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

Re: openldap db wiped?

2.3 alpha was released January 1st. Obviously as an alpha it's still going through some shakedown.

As always, your requests for changes should be submitted via ITS so they don't get lost.

Also, changes that are not flushed by a checkpoint are still present in the BDB cache. When you run db_recover, they will be flushed out correctly, so in practice you should never be losing any data. You just have to remember to run db_recover after any unclean shutdown. This is mandatory anyway for correct operation. (With Symas CDS, unclean shutdowns are automatically detected and recovery is performed automatically, so data integrity is always maintained without any outside intervention.)

Armbrust, Daniel C. wrote:

Gak! 

Unless you are planning on releasing the 2.3 branch soon, this seems like a bug that should be backported to 2.2.

At the very minimum, the man pages should be updated to say what was just said below.  Until this thread, I had no idea that this was going on.  At least now I know why I was still losing changes (even after setting the checkpoint) when it doesn't shut down cleanly....

The man pages give the impression that if you set the checkpoint to 5 minutes, that is how often things will be saved to disk.  Now I find out I actually have to set up a cron job to achieve this?  Yuk.

Dan


-----Original Message-----
From: owner-openldap-software@OpenLDAP.org [mailto:owner-openldap-software@OpenLDAP.org] On Behalf Of Howard Chu
Sent: Tuesday, January 11, 2005 9:45 AM
To: Karolis
Cc: Szuts Attila; openldap-software@OpenLDAP.org
Subject: Re: openldap db wiped?

Karolis Dautartas wrote:



what is the difference between running this script every 5 mins and
having a "checkpoint 5 X"   directive in slapd.conf?



In OpenLDAP 2.1 and 2.2 the checkpoint directive acts as follows - *when there is a write operation*, and more than <check> minutes have occurred since the last checkpoint, perform the checkpoint. If more than <check> minutes pass after a write without any other write operations occurring, no checkpoint is performed, so it's possible to lose the last write that occurred.

In OpenLDAP 2.3 this behavior has been changed; checkpoints occur every <check> minutes as I already mentioned in my previous email here.





--
 -- Howard Chu
 Chief Architect, Symas Corp.       Director, Highland Sun
 http://www.symas.com               http://highlandsun.com/hyc
 Symas: Premier OpenSource Development and Support