本节介绍了与日志相关的配置参数log_XXX。 ### 概览 在新initdb的数据库上查询pg_settings，可查询PostgreSQL中与log的参数包括： ``` [test@localhost ~]$ psql Expanded display is used automatically. psql (12.1) Type "help" for help. [local:/var/run/test]:5000 test@testdb=# select category,name,setting from pg_settings where name like 'log%' order by category; category | name | setting --------------------------------------+-----------------------------+-------------------------------- Reporting and Logging / What to Log | log_lock_waits | off Reporting and Logging / What to Log | log_checkpoints | off Reporting and Logging / What to Log | log_connections | off Reporting and Logging / What to Log | log_timezone | PRC Reporting and Logging / What to Log | log_temp_files | -1 Reporting and Logging / What to Log | log_disconnections | off Reporting and Logging / What to Log | log_duration | off Reporting and Logging / What to Log | log_error_verbosity | default Reporting and Logging / What to Log | log_statement | none Reporting and Logging / What to Log | log_replication_commands | off Reporting and Logging / What to Log | log_autovacuum_min_duration | -1 Reporting and Logging / What to Log | log_hostname | off Reporting and Logging / What to Log | log_line_prefix | %m [%p] Reporting and Logging / When to Log | log_min_duration_statement | -1 Reporting and Logging / When to Log | log_min_error_statement | error Reporting and Logging / When to Log | log_min_messages | warning Reporting and Logging / When to Log | log_transaction_sample_rate | 0 Reporting and Logging / Where to Log | log_destination | stderr Reporting and Logging / Where to Log | log_filename | postgresql-%Y-%m-%d_%H%M%S.log Reporting and Logging / Where to Log | logging_collector | off Reporting and Logging / Where to Log | log_truncate_on_rotation | off Reporting and Logging / Where to Log | log_rotation_size | 10240 Reporting and Logging / Where to Log | log_file_mode | 0600 Reporting and Logging / Where to Log | log_rotation_age | 1440 Reporting and Logging / Where to Log | log_directory | log Statistics / Monitoring | log_statement_stats | off Statistics / Monitoring | log_planner_stats | off Statistics / Monitoring | log_executor_stats | off Statistics / Monitoring | log_parser_stats | off (29 rows) [local:/var/run/test]:5000 test@testdb=# ``` log_打头的参数有29个，下面从where、when、what这几个维度来解析这些参数，本节是第三部分，介绍what to log。 ### What to log **debug_print_parse** 是否打印分析阶段的查询树，默认为off-不打印 **debug_print_rewritten** 是否打印查询重写阶段的查询树，默认为off-不打印 **debug_print_plan** 是否打印计划阶段的计划树，默认为off-不打印 **debug_pretty_print** 是否以优雅的方式打印？默认为on **log_checkpoints** 是否记录checkpoint信息，默认为off。 在checkpoint发生时，PG会记录相关的信息 ``` checkpoint_timeout = 1min # range 30s-1d checkpoint_completion_target = 0.5 # checkpoint target duration, 0.0 - 1.0 #checkpoint_flush_after = 256kB # measured in pages, 0 disables #checkpoint_warning = 30s # 0 disables log_checkpoints = on ``` 插入数据，在检查点发生时，可以看到检查点的相关信息输出 ``` 2019-12-30 14:27:03.618 CST [2224] LOG: duration: 106.374 ms statement: insert into tbl select x,'c1'||x from generate_series(1,10000) x; 2019-12-30 14:28:00.428 CST [2166] LOG: checkpoint starting: time 2019-12-30 14:28:06.387 CST [2166] LOG: checkpoint complete: wrote 59 buffers (0.4%); 0 WAL file(s) added, 0 removed, 0 recycled; write=5.949 s, sync=0.001 s, total=5.958 s; sync files=3, longest=0.001 s, average=0.000 s; distance=764 kB, estimate=764 kB ``` **log_connections** 记录登录信息，包括什么时候有连接请求，哪个用户连接的是哪个数据库；如连接失败，也会记录相关信息。 ``` ###修改配置信息 [test@localhost ~]$ grep 'log_connections' $PGDATA/postgresql.conf # "postgres -c log_connections=on". Some parameters can be changed at run time log_connections = on [test@localhost ~]$ pg_ctl reload server signaled [test@localhost ~]$ ###日志输出 2019-12-30 14:33:02.527 CST [2634] LOG: connection received: host=[local] 2019-12-30 14:33:02.531 CST [2634] LOG: connection authorized: user=test database=testdb application_name=psql ``` **log_disconnections** 记录连接断开登录信息 ``` ### [test@localhost ~]$ grep 'log_disconnections' $PGDATA/postgresql.conf log_disconnections = on [test@localhost ~]$ ###日志输出 2019-12-30 14:34:57.646 CST [2734] LOG: disconnection: session time: 0:00:04.885 user=test database=testdb host=[local] ``` **log_duration** 记录执行时间，仅记录执行时间，没有其他多余的信息。