Re: Synchronisation problem with replication

[Chris Hostetter <ho...@fucit.org>]

:   Let's say I post a document on the master server, and the slaves do
: a snappuller/installer via crontab every 1 minutes.
: 
:   Then between in average 30 seconds, all my search servers are not
: synchronized.
: 
:   Is there a way to improve this situation ?

If your slaves all use NTP to keep their clocks in sync then the snap 
pulls should be fairly well in sync.  

you're probably going to see more variation in how long it takes to warm 
newSearchers based on differences in cached queries on each slave then you 
are to notice network lag differences between slaves on neighboring racks.

if your slaves are very far about, and you have unpredictable network 
times, you could always use the stats of which slaves have which snapshots 
to know when to send a snapinstall command to all the slaves.



-Hoss