MySQL:RR分析死鎖一列
水平有限 有誤請指出
版本:Percona MySQL 5.7.22
對於鎖的學習我做了一些輸出詳細參考如下:
其中有readme
本文也是一個朋友問我死鎖問題。@越前
一、問題提出
如下構造方式,問為什麼RC模式下不會死鎖,RR模式下死鎖。
drop table tt; CREATE TABLE `tt` ( `id` int(11) NOT NULL, `c1` int(11) DEFAULT NULL, PRIMARY KEY (`id`), UNIQUE KEY `c1` (`c1`) ) ENGINE=InnoDB DEFAULT CHARSET=utf8; insert into tt values(1,1);
session 1 | session 2 |
---|---|
begin; |
|
select * from tt where c1=1 for update; |
|
update tt set id=2 where c1=1; |
|
|
begin;select * from tt where c1=1 for update;堵塞 |
select * from tt where c1=1 for update; |
|
|
死鎖回滾 |
二、分析方式
首先分析session 1 第一句:
select * from tt where c1=1 for update;
執行後的加鎖行為
- RR
---TRANSACTION 231106, ACTIVE 9 sec3 lock struct(s), heap size 1160, 2 row lock(s) MySQL thread id 11, OS thread handle 140737153623808, query id 303 localhost root TABLE LOCK table `test`.`tt` trx id 231106 lock mode IX RECORD LOCKS space id 127 page no 4 n bits 72 index c1 of table `test`.`tt` trx id 231106 lock_mode X(LOCK_X) locks rec but not gap(LOCK_REC_NOT_GAP) Record lock, heap no 2 PHYSICAL RECORD: n_fields 2; compact format; info bits 0 0: len 4; hex 80000001; asc ;; 1: len 4; hex 80000001; asc ;; RECORD LOCKS space id 127 page no 3 n bits 72 index PRIMARY of table `test`.`tt` trx id 231106 lock_mode X(LOCK_X) locks rec but not gap(LOCK_REC_NOT_GAP) Record lock, heap no 2 PHYSICAL RECORD: n_fields 4; compact format; info bits 0 0: len 4; hex 80000001; asc ;; 1: len 6; hex 0000000386c0; asc ;; 2: len 7; hex aa0000003f0110; asc ? ;; 3: len 4; hex 80000001; asc ;;
- RC
---TRANSACTION 231105, ACTIVE 7 sec3 lock struct(s), heap size 1160, 2 row lock(s) MySQL thread id 11, OS thread handle 140737153623808, query id 295 localhost root TABLE LOCK table `test`.`tt` trx id 231105 lock mode IX RECORD LOCKS space id 127 page no 4 n bits 72 index c1 of table `test`.`tt` trx id 231105 lock_mode X(LOCK_X) locks rec but not gap(LOCK_REC_NOT_GAP) Record lock, heap no 2 PHYSICAL RECORD: n_fields 2; compact format; info bits 0 0: len 4; hex 80000001; asc ;; 1: len 4; hex 80000001; asc ;; RECORD LOCKS space id 127 page no 3 n bits 72 index PRIMARY of table `test`.`tt` trx id 231105 lock_mode X(LOCK_X) locks rec but not gap(LOCK_REC_NOT_GAP) Record lock, heap no 2 PHYSICAL RECORD: n_fields 4; compact format; info bits 0 0: len 4; hex 80000001; asc ;; 1: len 6; hex 0000000386c0; asc ;; 2: len 7; hex aa0000003f0110; asc ? ;; 3: len 4; hex 80000001; asc ;;
我們可以看到因為 c1是主鍵因此加鎖方式不管怎麼樣都是LOCK_X|LOCK_REC_NOT_GAP,主鍵上也是同樣的。就是鎖住了二級唯一索引和主鍵的相關記錄。
然後分析session 1 第二句:
update tt set id=2 where c1=1;
執行後的加鎖行為
這一句比較重要,在二級 唯一索引c1 上會做一個刪除和插入操作,也就是會將 原來的1,1記錄標記為del flag ,同時插入2,1這條記錄,這會引起一個鎖的繼承操作(lock_rec_inherit_to_gap_if_gap_lock呼叫會出現GAP LOCK),但是之前還會涉及到唯一性檢查因此還涉及到LOCK_S鎖和next key lock的存在(對於二級索引做唯一性檢查始終是next key lock)。這裡的del flag也是形成死鎖的重要原因。(對於二級索引的update操作總是先刪除然後插入記錄,主鍵則會進行判斷是否可以容下新的記錄)
- RR
