PostgreSQL 未显示 RDS 日志记录
RDS logging not appearing for PostgreSQL
我按照说明进行操作 here，此处显示了我的参数组更改摘要：
我重新启动了实例和所有东西，我得到了这个新的日志文件：
里面error/postgres.log，我只有这个：
2018-03-13 12:39:38 UTC::@:[28236]:WARNING: unrecognized configuration parameter "rds.logical_replication"
2018-03-13 12:39:38 UTC::@:[28236]:LOG: redirecting log output to logging collector process
2018-03-13 12:39:38 UTC::@:[28236]:HINT: Future log output will appear in directory "/rdsdbdata/log/error".
2018-03-13 13:07:39 UTC::@:[11128]:WARNING: unrecognized configuration parameter "rds.logical_replication"
2018-03-13 13:07:39 UTC::@:[11128]:LOG: redirecting log output to logging collector process
2018-03-13 13:07:39 UTC::@:[11128]:HINT: Future log output will appear in directory "/rdsdbdata/log/error".
2018-03-13 14:13:53 UTC::@:[14981]:WARNING: unrecognized configuration parameter "rds.logical_replication"
2018-03-13 14:13:53 UTC::@:[14981]:LOG: redirecting log output to logging collector process
2018-03-13 14:13:53 UTC::@:[14981]:HINT: Future log output will appear in directory "/rdsdbdata/log/error".
----------------------- END OF LOG ----------------------
一个令人担忧的事情是，我从未收到此消息，正如文档中所说的那样，在应用新更改后我应该收到此消息：
2013-11-05 16:48:56 UTC::@:[2952]:LOG: received SIGHUP, reloading configuration files
2013-11-05 16:48:56 UTC::@:[2952]:LOG: parameter "log_min_duration_statement" changed to "1"
知道我做错了什么吗？
更新
回应：
forge=> select name,setting,source from pg_settings where name like 'log_%';
name | setting | source
-----------------------------+----------------------------+--------------------
log_autovacuum_min_duration | -1 | default
log_checkpoints | on | configuration file
log_connections | off | default
log_destination | stderr | configuration file
log_directory | /rdsdbdata/log/error | configuration file
log_disconnections | off | default
log_duration | off | default
log_error_verbosity | default | default
log_executor_stats | off | default
log_file_mode | 0644 | configuration file
log_filename | postgresql.log.%Y-%m-%d-%H | configuration file
log_hostname | on | configuration file
log_line_prefix | %t:%r:%u@%d:[%p]: | configuration file
log_lock_waits | off | default
log_min_duration_statement | 1 | configuration file
log_min_error_statement | error | default
log_min_messages | warning | default
log_parser_stats | off | default
log_planner_stats | off | default
log_replication_commands | off | default
log_rotation_age | 60 | configuration file
log_rotation_size | 10240 | default
log_statement | all | configuration file
log_statement_stats | off | default
log_temp_files | -1 | default
log_timezone | UTC | configuration file
log_truncate_on_rotation | off | configuration file
logging_collector | on | configuration file
更新
与下面的正确答案相同，我收到了 RDS 支持的回复并得到了这个：
I would like to let you know that I can see the Log files getting generated for your instance after applying the parameter group changes. Moreover I can also see the size of these log files getting larger. The log files that are getting generated after applying the modification for parameter group and reboot(attached parameter group at 2018-03-13 12:35:58 UTC and reboot at 2018-03-13 12:39:33 UTC), are as follows:
File Name Date and Time of generation Size
--------------------------------------------------------------------------------------------
error/postgresql.log.2018-03-13-12 2018-03-13T12:59:56+00:00 592550
error/postgresql.log.2018-03-13-13 2018-03-13T13:59:53+00:00 2258196
error/postgresql.log.2018-03-13-14 2018-03-13T15:00:00+00:00 2761617
error/postgresql.log.2018-03-13-15 2018-03-13T15:59:03+00:00 1000738
error/postgresql.log.2018-03-13-16 2018-03-13T16:59:02+00:00 582653
error/postgresql.log.2018-03-13-17 2018-03-13T17:59:01+00:00 111647
error/postgresql.log.2018-03-13-18 2018-03-13T18:59:02+00:00 66102
error/postgresql.log.2018-03-13-19 2018-03-13T19:59:02+00:00 45318
文档好像有点inconsistent/out-of-date。当我今天早上更改参数组并重新启动时，我得到了同样奇怪的 error/postgres.log 文件，内容相同。
不知道这是否适合您，但这对我来说只是以愚蠢的顺序对日志进行排序。我刷新，按 "Last written (descending)" 排序，实际上最新的日志弹出。
我按照说明进行操作 here，此处显示了我的参数组更改摘要：
我重新启动了实例和所有东西，我得到了这个新的日志文件：
里面error/postgres.log，我只有这个：
2018-03-13 12:39:38 UTC::@:[28236]:WARNING: unrecognized configuration parameter "rds.logical_replication"
2018-03-13 12:39:38 UTC::@:[28236]:LOG: redirecting log output to logging collector process
2018-03-13 12:39:38 UTC::@:[28236]:HINT: Future log output will appear in directory "/rdsdbdata/log/error".
