[Date Prev][Date Next] [Chronological] [Thread] [Top]

BDB Corruption problems



Hiya All,

I have been having some database corruption issues for the past few months. Here are the details:

I run

@(#) $OpenLDAP: slapd 2.2.26 (Jul 15 2005 14:05:34) $

with one master and one syncrepl slave that updates every 60 seconds.

About once every two weeks either the master or slave bdb becomes corrput, there are no
abnormal shutdowns or power failures. Both the master and slave are busy with reads and the
master has a write around every 10 minutes.


I just had the master slapd processes all running at very high processor utilisation and refusing to
answer queries. I slapcatted the database and slapcat died about half way through. db4_verify:


root@server1.odb/database/openldap/openldap-data# db4_verify PrimaryUser.bdb
db_verify: Page 2: out-of-order key at entry 2
db_verify: Page 2: out-of-order key at entry 5
db_verify: Page 2: out-of-order key at entry 6
db_verify: Page 2: out-of-order key at entry 7
db_verify: Page 2: out-of-order key at entry 10
db_verify: Page 2: out-of-order key at entry 12
db_verify: Page 4: out-of-order key at entry 116
db_verify: Page 4: out-of-order key at entry 288
db_verify: Page 6: out-of-order key at entry 67
db_verify: Page 6: out-of-order key at entry 263
db_verify: Page 7: out-of-order key at entry 65
db_verify: Page 7: out-of-order key at entry 275
db_verify: Page 8: out-of-order key at entry 11
db_verify: Page 8: out-of-order key at entry 82
db_verify: Page 8: out-of-order key at entry 166
db_verify: Page 8: out-of-order key at entry 238
db_verify: Page 8: out-of-order key at entry 298
db_verify: Page 8: out-of-order key at entry 347

......

There is similar output on all dbd files. What does this mean?

I recovered by grabbing a slapcat from the slave, slapadding it to the master and restarting it all.

What would cause such db4 issues? Should I be using 2.3 versions of OpenLdap?

Thanks folks!

--
Leigh