> From: Vick Khera <vivek at khera.org>
> On Thu, Sep 8, 2011 at 11:43 AM, Glyn Astill <glynastill at yahoo.co.uk> 
> wrote:
>>> Slony 1.2 and 2.0 both behave this way, this shouldn't be new 
>>> behaviour with 2.0.
>>> 
>> 
>>  Hmm, my nagios graphs disagree (or for some reason were oblivious), we 
> check syncs every 10 minutes and they only started showing warnings the night 
> after we moved to 2.0. The check is basically just looks at sl_status, and flags 
> up a problem if st_lag_num_events or st_lag_time go over a threshold via the 
> following query.
>> 
> 
> I disagree as well.  We still run 1.2.x and I never get replication
> lag alarms from our monitoring, and the nightly dump of my big DB
> takes a couple of hours.  I'm still going to add the --exclude-schema
> to it, just because it is useless to dump the slony schema anyway.

Great to hear I'm not the only one then Vick.

When we were on 1.2 I purposefully used to dump the slony schema so that we could restore from our subscribers data and use slony baremetal functions to fix the schema corruption automatically.