默认为on。 ``` 2019-12-30 14:36:49.149 CST [2224] LOG: duration: 12.178 ms ``` **log_error_verbosity** 出现错误时的日志诊断信息级别，可选项包括：terse, default, or verbose，默认值为default 使用verbose，可以看到哪个源文件的哪一行，方便诊断 ``` ###创建测试对象 [local:/data/run/test]:5000 test@testdb=# create view vw_tbl as select * from tbl; CREATE VIEW [local:/data/run/test]:5000 test@testdb=# ###terse 2019-12-30 14:41:30.395 CST [2163] LOG: parameter "log_error_verbosity" changed to "terse" 2019-12-30 14:41:39.689 CST [2224] ERROR: cannot drop table tbl because other objects depend on it ###default 2019-12-30 14:42:27.106 CST [2163] LOG: parameter "log_error_verbosity" changed to "default" 2019-12-30 14:42:33.287 CST [2224] ERROR: cannot drop table tbl because other objects depend on it 2019-12-30 14:42:33.287 CST [2224] DETAIL: view vw_tbl depends on table tbl 2019-12-30 14:42:33.287 CST [2224] HINT: Use DROP ... CASCADE to drop the dependent objects too. ###verbose 2019-12-30 14:43:29.654 CST [2163] LOG: 00000: parameter "log_error_verbosity" changed to "verbose" 2019-12-30 14:43:29.654 CST [2163] LOCATION: ProcessConfigFileInternal, guc-file.l:456 2019-12-30 14:43:32.790 CST [2224] ERROR: 2BP01: cannot drop table tbl because other objects depend on it 2019-12-30 14:43:32.790 CST [2224] DETAIL: view vw_tbl depends on table tbl 2019-12-30 14:43:32.790 CST [2224] HINT: Use DROP ... CASCADE to drop the dependent objects too. 2019-12-30 14:43:32.790 CST [2224] LOCATION: reportDependentObjects, dependency.c:1196 ``` **log_hostname** 是否记录主机名称，默认为off。 **log_line_prefix** 每一行日志前的前缀。默认为'%m [%p] '，可使用的通配符包括： >%a – application name %u – user name %d – database name %r – remote host and port %h – remote host %p – process ID %t – timestamp without milliseconds %m – timestamp with milliseconds %i – command tag %e – SQL state %c – session ID %l – session line number %s – session start timestamp %v – virtual transaction ID %x – transaction ID (0 if none) %q – stop here in non-session processes %% – ‘%' 把该参数修改为'%m %u@%d %p %r'，reload后的日志输出： ``` 2019-12-30 15:20:31.707 CST @ [2163] LOG: 00000: parameter "log_line_prefix" changed to "%m %u@%d [%p] %r " 2019-12-30 15:20:31.707 CST @ [2163] LOCATION: ProcessConfigFileInternal, guc-file.l:456 2019-12-30 15:20:40.148 CST test@testdb [2224] [local] ERROR: 2BP01: cannot drop table tbl because other objects depend on it 2019-12-30 15:20:40.148 CST test@testdb [2224] [local] DETAIL: view vw_tbl depends on table tbl 2019-12-30 15:20:40.148 CST test@testdb [2224] [local] HINT: Use DROP ... CASCADE to drop the dependent objects too. 2019-12-30 15:20:40.148 CST test@testdb [2224] [local] LOCATION: reportDependentObjects, dependency.c:1196 ``` **log_lock_waits** 记录等待时间超过deadlock_timeout（默认为1s）的lock。 ``` 2019-12-30 15:27:06.637 CST @ [2163] LOG: 00000: parameter "log_lock_waits" changed to "on" 2019-12-30 15:27:06.637 CST @ [2163] LOCATION: ProcessConfigFileInternal, guc-file.l:456 2019-12-30 15:27:18.462 CST test@testdb [2224] [local] LOG: 00000: duration: 0.542 ms 2019-12-30 15:27:18.462 CST test@testdb [2224] [local] LOCATION: exec_simple_query, postgres.c:1289 2019-12-30 15:27:36.559 CST test@testdb [5749] [local] LOG: 00000: duration: 0.639 ms 2019-12-30 15:27:36.559 CST test@testdb [5749] [local] LOCATION: exec_simple_query, postgres.c:1289 2019-12-30 15:27:44.470 CST test@testdb [2224] [local] LOG: 00000: duration: 3.592 ms 2019-12-30 15:27:44.470 CST test@testdb [2224] [local] LOCATION: exec_simple_query, postgres.c:1289 2019-12-30 15:27:58.024 CST test@testdb [5749] [local] LOG: 00000: process 5749 still waiting for AccessShareLock on relation 16385 of database 16384 after 1000.615 ms at character 22 2019-12-30 15:27:58.024 CST test@testdb [5749] [local] DETAIL: Process holding the lock: 2224. Wait queue: 5749. 