pgbadger 分析postgresql 产生的日志

os: centos 7.4
postgresql: 9.1
pgbadger:9.2

pgbadger 是一款分析postgresql 日志文件的开软软件。
pgBadger is a PostgreSQL log analyzer build for speed with fully detailed reports from your PostgreSQL log file.

下载

https://github.com/dalibo/pgbadger

安装

# cd /tmp
# git clone https://github.com/dalibo/pgbadger.git
# cd ./pgbadger
# perl Makefile.PL
# make
# make install 

# which pgbadger
/usr/local/bin/pgbadger
# pgbadger --version
pgBadger version 9.2

使用

https://github.com/dalibo/pgbadger 这个页面有相对比较详细的介绍说明。
另外也可以参考 http://dalibo.github.io/pgbadger/
自己写了几个定时分析的脚本：
crontab -l

# crontab -l
00 03 * * * script -c "/apps/log_done.sh >> /apps/log_done.log"

tree /apps

# tree  /apps
/apps
├── log_done.log
├── log_done.sh
├── postgresql_log_trade
│   ├── postgresql.log-20180619-192.168.56.100
│   └── postgresql.log-20180619-192.168.56.101
├── postgresql_out_trade
│   ├── pgbadger-postgresql.log-20180619-192.168.56.100.html
│   └── pgbadger-postgresql.log-20180619-192.168.56.101.html
└── postgresql_script
    ├── postgresql_log_scp.sh
    ├── postgresql_mail_trade.sh
    └── postgresql_out_trade.sh

postgresql_log_scp.sh

# cat postgresql_log_scp.sh
#!/bin/bash

#20180619  peiyb add pgbadger deal postgresql log
#CDATE=`date +%Y%m%d`
#CDATEM=`date +%Y-%m-%d`
CDATE=$1
CDATEM=${CDATE:0:4}'-'${CDATE:4:2}'-'${CDATE:6:2}

#############################################
#
#拷贝 postgresql 的日志
#
#############################################

rm -f /apps/postgresql_log_trade/*

echo "`date ` scp postgresql trade log file"
scp root@192.168.56.100:/var/log/postgresql/postgresql-$CDATEM.csv /apps/postgresql_log_trade/postgresql.log-$CDATE-192.168.56.100
scp root@192.168.56.101:/var/log/postgresql/postgresql-$CDATEM.csv /apps/postgresql_log_trade/postgresql.log-$CDATE-192.168.56.101

postgresql_out_trade.sh

# cat ./postgresql_out_trade.sh 
#!/bin/bash

#20180619  peiyb add pgbadger deal postgres log
CDATE=$1
LOGBASEDIR='/apps/postgresql_log_trade'
OUTBASEDIR='/apps/postgresql_out_trade'

rm -f ${OUTBASEDIR}/*

#############################################
#
#使用 pgbadger 处理 
#
#############################################

for LFILE in `ls ${LOGBASEDIR}/`
do
 echo $LFILE
 /usr/local/bin/pgbadger --prefix='%t ' -Z +08 ${LOGBASEDIR}/${LFILE} -f csv -o ${OUTBASEDIR}/pgbadger-${LFILE}.html
done

postgresql_mail_trade.sh

# cat postgresql_mail_trade.sh
#!/bin/bash

#20180619  peiyb  add mail send
CDATE=$1
LOGBASEDIR='/apps/postgresql_log_trade'
OUTBASEDIR='/apps/postgresql_out_trade'
MAILLIST='[e-mail]'
ATTACHLIST=''

#############################################
#
#使用 linux mailx 发送 postgresql trade 生成的out文件
#
#############################################

#获取不同机器个数
for LFILE in `ls ${OUTBASEDIR}/ |grep -i ".html"|cut -d- -f4 | sort | uniq`
do
    ATTACHLIST=''
    #拼接找出符合的文件
    for AFILE in `ls ${OUTBASEDIR}/ |grep -i $LFILE`
    do
        ATTACHLIST=$ATTACHLIST' -a '$OUTBASEDIR'/'$AFILE
    done
    #echo $ATTACHLIST
    
    #发送邮件
    echo "见附件" | /bin/mailx  -s "[db][pg001][log analyze][$LFILE]" $ATTACHLIST $MAILLIST   
done

postgresql一些参数

log_duration = off #每一个完成的语句的持续时间被记录,记录会很多，建议关闭
log_statement = 'ddl'
log_min_duration_statement = 10000 #单位为ms，大于这个时间的语句会被记录
log_line_prefix = '%t '

log_checkpoints = on
log_connections = on
log_disconnections = on
log_lock_waits = on
log_temp_files = 0
log_autovacuum_min_duration = 0
log_error_verbosity = default

lc_messages='C'

参考：
http://dalibo.github.io/pgbadger/
https://github.com/dalibo/pgbadger
https://github.com/dalibo/pgbadger/releases