oracle block 格式 (zt)

tolywang發表於2007-07-13
資訊參考:
特別感謝 overtime 大哥對我的無私的幫助和對我一直鼓勵支援我的網友這些資料是沒得到oracle 支援的 所以不能保證資訊的正確性 請謹慎使用 .

oracle 8.1.7 8k block windows xp
create table t(n number);
insert into t values(1); --- 從 dbms_rowid 中看出table t 的資料在datafile 4 的第3個 block 上
alter system dump datafile 4 block 3;
下面是 dump 的檔案的資訊
還可以用下面的語句然後再用上面的 dump 語句看記憶體中 block 的2進位制儲存格式
但這記憶體中的2進位制格式和datafile中block資料表示的順序會略有不同可能是記憶體定址的緣故吧
ALTER SESSION SET EVENTS '10289 trace name context forever, level 1';
ALTER SESSION SET EVENTS '10289 trace name context off';
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
下面的是 udump 目錄下面 dump 檔案的資訊

----------------------- 第一部分

buffer tsn: 7 rdba: 0x01000003 (4/3)
scn: 0x0000.000802a5 seq: 0x01 flg: 0x02 tail: 0x02a50601
frmt: 0x02 chkval: 0x0000 type: 0x06=trans data

--- buffer tsn: 資料檔案對應的 tablespace 的 number 這只是dump檔案中記錄的資料而已
--- block 中是沒有記錄 tablespace 的 number 的
--- rdba: 4 bytes datafile 中 block 的地址 標示的第幾個 block
--- scn: 6 bytes system change number
--- seq: 1 byte A sequence number incremented for each change to a block at the same SCN
--- A new SCN is allocated if the sequence number wraps.
--- 同一個SCN影響這個block中的行數大於 254 行就會為這個事務分配一個新的SCN
--- 如下面的操作就可能引起同一個SCN但影響的同一個block 中的行超過254行
--- "delete from table_name"
--- 影響的行數(最大254) 是用從 0x01 到 0xfe 表示的
--- 當這個byte 的資料為 0xff 的時候標誌這個 block 壞調了---&gt ora-01578
--- flg: 1 byte 1 = virgin block
--- 2 = last change to the block was for a cleanout operation
--- 4 = checksum value is set
--- 8 = temporary data
--- 這是一個可以組合的值 也就是說有為 6 的時候是 2,4 兩種情況的組合
--- tail: 4 bytes 這是記錄在 block 的最後面的 4 bytes 的資料
--- dump檔案中只是把他寫在了前面而已
--- frmt: 1 byte oracle 8 以後看見的都是 0x02
--- chkval: 2 bytes 在 init 檔案中設定了 db_block_checksum=true 才有值
--- type: 1 byte 這個 block 的型別 這裡主要看 0x06 這種 因為這種是用來存使用者資料的
--- 其他的型別可以參考

----------------------- 第二部分
Block header dump: 0x01000003
Object id on Block? Y
seg/obj: 0x614a csc: 0x00.802a3 itc: 1 flg: O typ: 1 - DATA
fsl: 0 fnx: 0x0 ver: 0x01

--- seg/obj: 4 bytes 這裡是16進位制的 對應 sys.obj$.obj# 資料字典的資料
--- csc: 6 bytes The SCN at which the last full cleanout was performed on the block
--- itc: 1 byte 下面的Itl事務條的個數 8.1.7的文件上面說可以使用 INITRANS 在建表的時候
--- 限制這個值的大小(max 255超過會報ORA-02207) 但要考慮block 的空間是否夠
--- 表在8i中 INITRANS default為1 , 9.2.0中 INITRANS default為2
--- Yong Huang 說有些時候發生ORA-00060可以把表的 INITRANS 設定大點
--- ixora 上說當block的空間不夠建立一個ITL的時候一樣可能引起ORA-00054
--- flg: 2 bytes 0 indicates that the block is on a freelist. Otherwise the flag is -
--- 9i 的ASSM 的情況下這個值為 E
--- ixora 上說他佔用 2 bytes 但我下面的試驗和他的結果有一定的出入
--- 我觀察到的情況是 :
--- Object id on Block? Y
--- flg: O
--- ver: 0x01
--- 上面的3項是用同一個 byte 來表示的
--- typ: 1 byte 1 為 table ; 2 為 index. oracle進行查詢的時候是根據 obj$表中的情況來
--- 判斷物件的型別的,不是根據這個typ。也就是說如果有一個表但改變表中block
--- 的這個標誌位,一樣可以查詢出資料來,但dump block 時會出錯,如下面的ora
--- ORA-00600: 內部錯誤程式碼,自變數: [4555], [0], [], [], [], [], [], []
--- 錯誤中的 [0] 就是typ對應的資料
--- fsl: 1 byte Index to the first slot on the ITL freelist. ITL TX freelist slot
--- fnx: 4 bytes 自由列表中下一塊的地址 Null if this block is not on a freelist
--- 有資料例如: fnx: 0x1000029
--- ver: 1 byte format (version) 這個資料沒有看到相關的文件介紹 從ixora上說是佔用1byte
--- 但我從下面的2進位制檔案中看到的有不同 下面有介紹
--- unused: 4 bytes 在這裡還有4 bytes 的空閒的空間 但在上面的 dump 檔案上是沒顯示出來的
--- 這個unused 的4 bytes是 ixora 上面的說法
--- 9i 的 ASSM 的 " fsl: 0 fnx: 0x0 ver: 0x01 "這一段資料的情況已經改變了

