We ran replication at ten minute intervals. One master, five slaves, and replication on the hour on the first slave, ten minutes after the hour on the second, twenty minutes after on the third, and so on.
You could do this with a single crontab on the master. Send requests to each slave to replicate. We had a small index (about 250K docs) that was updated once per day. The replication ran every hour, just in case we had to make a mid-day change, which did happen sometimes. wunder On Jul 15, 2013, at 9:20 AM, adityab wrote: > Walter, > Could you provide some more details about your staggered replication > approach? > We are currently running into similar issues and looks like staggered > replication is a better approach to address the performance issues on > Slaves. > > thanks > Aditya > > > > -- > View this message in context: > http://lucene.472066.n3.nabble.com/Replication-process-on-Master-Slave-slowing-down-slave-read-search-performance-tp707934p4078090.html > Sent from the Solr - User mailing list archive at Nabble.com.