---TRANSACTION 231106, ACTIVE 1626 sec5 lock struct(s), heap size 1160, 5 row lock(s), undo log entries 2MySQL thread id 11, OS thread handle 140737153623808, query id 305 localhost root TABLE LOCK table `test`.`tt` trx id 231106 lock mode IX RECORD LOCKS space id 127 page no 4 n bits 72 index c1 of table `test`.`tt` trx id 231106 lock_mode X(LOCK_X) locks rec but not gap(LOCK_REC_NOT_GAP) Record lock, heap no 2 PHYSICAL RECORD: n_fields 2; compact format; info bits 32 0: len 4; hex 80000001; asc ;; 1: len 4; hex 80000001; asc ;; RECORD LOCKS space id 127 page no 3 n bits 72 index PRIMARY of table `test`.`tt` trx id 231106 lock_mode X(LOCK_X) locks rec but not gap(LOCK_REC_NOT_GAP) Record lock, heap no 2 PHYSICAL RECORD: n_fields 4; compact format; info bits 32 0: len 4; hex 80000001; asc ;; 1: len 6; hex 0000000386c2; asc ;; 2: len 7; hex 2c000000410dca; asc , A ;; 3: len 4; hex 80000001; asc ;; RECORD LOCKS space id 127 page no 4 n bits 72 index c1 of table `test`.`tt` trx id 231106 lock mode S(LOCK_S) locks gap and rec(LOCK_ORDINARY[next_key_lock]) Record lock, heap no 1 PHYSICAL RECORD: n_fields 1; compact format; info bits 0 0: len 8; hex 73757072656d756d; asc supremum;; Record lock, heap no 2 PHYSICAL RECORD: n_fields 2; compact format; info bits 32 0: len 4; hex 80000001; asc ;; 1: len 4; hex 80000001; asc ;; RECORD LOCKS space id 127 page no 4 n bits 72 index c1 of table `test`.`tt` trx id 231106 lock mode S(LOCK_S) locks gap before rec(LOCK_GAP) Record lock, heap no 3 PHYSICAL RECORD: n_fields 2; compact format; info bits 0 0: len 4; hex 80000001; asc ;; 1: len 4; hex 80000002; asc ;;
- RC
5 lock struct(s), heap size 1160, 5 row lock(s), undo log entries 2MySQL thread id 11, OS thread handle 140737153623808, query id 316 localhost root TABLE LOCK table `test`.`tt` trx id 231123 lock mode IX RECORD LOCKS space id 128 page no 4 n bits 72 index c1 of table `test`.`tt` trx id 231123 lock_mode X(LOCK_X) locks rec but not gap(LOCK_REC_NOT_GAP) Record lock, heap no 2 PHYSICAL RECORD: n_fields 2; compact format; info bits 32 0: len 4; hex 80000001; asc ;; 1: len 4; hex 80000001; asc ;; RECORD LOCKS space id 128 page no 3 n bits 72 index PRIMARY of table `test`.`tt` trx id 231123 lock_mode X(LOCK_X) locks rec but not gap(LOCK_REC_NOT_GAP) Record lock, heap no 2 PHYSICAL RECORD: n_fields 4; compact format; info bits 32 0: len 4; hex 80000001; asc ;; 1: len 6; hex 0000000386d3; asc ;; 2: len 7; hex 370000003206e2; asc 7 2 ;; 3: len 4; hex 80000001; asc ;; RECORD LOCKS space id 128 page no 4 n bits 72 index c1 of table `test`.`tt` trx id 231123 lock mode S(LOCK_S) locks gap and rec(LOCK_ORDINARY[next_key_lock]) Record lock, heap no 1 PHYSICAL RECORD: n_fields 1; compact format; info bits 0 0: len 8; hex 73757072656d756d; asc supremum;; Record lock, heap no 2 PHYSICAL RECORD: n_fields 2; compact format; info bits 32 0: len 4; hex 80000001; asc ;; 1: len 4; hex 80000001; asc ;; RECORD LOCKS space id 128 page no 4 n bits 72 index c1 of table `test`.`tt` trx id 231123 lock mode S(LOCK_S) locks gap before rec(LOCK_GAP) Record lock, heap no 3 PHYSICAL RECORD: n_fields 2; compact format; info bits 0 0: len 4; hex 80000001; asc ;; 1: len 4; hex 80000002; asc ;;