----------------------- 第三部分
Itl Xid Uba Flag Lck Scn/Fsc
0x01 xid: 0x0003.045.000000b4 uba: 0x0080170a.00c7.36 --U- 1 fsc 0x0000.000802a5

--- 這是 oracle 用來記錄事務資訊的部分 這裡顯示的只有一個ITL條 有多少個ITL條是可以動態增加的
--- 只要 block 中的空間足夠 可以定義初始化的 ITL 條的個數 用 INITRANS 這storage 引數
--- 這裡有多少個 ITL 可以從上面 "第二部分" 的 "itc:" 看出來
--- 這部分牽扯 rollback segment 或 undo tablespace
--- Itl itl 的序號
--- xid: 8 bytes 值可以用select XIDUSN, XIDSLOT,XIDSQN from v$transaction;查到
--- This is comprised of the rollback segment number (2 bytes), the slot number
--- in the transaction table of that rollback segment (2 bytes), and the number
--- of times use of that transaction table has wrapped (4 bytes).
--- Uba: 8 bytes The location of the undo for the most recent change to this block by this
--- transaction. This is comprised of the DBA of the rollback segment block (4
--- bytes), the sequence number (2 bytes), and the record number for the change
--- in that undo block (1 byte), plus 1 unused byte.
--- flag 1 nibble ---- = transaction is active, or committed pending cleanout
--- C--- = transaction has been committed and locks cleaned out
--- -B-- = this undo record contains the undo for this ITL entry
--- --U- = transaction committed (maybe long ago); SCN is an upper bound
--- ---T = transaction was still active at block cleanout SCN
--- Lck 3 nibbles The number of row-level locks held in the block by this transaction.
--- Scn/Fsc 6 bytes If the transaction has been cleaned out, this is the commit SCN or an upper
--- bound thereof. Otherwise the leading two bytes contain the free space credit
--- for the transaction - that is, the number of bytes freed in the block by the
--- transaction
--- 參考

---

----------------------- 第四部分
data_block_dump
===============
tsiz: 0x1fb8
hsiz: 0x14
pbl: 0x02476c44
bdba: 0x01000003
flag=-----------
ntab=1
nrow=1
frre=-1
fsbo=0x14
fseo=0x1fb2
avsp=0x1f9b
tosp=0x1f9b
0xe:pti[0] nrow=1 offs=0
0x12:pri[0] offs=0x1fb2

--- tsiz: hsiz: pbl: bdba: 在資料檔案都是沒有儲存的
--- tsiz: 除了上面的3部分和block尾部的4個位元組剩下的空間 0x1fb8就是8120位元組 8k的block:
--- 8192-20(block head)-24(Transaction Header)-24(一個事務條)-4(block tail)=8120
--- hsiz: 資料塊頭20個位元組+資料塊尾4個位元組=24位元組(0x14)
--- pbl: ptr to buffer holding the block 我是用的專用模式dump的datafile中的block出來
--- 在同一個session的dump檔案中 dump 出來的block 的這個都是同一個值
--- bdba: 和第一部分中的rdba 一個意思
--- flag 1 byte N=pctfree hit(clusters), F=don't put on free list
--- K=flushable cluster keys. 當然還有別的標記: A ...
--- ntab 1 byte 這block中有幾個table的資料 cluster這個就可能大於1
--- nrow 2 bytes block 中有多少行資料
--- frre 2 bytes First free row index entry. -1=you have to add one.
--- fsbo 2 bytes Free Space Begin offset
--- fseo 2 bytes Free Space End offset
--- avsp 2 bytes Available space in the block
--- tosp 2 bytes Total available space when all TXs commit
--- 0xe: nrow 2 bytes block 中的這個table有多少行資料
--- 0xe: offs 2 bytes 偏移量 用 cluster 的時候可以看出值

