[Date Prev][Date Next]
Re: Ongoing issues with replication.
Anwering (some) of my own post:
Ron Chmara wrote:
> I discover my directories are no longer replicating, and the
> load on the machine seems high. Very high.
> 7:00pm up 35 days, 5:04, 1 user, load average: 2.39, 1.49, 1.19
> 44 processes: 38 sleeping, 5 running, 1 zombie, 0 stopped
> CPU states: 48.3% user, 19.2% system, 0.0% nice, 32.4% idle
> Mem: 517032K av, 515692K used, 1340K free, 3504K shrd, 10484K buff
> Swap: 1052216K av, 526212K used, 526004K free 10876K cached
> 3730 root 0 0 206M 206M 600 S 0 0.0 40.8 0:00 slurpd
> 3731 root 0 0 206M 206M 600 S 0 0.0 40.8 0:00 slurpd
> 3732 root 1 0 206M 206M 600 R 0 93.5 40.8 0:21 slurpd
This is what happens when cd /usr/tmp/replica/slurpd.replog hits a few
gigs. It tries to load the whole thing into RAM, apparently. I had to
throw it away.
> Config: (replica host=ns2.cmg.net:389 binddn="cn=manager,o=Alphagraphics" bindmethod=simple credentials=secret)
> Config: ** successfully added replica "ns2.cmg.net:389"
> Config: (replogfile /var/lib/ldap/slapd.replog)
> Config: ** configuration file successfully read and parsed
> Warning: unknown replica devworld.alphagraphicsinc.com:389 found in replication log
> Which seems odd, as well, as my replogfile is 0 bytes.... (where is it finding this?)
'man slurpd', note the "-f" option. The directive for one's replogfile is not
where they are being stored, but merely where they are placed before being moved to
the work area.
The replog itself had two entries, repeated a few million times, for
replication. I don't know why. (A bug in moving data over to the tmp area?)
Since I noted some errors on the bug pages, I also changed my replica
directives to use IP numbers, rather than dns, which may have helped.
Personal: firstname.lastname@example.org, 520-326-6109, http://www.opus1.com/ron/
Work: email@example.com, 520-546-8993, http://www.pnsinc.com/
The opinions expressed in this email are not necessarily those of myself,
my employers, or any of the other little voices in my head.