到這裡RR和RC加鎖並沒有什麼不同,因為都是唯一值,同時鎖繼承也都是二級索引上的都是LOCK_S|LOCK_ORDINARY[next_key_lock],但是下面就會出現不同了。
然後分析session 2的第一句:
select * from tt where c1=1 for update;
實際上這個時候存在2條c1=1的記錄只有1,1標記為刪除了,1,2沒有提交,都是需要訪問的。
然後堵塞行為為:
- RR
LOCK WAIT 2 lock struct(s), heap size 1160, 1 row lock(s) MySQL thread id 10, OS thread handle 140737153824512, query id 350 localhost root statistics select * from tt where c1=1 for update ------- TRX HAS BEEN WAITING 11 SEC FOR THIS LOCK TO BE GRANTED: RECORD LOCKS space id 129 page no 4 n bits 72 index c1 of table `test`.`tt` trx id 231146 lock_mode X(LOCK_X) locks gap and rec(LOCK_ORDINARY[next_key_lock]) waiting(LOCK_WAIT) Record lock, heap no 2 PHYSICAL RECORD: n_fields 2; compact format; info bits 32 0: len 4; hex 80000001; asc ;; 1: len 4; hex 80000001; asc ;;
- RC
LOCK WAIT 2 lock struct(s), heap size 1160, 1 row lock(s) MySQL thread id 10, OS thread handle 140737153824512, query id 325 localhost root statistics select * from tt where c1=1 for update ------- TRX HAS BEEN WAITING 9 SEC FOR THIS LOCK TO BE GRANTED: RECORD LOCKS space id 128 page no 4 n bits 72 index c1 of table `test`.`tt` trx id 231128 lock_mode X(LOCK_X) locks rec but not gap(LOCK_REC_NOT_GAP) waiting(LOCK_WAIT) Record lock, heap no 2 PHYSICAL RECORD: n_fields 2; compact format; info bits 32 0: len 4; hex 80000001; asc ;; 1: len 4; hex 80000001; asc ;;
我們這裡可以看到對於RR模式下唯一鍵c1的1,1已經刪除了。我做了debug發現這裡會在函式中row_search_mvcc加鎖前做判斷如下:
if (!set_also_gap_locks || srv_locks_unsafe_for_binlog || trx->isolation_level <= TRX_ISO_READ_COMMITTED || (unique_search && !rec_get_deleted_flag(rec, comp)) || dict_index_is_spatial(index)) { goto no_gap_lock; } else { lock_type = LOCK_ORDINARY; }
我們拋開其他來分析這兩句
-
trx->isolation_level <= TRX_ISO_READ_COMMITTED
如果是RC模式則直接上LOCK_REC_NOT_GAP及只鎖住記錄本身 -
(unique_search && !rec_get_deleted_flag(rec, comp))
如果不是RC,如果是二級唯一索引並且沒有被標記為del flag則標記為LOCK_REC_NOT_GAP。但是如果標記為del flag則標記為LOCK_ORDINARY就是next key lock。
分析session 1的最後一個語句也就是產生死鎖的語句:
select * from tt where c1=1 for update;
如上這個語句會訪問1,1標記為刪除了,1,2沒有提交 的兩個記錄。這個時候就出現了不同。
-
RC
只需要唯一索引 1,1上 LOCK_REC_NOT_GAP|LOCK_X 及記錄索引,這個鎖在本事物的第一句語句上已經獲得了,因此直接透過了,不需要做檢測。 -
RR
需要在唯一索引 1,1上 LOCK_ORDINARY|LOCK_X 也就是就是next key lock。這個鎖在本事物中並沒有獲取過,因此需要重新的檢測所的相容性,最終加入了等待列表。
我們來看一下函式lock_rec_lock_slow,我做debug的時候明顯看到了不同的邏輯:
if (lock_rec_has_expl(mode, block, heap_no, trx)) { /* The trx already has a strong enough lock on rec: do 1,1 key lock RR NEX KEY LOCK stronager nothing */ lock_rec_print(mode,block,heap_no,index,thr_get_trx(thr)); err = DB_SUCCESS; } else { const lock_t* wait_for = lock_rec_other_has_conflicting( mode, block, heap_no, trx,index); if (wait_for != NULL) { /* If another transaction has a non-gap conflicting request in the queue, as this transaction does not have a lock strong enough already granted on the record, we may have to wait. */ RecLock rec_lock(thr, index, block, heap_no, mode); err = rec_lock.add_to_waitq(wait_for); }