----------------------- 第五部分
block_row_dump:
tab 0, row 0, @0x1fb2
tl: 6 fb: --H-FL-- lb: 0x1 cc: 1
col 0: [ 2] c1 02
end_of_block_dump

--- tl: 這條記錄中的長度 包括row head 的一般情況的 3 位元組和表示資料長度的1位元組和資料本身的長度
--- fb: 1 byte K = Cluster Key (Flags may change meaning if this is set to show HASH cluster)
--- C = Cluster table member
--- H = Head piece of row
--- D = Deleted row
--- F = First data piece
--- L = Last data piece
--- P = First column continues from previous piece
--- N = Last column continues in next piece
--- lb: 1 byte 和上面第三部分的 ITL 的lck相對應 表示這行是否被 lock 了
--- cc: 1 byte 表示這行有幾列資料
--- col 0: [ 2] : 1 byte 表示這行的這列的長度
--- c1 02 : 這就是table中的資料 "1" 可以透過下面的語句看 oracle真正使用的是
--- 那些16進位制的資料來表示的使用者資料 select dump(col_name,16) from table_name;

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

下面的資料就是上面的 dump 檔案對應的2進位制檔案 我依照 2 進位制資料的順序解釋內容

------------------------ 下面的對應上面dump 檔案的第一部分
Offset 0 1 2 3 4 5 6 7 8 9 A B C D E F

00006000 06 02 00 00 03 00 00 01 A5 02 08 00 00 00 01 02
00006010 00 00 00 00

--- 06 : type: 0x06=trans data (對應上面dump檔案中的內容)
--- 02 : frmt: 0x02
--- 00 00 : filler 應該是 unused 上面的dump檔案中並沒有顯示這部分內容
--- 03 00 00 01 : rdba: 0x01000003 在同一個 tablespace 中是唯一的 當 datafile 中可能不唯一
--- 來看看他的規律 :連續建4個 datafile 他們的第一block為
--- 0x 01 00 04 01
--- 0x 01 00 08 01
--- 0x 01 00 0c 01
--- 0x 01 00 00 02
--- 對應的資料在是 0x01040001 , 0x01080001 , 0x010c0001 , 0x02000001
--- 這就可以看出一個 datafile 的最大的容量是 0x01040001~0x0107ffff 總共是
--- 4M(因為還有一個datafile head)個block 如果block 是8k 一個datafile就是 32G
--- 這就是oracle 的文件上面對 oracle 的 datafile (8k的時候)最大 32G 的原因
--- 同樣可以看出一個 tablespace 的datafile 可以從 0x0100 到 0xfffe 就是1023
--- 個 datafile 這也是 oracle 的文件上面說一個tablesapce可以有1022 個datafile
--- 的原因(其實一個 tablespace 可以有 1023 個 datafile 我建過)
--- oracle10G中 block 的big datafile 的這段資料的情況是0x00000001~0xffffffff
--- 這也是為什麼一個 big datafile 的tablesapce只有一個datafile 並且最大值為
--- 32T (8k的時候 : 4G*8k=32T ; 32k的資料塊的時候為128T)
--- A5 02 08 00 00 00 : scn: 0x0000.000802a5 oracle是c語言寫的 這6位不正好是一個 unsigned long
--- 和 unsigned int的組合
--- 01 : seq: 0x01
--- 02 : flg: 0x02
--- 00 00 : chkval: 0x0000 在 init 中設定了 db_block_checksum=true 才有值
--- 00 00 : unused 上面的dump檔案中並沒有顯示這部分內容
--- 上面的20個bytes的資料任改其中的值肯定發生ORA-XXXXX(不一定就是ora-01578 我還看見過ora-600)


