Fri May 23 01:27:07 PDT 2008
- Previous message: [Slony1-general] Slow replication issue
- Next message: [Slony1-general] Slow replication issue
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
On Fri, 2008-05-23 at 14:37 +0900, Stephane LAPIE wrote: > Hello, > > We have been using Slony-I for nearly two years now for our database > system, with the following cluster : > - 6 Postgres 8.2.4 servers, four in Asia, two in Europe > - Communication occurs over a VPN, and every server can reach the others > - Roundtrip average is 300ms for Europe and Asia servers > - Data providers are in Asia, 3 million row updates / day average > (mostly split in 4 big updates / day), and at peak time, 100 updates/s > > Replication basically occurs fine, to the exception that the European > servers accumulate lag for around two, three hours (six in worst cases > observed) when data processing occurs in Asia. We tried raising > "sync_group_maxsize" in every slon.conf to 24, which diminished the > number of events replication was lagging behind, but did not diminish > the lag time (Both are observed through the _SCHEMA.sl_status view, and > made into a graph through MRTG) My experience is that for high-latency links it is crucial to have the slon daemon doing the inserts/updates/deletes running on the receiving side (close to the slave tables) so that the actual data updating queries do not happen over WAN links. You may need to split your replication to 2 separate slony clusters executed by separate slony daemons to achieve this. > By reducing processed data quantities of 30%, we could reduce the lag > time by half. We also checked the VPN status, and there is no obvious > anomaly nor QoS. The only startling point would be that bandwidth usage > is extremely low, and that reducing processed data quantities did not > make the bandwidth usage fall any lower. > > It looks as though as Slony is actually limiting itself to not over-use > the network's bandwidth, but actually ends up using nearly nothing. > Given our current network environment, we could perfectly cope with > Slony using 10 times more bandwidth with no problem. > > Is there a way to force Slony out of it's "don't go too hard on the > network" mode (such as a setting in slon.conf)? > > Thanks in advance,
- Previous message: [Slony1-general] Slow replication issue
- Next message: [Slony1-general] Slow replication issue
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Slony1-general mailing list