PostgreSQL 10.0 preview 功能增强 - 新增数十个IO等待事件监控

标签

PostgreSQL , 10.0 , 等待事件 , IO

背景

PostgreSQL 10.0新增了数十个IO等待事件，描述系统调用的等待。

Hi All,  

Attached is the patch, which extend the existing wait event infrastructure  
to  
implement the wait events for the disk I/O. Basically pg_stat_activity's  
wait  
event information to show data about disk I/O as well as IPC primitives.  

Implementation details:  

- Added PG_WAIT_IO to pgstat.h and a new enum WaitEventIO  
- Added a wait_event_info argument to FileRead, FileWrite, FilePrefetch,  
FileWriteback, FileSync, and FileTruncate. Set this wait event just before  
performing the file system operation and clear it just after.  
- Pass down an appropriate wait event from  caller of any of those  
functions.  
- Also set and clear a wait event around standalone calls to read(),  
write(), fsync() in other parts of the system.  
- Added documentation for all newly added wait event.  

Open issue:  
- Might missed few standalone calls to read(), write(), etc which need  
to pass the wait_event_info.  

Thanks to my colleague Robert Haas for his help in design.  

Please let me know your thought, and thanks for reading.  

Thanks,  
Rushabh Lathia  
www.EnterpriseDB.com  

IO等待事件如下

 /* ----------  
+ * Wait Events - IO  
+ *  
+ * Use this category when a process is waiting for a IO.  
+ * ----------  
+ */  
+typedef enum  
+{  
+   WAIT_EVENT_READ_DATA_BLOCK,  
+   WAIT_EVENT_WRITE_DATA_BLOCK,  
+   WAIT_EVENT_SYNC_DATA_BLOCK,  
+   WAIT_EVENT_EXTEND_DATA_BLOCK,  
+   WAIT_EVENT_FLUSH_DATA_BLOCK,  
+   WAIT_EVENT_PREFETCH_DATA_BLOCK,  
+   WAIT_EVENT_WRITE_REWRITE_DATA_BLOCK,  
+   WAIT_EVENT_SYNC_REWRITE_DATA_BLOCK,  
+   WAIT_EVENT_TRUNCATE_RELATION_DATA_BLOCKS,  
+   WAIT_EVENT_SYNC_RELATION,  
+   WAIT_EVENT_SYNC_IMMED_RELATION,  
+   WAIT_EVENT_READ_BUFFILE,  
+   WAIT_EVENT_WRITE_BUFFILE,  
+   /* Wait event for XLOG */  
+   WAIT_EVENT_READ_XLOG,  
+   WAIT_EVENT_READ_COPY_XLOG,  
+   WAIT_EVENT_WRITE_XLOG,  
+   WAIT_EVENT_WRITE_INIT_XLOG_FILE,  
+   WAIT_EVENT_WRITE_COPY_XLOG_FILE,  
+   WAIT_EVENT_WRITE_BOOTSTRAP_XLOG,  
+   WAIT_EVENT_SYNC_INIT_XLOG_FILE,  
+   WAIT_EVENT_SYNC_COPY_XLOG_FILE,  
+   WAIT_EVENT_SYNC_BOOTSTRAP_XLOG,  
+   WAIT_EVENT_SYNC_ASSIGN_XLOG_SYNC_METHOD,  
+   /* Wait event for CONTROL_FILE */  
+   WAIT_EVENT_WRITE_CONTROL_FILE,  
+   WAIT_EVENT_WRITE_UPDATE_CONTROL_FILE,  
+   WAIT_EVENT_SYNC_WRITE_CONTROL_FILE,  
+   WAIT_EVENT_SYNC_UPDATE_CONTROL_FILE,  
+   WAIT_EVENT_READ_CONTROL_FILE,  
