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

Replication and related problems ...

Another oddity I have noticed with the development server is that it seems to crash after about a day in operation. This is noticeable because we have added it to our nightly modifications we make to our production LDAP, and it seems to crash during the heaviest part of the changes (when we perform adds/modifications/deletes to our student directory at 4am). Unfortunately, I have not yet been able to reproduce this, and we no longer have the core file (I will be sure to save it next time).

The other item of interest is replication. I successfully setup a master and slave beta, and had replication functioning properly. However, after the first 10 minutes or so, we discovered that replication was no longer occurring. I gave it 27 hours, and still nothing was being replicated. Then, after the master crashed at 4am, we killed the old slurpd daemon restarted both servers. Slurpd hung when attempting to start (I could see a second process running with the slurpd name at that point), and never came back. After about 40 minutes I killed the process, and checked the log files. The time/date stamps had NOT changed since yesterday on any of the files.

To provide debugging information, I ran slurpd with:

/opt/ldap/libexec/slurpd -f /opt/ldap/etc/openldap/slapd.conf -d 1 -o -r /opt/ldap/var/slapd.replog > debug.txt 2>&1

The replication was performed flawlessly, and all updates committed. The fact that any time I turn on debugging, problems vanish is VERY frustrating :/ Any ideas are appreciated.


P.S. Should I be downloading the CVS versions for this project, or continuing to use the original Beta release version?
* Anthony Brock abrock@georgefox.edu *
* Director of Network Services George Fox University *