------------------------ 下面的對應上面dump 檔案的第二部分

01 00 00 00 4A 61 00 00 A3 02 08 00
00006020 00 00 00 00 01 00 03 00 00 00 00 00

--- 01 : typ: 1 - DATA
--- 00 : 只見過 0x00 沒見過其他的值 don't know
--- 00 : 見過有其他值 但用編輯器改這個值 在 dump 檔案中顯示不出來變化
--- 00 : 沒見變過其他的值 don't know
--- 4A 61 00 00 : seg/obj: 0x614a
--- A3 02 08 00 00 00 : csc: 0x00.802a3
--- 00 00 : 見過有其他值 但用編輯器改這個值 在 dump 檔案中顯示不出來變化
--- 01 : itc: 1 下面的 ITL 條目的個數
--- 00 : 見過有其他值 但用編輯器改這個值 在 dump 檔案中顯示不出來變化
--- 03 : flg: O ver: 0x01 Object id on Block? Y
--- 從我的觀察中 dump 出來的檔案中 flg ver Object id on Block
--- 他們共同佔用的這個一個位元組 他的規律可以從下面的情況看出
--- 2進位制資料 flg ver Object id on Block?
--- 0x00 - 0x00 N
--- 0x01 0 0x00 N
--- 0x02 - 0x01 Y
--- 0x03 0 0x01 Y
--- 0x04 - 0x02 Y
--- 0x05 0 0x02 Y
--- 0x06 - 0x03 Y
--- 0x07 0 0x03 Y
--- 0x08 - 0x04 N
--- 0x09 0 0x04 N
--- 0x0a - 0x05 Y
--- 0x0b 0 0x05 Y
--- 0x0c - 0x06 Y
--- 0x0d 0 0x06 Y
--- 0x0e - 0x07 Y
--- 0x0f 0 0x07 Y
--- 0x10 ... 類似上面的迴圈了 這種情況在9i上已經改變因為ASSM的出現
--- 00 : fsl: 0
--- 00 00 00 00 : fnx: 0x0

------------------------ 下面的對應上面dump 檔案的第三部分

03 00 45 00
00006030 B4 00 00 00 0A 17 80 00 C7 00 36 00 01 20 00 00
00006040 A5 02 08 00

--- 03 00 45 00 B4 00 00 00 : xid: 0x0003.045.000000b4
--- 0A 17 80 00 C7 00 36 00 : uba: 0x0080170a.00c7.36
--- 01 0 : Lck 鎖定的row數 這裡還用到了下一個 byte 的資料
--- 2 : Flag 2 對應的二進位制表示為 0010 正好和dump檔案中的 --U- 吻合
--- 00 00 A5 02 08 00 : Scn/Fsc

------------------------ 下面的對應上面dump 檔案的第四部分

00 01 01 00 FF FF 14 00 B2 1F 9B 1F
00006050 9B 1F 00 00 01 00 B2 1F

--- 00 : flag
--- 01 : ntab
--- 01 00 : nrow
--- FF FF : frre
--- 14 00 : fsbo
--- B2 1F : fseo
--- 9B 1F : avsp
--- 9B 1F : tosp
--- 00 00 : 0xe: offs
--- 01 00 : 0xe: nrow
--- B2 1F : 0x12:pri[0] offs=0x1fb2

------------------------ 下面的對應上面dump 檔案的第五部分

--- 這部分和上面的資料中間省略了很多 因為這列子中這些部分沒儲存資料

00007FF0 00 00 00 00 00 00 2C 01 01 02 C1 02

--- 這是 block 中存使用者資料的地方
--- 2C : fb:
--- 01 : lb: 這一行是否被lock
--- 01 : 這條記錄中有多少列的資料 從這裡看出因為只用一個byte去記錄這一行有多少列
--- 所以最多是255列 但一個表可以最多是1000列 如果table 的列大於255列 這裡就會
--- 發生連結 根據一行記錄的長度來看是 塊內的連結或者是塊與塊直接的連結
--- 這也是為什麼table 的設計會盡量少於255列的原因
--- 02 : 第1列的資料的長度是多少
--- C1 02 : 儲存在 block 中的資料 "1"

------------------------ 下面的對應上面dump 檔案的第一部分的 "tail: 0x02a50601"
01 06 A5 02
--- 這是用來效驗 block 是否完整的標誌 改這 block 最後的4 bytes 資料中的任意肯定ora-1578

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
上面只是講解了一個最簡單的oracle 的block 的格式 其他還有很多的情況 如 cluster , index 等
long , lob , long raw 這些在 block 中的表示因為資料長度的不同都會有不同的體現

有的時候改動block 中的任意資料不一定會發生 ORA-01578 因為下面的兩個引數沒有設定為 true
db_block_checking db_block_checksum
這兩個引數對於system tablespace 預設都是true 所以改變 system tablespace 中的
資料肯定會發生ORA-01578
隨著oracle版本的不斷升級 oracle 對block中儲存資料的正確性也在不斷的提升要求 而且透過常規的
手段想去了解 oracle block 的格式也可能越來越困難

來自 “ ITPUB部落格 ” ,連結:http://blog.itpub.net/35489/viewspace-84771/,如需轉載,請註明出處,否則將追究法律責任。

相關文章