三、總結
最終RR下形成了環路
- session1 首先獲得唯一索引上的 1,1記錄的 LOCK_REC_NOT_GAP|LOCK_X
- 然後session 1做update 獲得唯一索引上的 1,1記錄的 LOCK_ORDINARY(next key lock)|LOCK_S
- 然後session 2需要獲取唯一索引上的 1,1記錄的 LOCK_ORDINARY(next key lock)|LOCK_X 發生等待
- 然後session 1需要獲取唯一索引上的 1,1記錄的 LOCK_ORDINARY(next key lock)|LOCK_X 加入等待佇列進行等待
死鎖發生因此發生,而RC模式下最後兩部需要獲取的都是 LOCK_REC_NOT_GAP|LOCK_X,雖然session 2處於等待但是session因為已經獲得相同級別的鎖不會在進行檢測加鎖等待,而直接透過了。
下面是死鎖的記錄:
*** (1) WAITING FOR THIS LOCK TO BE GRANTED: RECORD LOCKS space id 117 page no 4 n bits 72 index c1 of table `t1`.`tt` trx id 230530 lock_mode X(LOCK_X) locks gap and rec(LOCK_ORDINARY[next_key_lock]) waiting(LOCK_WAIT) Record lock, heap no 2 PHYSICAL RECORD: n_fields 2; compact format; info bits 32 0: len 4; hex 80000001; asc ;; 1: len 4; hex 80000001; asc ;; *** (2) TRANSACTION: TRANSACTION 230525, ACTIVE 68 sec starting index read mysql tables in use 1, locked 16 lock struct(s), heap size 1160, 6 row lock(s), undo log entries 2MySQL thread id 6, OS thread handle 140737153423104, query id 156 localhost root statistics select * from tt where c1=1 for update *** (2) HOLDS THE LOCK(S): RECORD LOCKS space id 117 page no 4 n bits 72 index c1 of table `t1`.`tt` trx id 230525 lock_mode X(LOCK_X) locks rec but not gap(LOCK_REC_NOT_GAP) Record lock, heap no 2 PHYSICAL RECORD: n_fields 2; compact format; info bits 32 0: len 4; hex 80000001; asc ;; 1: len 4; hex 80000001; asc ;; *** (2) WAITING FOR THIS LOCK TO BE GRANTED: RECORD LOCKS space id 117 page no 4 n bits 72 index c1 of table `t1`.`tt` trx id 230525 lock_mode X(LOCK_X) locks gap and rec(LOCK_ORDINARY[next_key_lock]) waiting(LOCK_WAIT) Record lock, heap no 2 PHYSICAL RECORD: n_fields 2; compact format; info bits 32 0: len 4; hex 80000001; asc ;; 1: len 4; hex 80000001; asc ;; *** WE ROLL BACK TRANSACTION (1)
四、棧幀參考
最後留下幾個棧幀以備分析使用
- 鎖繼承
#0 lock_rec_set_nth_bit (lock=0x30b1230, i=3) at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/include/lock0priv.ic:91#1 0x0000000001a5d44a in RecLock::lock_alloc (trx=0x7fffd7803b10, index=0x7ffe7459ff80, mode=546, rec_id=..., size=9) at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/lock/lock0lock.cc:1484#2 0x0000000001a5d826 in RecLock::create (this=0x7fffec0c0eb0, trx=0x7fffd7803b10, owns_trx_mutex=false, add_to_hash=true, prdt=0x0) at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/lock/lock0lock.cc:1537#3 0x0000000001a5e60c in lock_rec_add_to_queue (type_mode=546, block=0x7fff9adb8150, heap_no=3, index=0x7ffe7459ff80, trx=0x7fffd7803b10, caller_owns_trx_mutex=false) at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/lock/lock0lock.cc:1853#4 0x0000000001a611ec in lock_rec_inherit_to_gap_if_gap_lock (block=0x7fff9adb8150, heir_heap_no=3, heap_no=1) at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/lock/lock0lock.cc:2829#5 0x0000000001a62ea3 in lock_update_insert (block=0x7fff9adb8150, rec=0x7fff9b9c408c "\200") at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/lock/lock0lock.cc:3659#6 0x0000000001c53c25 in btr_cur_optimistic_insert (flags=0, cursor=0x7fffec0c23f0, offsets=0x7fffec0c24c8, heap=0x7fffec0c13e0, entry=0x7ffe7403bb70, rec=0x7fffec0c24c0, big_rec=0x7fffec0c24b8, n_ext=0, thr=0x7ffe7403ba00, mtr=0x7fffec0c1bc0) at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/btr/btr0cur.cc:3346#7 0x0000000001b195fe in row_ins_sec_index_entry_low (flags=0, mode=2, index=0x7ffe7459ff80, offsets_heap=0x7ffe7403bf98, heap=0x7ffe7403c448, entry=0x7ffe7403bb70, trx_id=0, thr=0x7ffe7403ba00, dup_chk_only=false) at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/row/row0ins.cc:3166#8 0x0000000001b1a15e in row_ins_sec_index_entry (index=0x7ffe7459ff80, entry=0x7ffe7403bb70, thr=0x7ffe7403ba00, dup_chk_only=false) at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/row/row0ins.cc:3421#9 0x0000000001b9e053 in row_upd_sec_index_entry (node=0x7ffe7403b6f8, thr=0x7ffe7403ba00) at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/row/row0upd.cc:2337#10 0x0000000001b9e1c3 in row_upd_sec_step (node=0x7ffe7403b6f8, thr=0x7ffe7403ba00) at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/row/row0upd.cc:2364
- RR加鎖del flag記錄
#0 lock_rec_set_nth_bit (lock=0x30b28b8, i=2) at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/include/lock0priv.ic:91#1 0x0000000001a5d44a in RecLock::lock_alloc (trx=0x7fffd7804080, index=0x7ffe74064ea0, mode=259, rec_id=..., size=9) at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/lock/lock0lock.cc:1484#2 0x0000000001a5d826 in RecLock::create (this=0x7fffec0c1e00, trx=0x7fffd7804080, owns_trx_mutex=true, add_to_hash=true, prdt=0x0) at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/lock/lock0lock.cc:1537#3 0x0000000001a5e1c4 in RecLock::add_to_waitq (this=0x7fffec0c1e00, wait_for=0x30b0e58, prdt=0x0) at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/lock/lock0lock.cc:1731#4 0x0000000001a5ee37 in lock_rec_lock_slow (impl=0, mode=3, block=0x7fff4d027b20, heap_no=2, index=0x7ffe74064ea0, thr=0x7ffe7459ff60) at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/lock/lock0lock.cc:2004#5 0x0000000001a5f1ce in lock_rec_lock (impl=false, mode=3, block=0x7fff4d027b20, heap_no=2, index=0x7ffe74064ea0, thr=0x7ffe7459ff60) at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/lock/lock0lock.cc:2082#6 0x0000000001a69a02 in lock_sec_rec_read_check_and_lock (flags=0, block=0x7fff4d027b20, rec=0x7fff4da8c07e "\200", index=0x7ffe74064ea0, offsets=0x7fffec0c2690, mode=LOCK_X, gap_mode=0, thr=0x7ffe7459ff60) at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/lock/lock0lock.cc:6457#7 0x0000000001b717f7 in sel_set_rec_lock (pcur=0x7ffe7459f6d0, rec=0x7fff4da8c07e "\200", index=0x7ffe74064ea0, offsets=0x7fffec0c2690, mode=3, type=0, thr=0x7ffe7459ff60, mtr=0x7fffec0c2180) at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/row/row0sel.cc:1270#8 0x0000000001b7ab6a in row_search_mvcc (buf=0x7ffe7405b9c0 "\375\002", mode=PAGE_CUR_GE, prebuilt=0x7ffe7459f4b0, match_mode=1, direction=0) at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/row/row0sel.cc:5591
