On 7/14/2011 3:50 PM, Christopher Browne wrote:
> On Thu, Jul 14, 2011 at 2:54 PM, Frank McGeough
> <Frank.Mcgeough at vocalocity.com>  wrote:
>>  I have a 3 server cluster with slow replication currently. I believe it is because a very large transaction (>  1,000,000 rows) yielded a low of dead tuples in the slony log tables. I see a lot index bloat in the slony tables - sl_seqlog has a index_ratio of 68.65, for example. sl_confirm is 78.
>>
>>  Could I run
>>
>>  cluster sl_seqlog using sl_seqlog;
>>
>>  and help this situation?
>
> Yep, that should be helpful.
>
> In version 2.0, we changed strategy for sl_seqlog, so that it only
> collects values when sequences actually change, so it should be
> expected to churn a whole lot less.

We also identified a problem with large backlog in the selection from 
sl_log_n. This has been fixed in the upcoming 2.1 but not yet been 
backported into 2.0.


Jan

-- 
Anyone who trades liberty for security deserves neither
liberty nor security. -- Benjamin Franklin