Shulou(Shulou.com)06/01 Report--

Failure 1: failure 1 encountered in the initial startup

[ERROR] InnoDB: auto-extending data file. / ibdata1 is of a different size 640 pages (rounded down to MB) than specified in the .cnf file: initial 768 pages, max 0 (relevant if non-zero) pages!

The cause of the problem:

The system automatically establishes several ibdata1,ib_logfile0,ib_logfile1 files when initializing the database.

Caused by the configuration of: innodb_data_file_path = ibdata1:10M;ibdata2:10M:autoextend in the configuration file, which is inconsistent with the size of the initialization ibdata1

The default ibdata1 size is 8m, while the one I configure here is 10m, which is the cause of the problem

Solution:

Under the data directory of mysql, delete ibdata1,ib_logfile0,ib_logfile1

After restarting mysql, the system will generate the ibdata file again according to the configuration

Failure 2: failure 2 encountered in the initial startup

Which: no socat in (/ usr/sbin:/sbin:/usr//bin:/sbin:/usr/sbin:/bin:/usr/bin:/usr/bin) WSREP_SST: [ERROR] socat not found in path: / usr/sbin:/sbin:/usr//bin:/sbin:/usr/sbin:/bin:/usr/bin:/usr/bin (20170323 1715 32 which 25.176)

The cause of the problem:

Wsrep_sst_auth= "sst:86rt6y1adf@abc" is configured in server.cnf

Need to use socat software package

Solution:

# yum install epel-release-y (epel source can provide socat package, other yum does not have socat package)

# yum install socat-y

Just start it again after installation

Failure 3: a large number of slow queries caused the cpu service to surge, resulting in server downtime

Cause of the problem: when a slow query is blocked, all subsequent requests will wait for the query to be completed before execution

Solution:

If there are only a few slow queries, you can directly use the kill slow query ID number to end the slow query; for example: mysql > kill 32232

If there are a large number of slow queries, you can only shut down the current database service.

Problems encountered when closing with / etc/init.d/mysql stop when there are a large number of slow queries:

1. The shutdown was very slow. After waiting for about 15 minutes, the shutdown failed, and the database could not log in.

two。 At this time, the kill-9 PID number is used to forcibly shut down the mysql process.

3. When / etc/init.d/mysql start is used at this time, the startup fails

4. Use / etc/init.d/mysql start to start again, and you can start successfully.

Temporary solution:

1. Find a database that can be used normally

two。 Remove databases that cannot be accessed normally in the front-end haproxy

3. Wait for the failed data to return to normal before adding this host from haproxy

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.