0

I have a 2-node fresh Galera Cluster in 2 DCs. I Import a 3GB dump into one node and the other starts to sync. After importing dump, I check the size and rows on both nodes and they are not equal. This is my galera.cnf ( Customized for WAN env )

[mysqld]
binlog_format=ROW
default-storage-engine=innodb
innodb_autoinc_lock_mode=2
bind-address=0.0.0.0

# Galera Provider Configuration
wsrep_on=ON
wsrep_provider=/usr/lib/galera/libgalera_smm.so

# Galera Cluster Configuration
wsrep_cluster_name="my-company"
wsrep_cluster_address="gcomm://XXXX,YYYY"

# Galera Synchronization Configuration
wsrep_sst_method=rsync

# Galera Node Configuration
wsrep_node_address="XXXX"
wsrep_node_name="Node-1"

# Galera Custome setting for WAN
wsrep_provider_options="evs.suspect_timeout=PT1M"
wsrep_provider_options="evs.inactive_timeout=PT1M"
wsrep_provider_options="evs.send_window=512"
wsrep_provider_options="evs.user_send_window=512"
wsrep_provider_options="evs.inactive_check_period=PT30S"
wsrep_provider_options="gcs.fc_factor=1"
wsrep_provider_options="gcs.fc_limit=260"
wsrep_provider_options="gcs.max_packet_size=2097152"

This is Databases size after import dump:

Node1: 27329.4 MB
Node2: 17570.1 MB

This is bandwidth usage on the node that I imported dump: enter image description here

Why many record did't insert? It's about mysql tuning or something else?

Ubuntu 18.0.4
mysql-wsrep-server-5.7
8 Core cpu, 16GB Ram

I hope someone can help, Thanks a lot.

3
  • One node in each DCs. Dec 23, 2019 at 20:38
  • I edited your question to clarify it. Was the dump from mysqldump? Did it give any errors?
    – Rick James
    Dec 23, 2019 at 22:51
  • @RickJames mysqldump -uMyUSER -pMyPASS --single-transaction db > db.dump mysql -uMyUSER -p NewDB < db.dump It didn't return any error on command line or mysql error log file. Dec 24, 2019 at 6:35

2 Answers 2

1

From the documentation:

Options need to be provided as a semicolon (;) separated list on a single line.

Only wsrep_provider_options="gcs.max_packet_size=2097152" is applied, the rest of the specified options are ignored. - https://dba.stackexchange.com/u/217784

-1

There a few factors that can provoke a missmatch between datasets size (the most often one is when many UPDATES or DELETE happens in one of the servers). What you can do to ensure data consistency across nodes is table checksum. If you don't have many tables this should be simple to do. Of course you need the tables without traffic to do this, not sure if this is possible.

1
  • As it’s currently written, your answer is unclear. Please edit to add additional details that will help others understand how this addresses the question asked. You can find more information on how to write good answers in the help center.
    – Community Bot
    Sep 15 at 20:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.