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

Re: OL 2.3.18 syncrepl vs slurpd

On 1/19/06 5:01 PM, Quanah Gibson-Mount wrote:

--On Thursday, January 19, 2006 4:01 PM -0500 Aaron Richton <richton@nbcs.rutgers.edu> wrote:

   Anyone have any insight to what I've done wrong or where I should
start looking?

Have you considered delta syncrepl? Look in the tests directory.

I note that Symas has a tech tip available on setting up delta-syncrepl:


There is also a tech tip there discussing the pros and cons of the different replication mechanisms.

Thank you, I followed that tech tip and after fixing the index statements to not include spaces (will you take this as an issue report to fix the tech tip or should I file an ITS somewhere?) I have it working.

Now, I have discovered three things.

1) delta-syncrepl doesn't seem to have any way to filter the amount of what is sent -- so, it has the same issues that I'm fighting with slurpd of sending every update to all the replicas and perhaps I do not want all the updates on all the replicas (this was the reason for me going to syncrepl).

2) There is a DOS against the master server if the consumer codes a bad logfilter. You will see a bad filter indication in the log on the master (with loglevel stats) when the consumer starts up. The first update to the master after that will cause slapd to end with nothing going into the syslog at all, but suddenly, it's not running anymore. Given that anyone could fire up a syncrepl consumer and point it at my master... that's a rather nasty one... Has anyone else noticed it (I honestly just found it and have not searched in the ITS yet)?

3) loglevel sync on the syncrepl consumer doesn't log anything with delta-syncrepl. Did I miss something in the slapd.conf(5) man page about the loglevel to get delta-syncrepl actions logged? Or is it in a different man page, that I didn't think to look at?