Re: Power outage db down - need help

[Bob Bankay]

uname -a
Linux jocelyn 2.4.21-27.ELsmp #1 SMP Wed Dec 1 21:53:57 EST 2004 x86_64 x86_64 x86_64 GNU/Linux


I set innodb_forced_recoery=4
The result is below:

[mysql@jocelyn ~]$ bin/mysqld_safe start
Starting mysqld daemon with databases from /mydisks/a/apps/mysql/data/
STOPPING server from pid file /mydisks/a/apps/mysql/jocelyn.pid
050224 09:13:04 mysqld ended

server.err - tail a small bit of page dump included

t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t t= = =O= < <E;t: :}:19 9:8 8C7 3 % &'7r/H0?'}6 6 6U5 5 5%4 4A3 3 3]2 2@1 1 0 0e/ / . . - - -4, ,c+ +Y+ *u*)) ) ( ( ( 'W & c y& &s & % %C$ $ $L# # " "q"% # [ ! " > 4 7 3 O k ; W ' C { / K g 7 S I ? [ w + G 3 O k ; W e w ('(;InnoDB: End of page dump
050224 9:13:04 InnoDB: Page checksum 2719677399, prior-to-4.0.14-form checksum 3386652281
InnoDB: stored checksum 2424901223, prior-to-4.0.14-form stored checksum 3791550236
InnoDB: Page lsn 12 4083660083, low 4 bytes of lsn at page end 4180158248
InnoDB: Page may be an index page where index id is 0 88
InnoDB: Also the page in the doublewrite buffer is corrupt.
InnoDB: Cannot continue operation.
InnoDB: You can try to recover the database with the my.cnf
InnoDB: option:
InnoDB: set-variable=innodb_force_recovery=6
050224 09:13:04 mysqld ended


So it will not start as is.

In the 2nd worst case we may have to remigrate the db from a replica that was 30 minutes behind. This means going from Solaris sparc to RHEL3 x86. (The replica server is on slower Solaris Sparc and it seems to have survived the disaster with very little db damage)
.
In the worst case it will be from a 1 week old backup.