PostgreSQL DBA(17) - XLOG Record data內部結構

上節介紹了WAL segment file的總體內部結構和其中的部分結構，本節繼續介紹其中的XLOG Record data結構。

一、XLOG Record data

WAL segment file預設大小為16MB,其內部結構如下圖所示:

WAL segment file內部結構

其中XLOG Record data是儲存實際資料的結構,由以下幾部分組成:
1、0..N個XLogRecordBlockHeader,每個XLogRecordBlockHeader對應一個block data;
注意:如設定了BKPBLOCK_HAS_IMAGE標記,則在XLogRecordBlockHeader結構體後跟XLogRecordBlockImageHeader結構體;如設定了BKPIMAGE_HAS_HOLE和 BKPIMAGE_IS_COMPRESSED則在XLogRecordBlockImageHeader後跟XLogRecordBlockCompressHeader結構體;
2、XLogRecordDataHeader[Short|Long]:如資料<256Bytes,則使用Short格式,否則使用Long格式;
3、block data:full-write-block資料,如啟用了壓縮,則壓縮儲存,相關後設資料儲存在XLogRecordBlockHeader中的XLogRecordBlockCompressHeader中.
4、main data:(tuple) data/checkpoint等日誌資料.
插入資料時的XLOG Record data內部結構如下圖所示:

XLOG Record data for DML

下面逐一介紹上述幾個部分,透過使用hexdump工具檢視相關資料。

1、XLogRecordBlockHeader

uint8 id

[xdb@localhost pg_wal]$ hexdump -C 000000010000000100000042 -s 80 -n 1
00000050  00                                                |.|
00000051

塊引用ID為0x00,即0號Block.

uint8 fork_flags

[xdb@localhost pg_wal]$ hexdump -C 000000010000000100000042 -s 81 -n 1
00000051  20                                                | |
00000052

值為0x20,高4位用於標記,即BKPBLOCK_HAS_DATA

uint16 data_length

[xdb@localhost pg_wal]$ hexdump -C 000000010000000100000042 -s 82 -n 2
00000052  1e 00                                             |..|
00000054

payload bytes = 0x001E,十進位制數值為30.
接下來是RelFileNode

RelFileNode
tablespace/database/relation,均為Oid型別(unsigned int)
1.tablespace

[xdb@localhost pg_wal]$ hexdump -C 000000010000000100000042 -s 84 -n 4
00000054  7f 06 00 00                                       |....|
00000058

值為0x0000067F,十進位制值為1663
表空間為default

testdb=# select * from pg_tablespace where oid=1663;
  spcname   | spcowner | spcacl | spcoptions 
------------+----------+--------+------------
 pg_default |       10 |        | 
(1 row)

2.database

[xdb@localhost pg_wal]$ hexdump -C 000000010000000100000042 -s 88 -n 4
00000058  12 40 00 00                                       |.@..|
0000005c

值為0x00004012,十進位制值為16402,資料庫為testdb

testdb=# select * from pg_database where oid=16402;
 datname | datdba | encoding | datcollate | datctype | datistemplate | datallowconn | datconnlimit | datlastsysoid | datfroze
nxid | datminmxid | dattablespace | datacl 
---------+--------+----------+------------+----------+---------------+--------------+--------------+---------------+---------
-----+------------+---------------+--------
 testdb  |     10 |        6 | C          | C        | f             | t            |           -1 |         13284 |         
 561 |          1 |          1663 | 
(1 row)

3.relation

[xdb@localhost pg_wal]$ hexdump -C 000000010000000100000042 -s 92 -n 4
0000005c  56 42 00 00                                       |VB..|
00000060

值為0x00004256,十進位制值為16982

testdb=# select oid,relfilenode,relname from pg_class where relfilenode = 16982;
  oid  | relfilenode | relname 
-------+-------------+---------
 16982 |       16982 | t_jfxx
(1 row)

相應的關係為t_jfxx

BlockNumber

[xdb@localhost pg_wal]$ hexdump -C 000000010000000100000042 -s 96 -n 4
00000060  85 00 00 00                                       |....|
00000064

值為0x00000085,十進位制值為133,這是對應的資料塊號.

2、XLogRecordDataHeaderShort

接下來是XLogRecordDataHeaderShort/Long,由於資料小於256B,使用XLogRecordDataHeaderShort結構
unit8 id

[xdb@localhost pg_wal]$ hexdump -C 000000010000000100000042 -s 100 -n 1
00000064  ff                                                |.|
00000065

值為0xFF --> XLR_BLOCK_ID_DATA_SHORT 255
uint8 data_length

[xdb@localhost pg_wal]$ hexdump -C 000000010000000100000042 -s 101 -n 1
00000065  03                                                |.|
00000066

值為0x03,3個位元組,指的是main data的大小,3個位元組是xl_heap_insert結構體的大小.

3、block data

XLogRecordDataHeaderShort之後是block data,由兩部分組成:
1.xl_heap_header
2.Tuple data

xl_heap_header
1.uint16 t_infomask2

[xdb@localhost pg_wal]$ hexdump -C 000000010000000100000042 -s 102 -n 2
00000066  03 00                                             |..|
00000068

t_infomask2值為0x03,二進位制值為00000000 00000011

2.uint16 t_infomask

[xdb@localhost pg_wal]$ hexdump -C 000000010000000100000042 -s 104 -n 2
00000068  02 08                                             |..|
0000006a

t_infomask值為0x0802,二進位制值為00001000 00000010

3.uint8 t_hoff

[xdb@localhost pg_wal]$ hexdump -C 000000010000000100000042 -s 106 -n 1
0000006a  18                                                |.|
0000006b

t_hoff值(偏移)為0x18,十進位制值為24

Tuple data
XLOG Record的大小是0x4F,即79B,減去頭部資料XLogRecord(24B) + XLogRecordBlockHeader(20B) + XLogRecordDataHeaderShort(2B) + xl_heap_header(5B) + main data(3B),剩餘25B

[xdb@localhost pg_wal]$ hexdump -C 000000010000000100000042 -s 107 -n 25
0000006b  00 0d 32 30 39 31 39 0f  32 30 31 33 30 37 00 00  |..20919.201307..|
0000007b  00 00 00 00 00 00 03 b3  40                       |........@|
00000084

4、main data

這是xl_heap_insert結構體
uint16 OffsetNumber

[xdb@localhost pg_wal]$ hexdump -C 000000010000000100000042 -s 132 -n 2
00000084  26 00                                             |&.|
00000086

插入的tuple的偏移為0x0026,十進位制為38

uint8 flags

[xdb@localhost pg_wal]$ hexdump -C 000000010000000100000042 -s 134 -n 1
00000086  00                                                |.|
00000087

標誌位為0x00

二、參考資料

WAL Internals Of PostgreSQL
PostgreSQL 原始碼解讀（109）- WAL#5（相關資料結構）
PostgreSQL DBA(16) - WAL segment file內部結構
關於結構體佔用空間大小總結