Files
gholt
52ba08d67d
Improvements to database replication.
Note: I'd like to get this released as soon as possible as it is a data durability issue. 1) Orders nodes so that none get starved (see code and footnote). 2) New max_diffs setting that caps how long the replicator will spend trying to sync a given database per pass so the other databases don't get starved. 3) Replaces run_pause with the more standard "interval", which means the replicator won't pause unless it takes less than the interval set. Change-Id: I986742229e65031df88f5251ca61746b7c8d2bde
Updated to use standard shebang and explicitly flush output; added deprecation warning to stats.conf