2019-12-30 15:27:58.024 CST test@testdb [5749] [local] LOCATION: ProcSleep, proc.c:1493 ... ``` **log_statement** 记录哪些语句，可选项包括none, ddl, mod, all。 设置该参数为all，而log_min_duration_statement设置为600 执行SQL ``` [local:/data/run/test]:5000 test@testdb=# select 1; ?column? ---------- 1 (1 row) [local:/data/run/test]:5000 test@testdb=# select pg_sleep(1); pg_sleep ---------- (1 row) ``` 虽然select 1;执行得很快，没有超过600ms，但由于设置了该参数为all，因此也会在日志中出现 ``` 2019-12-30 15:32:41.934 CST @ [2163] LOG: 00000: received SIGHUP, reloading configuration files 2019-12-30 15:32:41.934 CST @ [2163] LOCATION: SIGHUP_handler, postmaster.c:2635 2019-12-30 15:32:41.939 CST @ [2163] LOG: 00000: parameter "log_min_duration_statement" changed to "600" 2019-12-30 15:32:41.939 CST @ [2163] LOCATION: ProcessConfigFileInternal, guc-file.l:456 2019-12-30 15:32:41.939 CST @ [2163] LOG: 00000: parameter "log_statement" changed to "all" 2019-12-30 15:32:41.939 CST @ [2163] LOCATION: ProcessConfigFileInternal, guc-file.l:456 2019-12-30 15:32:51.932 CST test@testdb [2224] [local] LOG: 00000: statement: select 1; 2019-12-30 15:32:51.932 CST test@testdb [2224] [local] LOCATION: exec_simple_query, postgres.c:1045 2019-12-30 15:32:51.932 CST test@testdb [2224] [local] LOG: 00000: duration: 0.375 ms 2019-12-30 15:32:51.932 CST test@testdb [2224] [local] LOCATION: exec_simple_query, postgres.c:1289 2019-12-30 15:33:02.686 CST test@testdb [2224] [local] LOG: 00000: statement: select pg_sleep(1); 2019-12-30 15:33:02.686 CST test@testdb [2224] [local] LOCATION: exec_simple_query, postgres.c:1045 2019-12-30 15:33:03.691 CST test@testdb [2224] [local] LOG: 00000: duration: 1005.297 ms 2019-12-30 15:33:03.691 CST test@testdb [2224] [local] LOCATION: exec_simple_query, postgres.c:1289 ``` 把该参数设置为none，执行同样的SQL，日志输出中只有select pg_sleep(1); ``` 2019-12-30 15:35:06.740 CST @ [2163] LOG: 00000: received SIGHUP, reloading configuration files 2019-12-30 15:35:06.740 CST @ [2163] LOCATION: SIGHUP_handler, postmaster.c:2635 2019-12-30 15:35:06.743 CST @ [2163] LOG: 00000: parameter "log_statement" changed to "none" 2019-12-30 15:35:06.743 CST @ [2163] LOCATION: ProcessConfigFileInternal, guc-file.l:456 2019-12-30 15:35:09.995 CST test@testdb [2224] [local] LOG: 00000: duration: 0.325 ms 2019-12-30 15:35:09.995 CST test@testdb [2224] [local] LOCATION: exec_simple_query, postgres.c:1289 2019-12-30 15:35:12.441 CST test@testdb [2224] [local] LOG: 00000: duration: 1001.645 ms statement: select pg_sleep(1); 2019-12-30 15:35:12.441 CST test@testdb [2224] [local] LOCATION: exec_simple_query, postgres.c:1296 ``` **log_replication_commands** 是否记录复制命令。 **log_temp_files** 是否记录大小超过该参数配置大小的临时文件。在执行大批量数据排序或者使用临时表时可以用于诊断。 ``` 2019-12-30 15:40:37.992 CST @ [2163] LOG: 00000: received SIGHUP, reloading configuration files 2019-12-30 15:40:37.992 CST @ [2163] LOCATION: SIGHUP_handler, postmaster.c:2635 2019-12-30 15:40:37.993 CST @ [2163] LOG: 00000: parameter "log_replication_commands" changed to "on" 2019-12-30 15:40:37.993 CST @ [2163] LOCATION: ProcessConfigFileInternal, guc-file.l:456 2019-12-30 15:40:37.993 CST @ [2163] LOG: 00000: parameter "log_temp_files" changed to "1024" 2019-12-30 15:40:37.993 CST @ [2163] LOCATION: ProcessConfigFileInternal, guc-file.l:456 ###日志输出 2019-12-30 15:42:07.897 CST test@testdb [2224] [local] LOG: 00000: temporary file: path "base/pgsql_tmp/pgsql_tmp2224.4", size 29507584 2019-12-30 15:42:07.897 CST test@testdb [2224] [local] LOCATION: ReportTemporaryFileUsage, fd.c:1285 ``` **log_timezone** 是否记录时区信息。 **参考资料** [Understanding postgresql.conf : log](https://www.depesz.com/2011/05/06/understanding-postgresql-conf-log/)