Hello.
If I use the checkpointing feature in the bdb backend, if there were no changes made to the directory since the previous checkpoint, is there any overhead other than waking up the code to check to see if the database is to be synced?
If there are no unsynced changes, does anything get written to disk or logged in any way when the checkpointing code wakes up?
Thank you.
____________________________________________________________________________________ Bored stiff? Loosen up... Download and play hundreds of games for free on Yahoo! Games. http://games.yahoo.com/games/front
On 6/13/07, Bill Johnstone beejstone3@yahoo.com wrote:
Hello.
If I use the checkpointing feature in the bdb backend, if there were no changes made to the directory since the previous checkpoint, is there any overhead other than waking up the code to check to see if the database is to be synced?
If there are no unsynced changes, does anything get written to disk or logged in any way when the checkpointing code wakes up?
When you specify the _min_ argument to checkpoint, it will do something regardless of wether there were transactions or not, but I'm not actually sure how much it does.
If you set _min_ to 0 then it will only flush at _kbyte_, which sounds more like what you're looking for.
See slapd-bdb(5) http://www.openldap.org/software/man.cgi?query=slapd-bdb&apropos=0&s...
Bill Johnstone wrote:
Hello.
If I use the checkpointing feature in the bdb backend, if there were no changes made to the directory since the previous checkpoint, is there any overhead other than waking up the code to check to see if the database is to be synced?
If there are no unsynced changes, does anything get written to disk or logged in any way when the checkpointing code wakes up?
A timestamp will be written to the transaction log files.
openldap-software@openldap.org
-
Bill Johnstone -
Howard Chu -
matthew sporleder