Wed Dec 20 07:13:51 PST 2006
- Previous message: [Slony1-general] how to drop node when replication is lagging
- Next message: [Slony1-general] how to troubleshoot/fix slow initial replication --
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
(I asked a related question in a previous post -- ) I have a medium-sized table (about 5G, with 4G worth of indexes) in a database that's about 37G is size. On November 30, I created a slony cluster using Slony 1.2.1with one master and two slave nodes, all running postgresql 8.1.3 on Debian testing on a 1G local area network. The initial replication of the full database took about 3 hours, with the medium sized table taking abot 25 minutes to replicate. Yesterday, I dropped one of the slave nodes from the cluster and built postgresql 8.2.0 on it. I then killed all the slon processes, upgraded master and both slave boxes to slony 1.2.2, brought back up the slony cluster and readded the second slave back to the node. Initial replication on the medium sized table on this slave node took about 16 hours(!), and the replication of the entire database is still running some 20 hours after I started. I've done a low level trace of the slony process and I can see that it is indeed still moving packets from the master to the slave, so it's working -- but what can I use to troubleshoot why on earth this replication is so slooooooow? 1) Are there any known issues replicating between 8.1.3 and 8.2.0? 2) Are there any known issues with initial replications in 8.2.0? 3) Could I have a misconfiguration in my 8.2.0 postgresql.conf file? I'm not sure what to look for; all I know is that if the initial replication time shoots from 2 hours to over 20 hours, something's going wrong somewhere. Any help would be greatly appreciated. ###################################################################### This e-mail is confidential and should not be redistributed or forwarded by the recipient except with the permission of the sender. The information contained in this e-mail message is intended only for the use of the individual or entity named above. If the reader of this message is not the intended recipient or you have received this communication in error, please immediately notify us by telephone. Receipt by anyone other than the intended recipient is not a waiver of any work-product or, if applicable, attorney-client privilege. This e-mail does not constitute an offering of any security. Such an offering may only be made by means of a private placement memorandum or other disclosure document. Nothing in this e-mail constitutes investment advice. Past performance is not indicative of future results. All e-mail to and from Millburn Ridgefield Corporation and its affiliates is monitored, stored and made available to regulators if requested. ######################################################################
- Previous message: [Slony1-general] how to drop node when replication is lagging
- Next message: [Slony1-general] how to troubleshoot/fix slow initial replication --
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Slony1-general mailing list