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

Re: Two contextCSNs

Peter Mogensen wrote:
Howard Chu wrote:
There is no need for your step #2.

(step #2 is removing all entryCSN, contextCSN lines).

I did so because the SID of CSN values from the 2.3.30 dump is 00:
entryCSN: 20071214130312Z#000000#00#000000

Current SID is 001 or 002.

Given a valid slapcat from OpenLDAP 2.3 you should be able to slapadd it
directly in 2.4 without using -S or -w in your step #3. Therefore you
don't need step #4.

Then I misunderstood your post:

I read it as the SID of "00" from 2.3 was not a correct CSN from a 2.4

It's fine. OpenLDAP 2.4 will accept CSNs from all previous OpenLDAP releases. You only need to worry about these steps if there are no CSNs at all, or they came from some non-OpenLDAP software.

No idea what that is. Your debug logs should tell what it was doing.

I've tried a lot of loglevels and look for anything suspicious.
I noticed (as I mentioned) this:
bdb_index_read: failed (-30989)

... and another thing I find weird.
The last entry in the LDIF is special in the log, like:

Dec  4 14:14:30 server1 slapd[5433]:<=
Dec  4 14:14:30 server1 slapd[5433]: Entry
cn=me,ou=3,uid=apm,o=net,cn=data,dc=example,dc=com changed by peer, ignored
Dec  4 14:14:30 server1 slapd[5433]: send_ldap_result: conn=1004 op=1 p=3
Dec  4 14:14:30 server1 slapd[5433]: send_ldap_result: err=0 matched=""
Dec  4 14:14:30 server1 slapd[5433]: syncprov_search_response:

Sounds like a change was written on server2 and received on server1. The above message just means that server1 is not going to try to send the same change back to server2. (I.e., perfectly normal.) So the question is, why are you saying that no writes have occurred on server2, when the logs and contextCSNs show otherwise?

  -- Howard Chu
  CTO, Symas Corp.           http://www.symas.com
  Director, Highland Sun     http://highlandsun.com/hyc/
  Chief Architect, OpenLDAP  http://www.openldap.org/project/