, see README.Debian about
#       other settings you may need to change.
#server-id              = 1
#log_bin                        = /var/log/mysql/mysql-bin.log
expire_logs_days        = 10
max_binlog_size   = 100M
#binlog_do_db           = include_database_name
#binlog_ignore_db       = include_database_name
#
# * InnoDB
#
# InnoDB is enabled by default with a 10MB datafile in /var/lib/mysql/.
# Read the manual for more InnoDB related options. There are many!
max_heap_table_size=2G
tmp_table_size=2G


innodb_write_io_threads=8
innodb_read_io_threads=8

innodb_log_file_size = 1G
innodb_buffer_pool_size = 50G
innodb_log_buffer_size = 20M
#wait_timeout = 28800
#wait_timeout = 20
#interactive_timeout = 28800
interactive_timeout = 120
innodb_flush_log_at_trx_commit=0
innodb_flush_method = O_DIRECT
innodb_buffer_pool_instances = 8
innodb_sort_buffer_size = 2M

Please let me know if I've missed adding some important info, and I'll supply it asap. NOTE that the config above is for the primary. Config for secondary is identical except for the different IPs.

One question I have is: Shouldn't the MEMBER_PORT be 33061 instead of 3306 since we set that in the config as the replication port?

Thanks in advance!

Best Answer

Please make sure that you have opened all the required ports used by MySQL group replication (as per your config):

1) 3306

2) 33061

Also, if the member is being removed from the MySQL cluster, there must be something in the mysql logs. Please provide the detailed mysql log.

I have setup a group replication cluster in production with 3 nodes and it is working fine.

Related Solutions

Mysql – How to debug a db memory-leak causing thesql to go before it’s own limits

...even surpassing it's theorically maximum possible allocation.

[OK] Maximum possible memory usage: 7.3G (46% of installed RAM)

There is not actually a way to calculate maximum possible memory usage for MySQL, because there is no cap on the memory it can request from the system.

The calculation done by mysqltuner.pl is only an estimate, based on a formula that doesn't take into account all possible variables, because if all possible variables were taken into account, the answer would always be "infinite." It's unfortunate that it's labeled this way.

Here is my theory on what's contributing to your excessive memory usage:

thread_cache_size       = 128

Given that your max_connections is set to 200, the value of 128 for thread_cache_size seems far too high. Here's what makes me think this might be contributing to your problem:

When a thread is no longer needed, the memory allocated to it is released and returned to the system unless the thread goes back into the thread cache. In that case, the memory remains allocated.

^{http://dev.mysql.com/doc/refman/5.6/en/memory-use.html}

If your workload causes even an occasional client thread to require a large amount of memory, those threads may be holding onto that memory, then going back to the pool and sitting around, continuing to hold on to memory they don't technically "need" any more, on the premise that holding on to the memory is less costly than releasing it if you're likely to need it again.

I think it's worth a try to do the following, after first making a note of how much memory MySQL is using at the moment.

Note how many threads are currently cached:

mysql> show status like 'Threads_cached';
+----------------+-------+
| Variable_name  | Value |
+----------------+-------+
| Threads_cached | 9     |
+----------------+-------+
1 row in set (0.00 sec)

Next, disable the thread cache.

mysql> SET GLOBAL thread_cache_size = 0;

This disables the thread cache, but the cached threads will stay in the pool until they're used one more time. Disconnect from the server, then reconnect and repeat.

mysql> show status like 'Threads_cached';

Continue disconnecting, reconnecting, and checking until the counter reaches 0.

Then, see how much memory MySQL is holding.

You may see a decrease, possibly significant, and then again you may not. I tested this on one of my systems, which had 9 threads in the cache. Once those threads had all been cleared out of the cache, the total memory held by MySQL did decrease... not by much, but it does illustrate that threads in the cache do release at least some memory when they are destroyed.

If you see a significant decrease, you may have found your problem. If you don't, then there's one more thing that needs to happen, and how quickly it can happen depends on your environment.

If the theory holds that the other threads -- the ones currently servicing active client connections -- have significant memory allocated to them, either because of recent work in their current client session or because of work requiring a lot of memory that was done by another connection prior to them languishing in the pool, then you won't see all of the potential reduction in memory consumption until those threads are allowed to die and be destroyed. Presumably your application doesn't hold them forever, but how long it will take to know for sure whether there's a difference will depend on whether you have the option of cycling your application (dropping and reconnecting the client threads) or if you'll have to just wait for them to be dropped and reconnected over time on their own.

But... it seems like a worthwhile test. You should not see a substantial performance penalty by setting thread_cache_size to 0. Fortunately, thread_cache_size is a dynamic variable, so you can freely change it with the server running.

MySQL 5.6 – No Longer Defaulting to /tmp/mysql.sock

The MySQL documentation states that the default path to mysql.sock is /var/lib/mysql/ on RPM distributions. The path is set at compile-time depending on the target OS and can be overridden by configuration or command line, so for CentOS the behavior you are seeing is correct.

This means that the socket path must have originally been set to /tmp by some other mechanism. There are several places this could have been done; likely in the init script /etc/init.d/mysql.server (to set socket=/tmp/mysql.sock on the command line - this can be done in an init script even though it is unsupported) or a local options file. See this page in the documentation for MySQL's precedence order when processing startup options. Entries listed later in the table will override the preceding ones.

It probably isn't possible to trace the sequence of events leading up to the socket path change, but fortunately you can change it via ~/.my.cnf, the user-specific option file.

Best Answer

Related Solutions

Mysql – How to debug a db memory-leak causing thesql to go before it’s own limits

MySQL 5.6 – No Longer Defaulting to /tmp/mysql.sock

Related Question