- RC加鎖del flag記錄
#0 lock_rec_lock_slow (impl=0, mode=1027, block=0x7fff3310cdf0, heap_no=2, index=0x7ffe74076d90, thr=0x7ffe7459fc20) at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/lock/lock0lock.cc:1962#1 0x0000000001a5f1ce in lock_rec_lock (impl=false, mode=1027, block=0x7fff3310cdf0, heap_no=2, index=0x7ffe74076d90, thr=0x7ffe7459fc20) at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/lock/lock0lock.cc:2082#2 0x0000000001a69a02 in lock_sec_rec_read_check_and_lock (flags=0, block=0x7fff3310cdf0, rec=0x7fff33bdc07e "\200", index=0x7ffe74076d90, offsets=0x7fffec0c2690, mode=LOCK_X, gap_mode=1024, thr=0x7ffe7459fc20) at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/lock/lock0lock.cc:6457#3 0x0000000001b717f7 in sel_set_rec_lock (pcur=0x7ffe7459f6d0, rec=0x7fff33bdc07e "\200", index=0x7ffe74076d90, offsets=0x7fffec0c2690, mode=3, type=1024, thr=0x7ffe7459fc20, mtr=0x7fffec0c2180) at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/row/row0sel.cc:1270#4 0x0000000001b7ab6a in row_search_mvcc (buf=0x7ffe7403ae80 "\375\002", mode=PAGE_CUR_GE, prebuilt=0x7ffe7459f4b0, match_mode=1, direction=0) at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/row/row0sel.cc:5591#5 0x00000000019d5493 in ha_innobase::index_read (this=0x7ffe7403cda0, buf=0x7ffe7403ae80 "\375\002", key_ptr=0x7ffe74095600 "", key_len=5, find_flag=HA_READ_KEY_EXACT) at /root/mysqlall/percona-server-locks-detail-5.7.22/storage/innobase/handler/ha_innodb.cc:9536#6 0x0000000000f934aa in handler::index_read_map (this=0x7ffe7403cda0, buf=0x7ffe7403ae80 "\375\002", key=0x7ffe74095600 "", keypart_map=1, find_flag=HA_READ_KEY_EXACT) at /root/mysqlall/percona-server-locks-detail-5.7.22/sql/handler.h:2942
作者微信:gp_22389860
來自 “ ITPUB部落格 ” ,連結:http://blog.itpub.net/7728585/viewspace-2638386/,如需轉載,請註明出處,否則將追究法律責任。
相關文章
- MySQL:RR模式下死鎖一列MySql模式
- MySQL:一個死鎖分析 (未分析出來的死鎖)MySql
- 故障分析 | MySQL死鎖案例分析MySql
- MySQL 死鎖問題分析MySql
- MySQL鎖等待與死鎖問題分析MySql
- MySQL批量更新死鎖案例分析MySql
- MySQL死鎖分析與解決之路MySql
- MySQL死鎖系列-常見加鎖場景分析MySql
- 線上BUG:MySQL死鎖分析實戰MySql
- MySQL 死鎖和鎖等待MySql
- MySQL死鎖案例分析一(先delete,再insert,導致死鎖)MySqldelete
- 死鎖案例分析
- MySQL rr下幻讀問題分析MySql
- 面試:什麼是死鎖,如何避免或解決死鎖;MySQL中的死鎖現象,MySQL死鎖如何解決面試MySql
- MySQL 死鎖解決MySql
- MySQL解決死鎖MySql
- MySQL死鎖問題MySql
- 一次 MySQL 線上死鎖分析實戰MySql
- MySQL死鎖系列-線上死鎖問題排查思路MySql
- GreatSQL 死鎖案例分析SQL
- mysql行鎖和死鎖檢測MySql
- Mysql 兩階段鎖和死鎖MySql
- 手把手教你分析解決MySQL死鎖問題MySql
- SQLServer的死鎖分析(1):頁鎖SQLServer
- mysql死鎖最佳化MySql
- MySQL:死鎖一例MySql
- 【MySQL】死鎖案例之六MySql
- 【MySQL】死鎖案例之七MySql
- 【MySQL】死鎖案例之八MySql
- Mysql如何處理死鎖MySql
- MySQL列印死鎖日誌MySql
- MySQL:MTS和mysqldump死鎖MySql
- SQL SERVER死鎖查詢,死鎖分析,解鎖,查詢佔用SQLServer
- RR與RC隔離級別下MySQL不同的加鎖解鎖方式MySql
- MySQL:Innodb 一個死鎖案例MySql
- MySQL insert on duplicate key update 死鎖MySql
- MySQL死鎖案例一(回滾導致死鎖)MySql
- MySQL死鎖案例二(自增列導致死鎖)MySql