2018-03-13 13:07:39 UTC::@:[11128]:WARNING: unrecognized configuration parameter "rds.logical_replication"
2018-03-13 13:07:39 UTC::@:[11128]:LOG: redirecting log output to logging collector process
2018-03-13 13:07:39 UTC::@:[11128]:HINT: Future log output will appear in directory "/rdsdbdata/log/error".
2018-03-13 14:13:53 UTC::@:[14981]:WARNING: unrecognized configuration parameter "rds.logical_replication"
2018-03-13 14:13:53 UTC::@:[14981]:LOG: redirecting log output to logging collector process
2018-03-13 14:13:53 UTC::@:[14981]:HINT: Future log output will appear in directory "/rdsdbdata/log/error".
----------------------- END OF LOG ----------------------
一个令人担忧的事情是，我从未收到此消息，正如文档中所说的那样，在应用新更改后我应该收到此消息：
2013-11-05 16:48:56 UTC::@:[2952]:LOG: received SIGHUP, reloading configuration files
2013-11-05 16:48:56 UTC::@:[2952]:LOG: parameter "log_min_duration_statement" changed to "1"
知道我做错了什么吗？
更新
回应
forge=> select name,setting,source from pg_settings where name like 'log_%';
name | setting | source
-----------------------------+----------------------------+--------------------
log_autovacuum_min_duration | -1 | default
log_checkpoints | on | configuration file
log_connections | off | default
log_destination | stderr | configuration file
log_directory | /rdsdbdata/log/error | configuration file
log_disconnections | off | default
log_duration | off | default
log_error_verbosity | default | default
log_executor_stats | off | default
log_file_mode | 0644 | configuration file
log_filename | postgresql.log.%Y-%m-%d-%H | configuration file
log_hostname | on | configuration file
log_line_prefix | %t:%r:%u@%d:[%p]: | configuration file
log_lock_waits | off | default
log_min_duration_statement | 1 | configuration file
log_min_error_statement | error | default
log_min_messages | warning | default
log_parser_stats | off | default
log_planner_stats | off | default
log_replication_commands | off | default
log_rotation_age | 60 | configuration file
log_rotation_size | 10240 | default
log_statement | all | configuration file
log_statement_stats | off | default
log_temp_files | -1 | default
log_timezone | UTC | configuration file
log_truncate_on_rotation | off | configuration file
logging_collector | on | configuration file
更新
与下面的正确答案相同，我收到了 RDS 支持的回复并得到了这个：
I would like to let you know that I can see the Log files getting generated for your instance after applying the parameter group changes. Moreover I can also see the size of these log files getting larger. The log files that are getting generated after applying the modification for parameter group and reboot(attached parameter group at 2018-03-13 12:35:58 UTC and reboot at 2018-03-13 12:39:33 UTC), are as follows:
File Name Date and Time of generation Size -------------------------------------------------------------------------------------------- error/postgresql.log.2018-03-13-12 2018-03-13T12:59:56+00:00 592550 error/postgresql.log.2018-03-13-13 2018-03-13T13:59:53+00:00 2258196 error/postgresql.log.2018-03-13-14 2018-03-13T15:00:00+00:00 2761617 error/postgresql.log.2018-03-13-15 2018-03-13T15:59:03+00:00 1000738 error/postgresql.log.2018-03-13-16 2018-03-13T16:59:02+00:00 582653 error/postgresql.log.2018-03-13-17 2018-03-13T17:59:01+00:00 111647 error/postgresql.log.2018-03-13-18 2018-03-13T18:59:02+00:00 66102 error/postgresql.log.2018-03-13-19 2018-03-13T19:59:02+00:00 45318
文档好像有点inconsistent/out-of-date。当我今天早上更改参数组并重新启动时，我得到了同样奇怪的 error/postgres.log 文件，内容相同。
不知道这是否适合您，但这对我来说只是以愚蠢的顺序对日志进行排序。我刷新，按 "Last written (descending)" 排序，实际上最新的日志弹出。