+   /* Wait event for REORDER BUFFER */  
+   WAIT_EVENT_READ_REORDER_BUFFER,  
+   WAIT_EVENT_WRITE_REORDER_BUFFER,  
+   /* Wait event for LOGICAL MAPPING */  
+   WAIT_EVENT_READ_APPLY_LOGICAL_MAPPING,  
+   WAIT_EVENT_WRITE_LOGICAL_MAPPING_REWRITE,  
+   WAIT_EVENT_SYNC_LOGICAL_MAPPING_REWRITE,  
+   WAIT_EVENT_SYNC_LOGICAL_MAPPING_REWRITE_HEAP,  
+   WAIT_EVENT_TRUNCATE_LOGICAL_MAPPING_REWRITE,  
+   /* Wait event for SNAPBUILD */  
+   WAIT_EVENT_WRITE_SNAPBUILD_SERIALIZE,  
+   WAIT_EVENT_READ_SNAPBUILD_RESTORE,  
+   WAIT_EVENT_SYNC_SNAPBUILD_SERIALIZE,  
+   /* Wait event for SNRU */  
+   WAIT_EVENT_READ_SLRU_PAGE,  
+   WAIT_EVENT_WRITE_SLRU_PAGE,  
+   WAIT_EVENT_SYNC_SLRU_FLUSH,  
+   WAIT_EVENT_SYNC_SLRU_WRITE_PAGE,  
+   /* Wait event for TIMELINE HISTORY */  
+   WAIT_EVENT_READ_TIMELINE_HISTORY_WALSENDER,  
+   WAIT_EVENT_READ_TIMELINE_HISTORY_WRITE,  
+   WAIT_EVENT_WRITE_TIMELINE_HISTORY,  
+   WAIT_EVENT_WRITE_TIMELINE_HISTORY_FILE,  
+   WAIT_EVENT_SYNC_TIMELINE_HISTORY_WRITE,  
+   WAIT_EVENT_SYNC_TIMELINE_HISTORY_FILE,  
+   /* Wait event for TWOPHASE FILE */  
+   WAIT_EVENT_READ_TWOPHASE_FILE,  
+   WAIT_EVENT_WRITE_RECREATE_TWOPHASE_FILE,  
+   WAIT_EVENT_SYNC_RECREATE_TWOPHASE_FILE,  
+   /* Wait event for SYSLOGGER */  
+   WAIT_EVENT_READ_SYSLOGGER_FILE,  
+   WAIT_EVENT_WRITE_SYSLOGGER_FILE,  
+   /* Wait event for REPLSLOT */  
+   WAIT_EVENT_READ_RESTORE_REPLSLOT,  
+   WAIT_EVENT_WRITE_REPLSLOT,  
+   WAIT_EVENT_SYNC_RESTORE_REPLSLOT,  
+   WAIT_EVENT_SYNC_SAVE_REPLSLOT,  
+   /* Wait event for copydir */  
+   WAIT_EVENT_READ_COPY_FILE,  
+   WAIT_EVENT_WRITE_COPY_FILE,  
+   /* Wait event RELMAP FILE */  
+   WAIT_EVENT_READ_LOAD_RELMAP_FILE,  
+   WAIT_EVENT_WRITE_RELMAP_FILE,  
+   WAIT_EVENT_SYNC_WRITE_RELMAP_FILE,  
+   /* Wait event for LOCK FILE */  
+   WAIT_EVENT_READ_CREATE_LOCK_FILE,  
+   WAIT_EVENT_READ_ADDTODATEDIR_LOCK_FILE,  
+   WAIT_EVENT_READ_RECHECKDATADIR_LOCK_FILE,  
+   WAIT_EVENT_WRITE_CREATE_LOCK_FILE,  
+   WAIT_EVENT_WRITE_ADDTODATEDIR_LOCK_FILE,  
+   WAIT_EVENT_SYNC_ADDTODATEDIR_LOCK_FILE,  
+   WAIT_EVENT_SYNC_CREATE_LOCK_FILE  
+} WaitEventIO;  

对于判断系统瓶颈，又多了一个有力信息。

这个patch的讨论，详见邮件组，本文末尾URL。

PostgreSQL社区的作风非常严谨，一个patch可能在邮件组中讨论几个月甚至几年，根据大家的意见反复的修正，patch合并到master已经非常成熟，所以PostgreSQL的稳定性也是远近闻名的。

参考

https://commitfest.postgresql.org/13/975/

https://www.postgresql.org/message-id/flat/CAGPqQf0LsYHXREPAZqYGVkDqHSyjf=KsD=k0GTVPAuzyThh-VQ@mail.gmail.com#CAGPqQf0LsYHXREPAZqYGVkDqHSyjf=KsD=k0GTVPAuzyThh-VQ@mail.gmail.com