Hello all,
We like to use bucardo to enhance the databases scalability : We
currently have 2 machines M1, M2 in multi master mode (defined both as
'source' databases), and we plan to add other nodes S3, S4 ... that
would be never writable (they would be defined as 'target' databases).
Some tests show that is that if one of the S3, S4 machine goes down,
this blocks the replication on all machines.
Is it a way that a sync continues to be active even if one referenced
machine is down? The same problem is related in some old posts, so I
fear it is a well known problem.
Note: We were thinking of creating syncs per couple of machines, but
this tends to complexify our architecture, and we fear the scalability
won't be linear with machine addition, because creating lots of syncs
means more work for bucardo (lots of triggers) and lots of postgres
connections. Is there something 'simpler' we could do with bucardo to be
able to support "down" nodes ?
Additional question:
Is it possible to use a dedicated pgbouncer with bucardo that will only
be used by bucardo. This may reduce the number of postgres instances.
Thanks and regards,
Sylvain
_______________________________________________
Bucardo-general mailing list
[email protected]
https://mail.endcrypt.com/mailman/listinfo/bucardo-general
