[Bucardo-general] Recover bucardo "hole" in sync

Laurent GARCES laurent.garces at l-rd.fr
Mon Jan 16 11:34:37 UTC 2017


Hi,

Thank you very much for the "id=id trick", that did it.

As a side question, could you advice a way to monitor the bucardo job on 
postgres?

During the tests phase we did some things like, on each synced database:

pg_dump <database> --exclude-schema=bucardo -t <syncedTable1> -t 
<syncTable2> | sort | md5sum

Not really smart but enough for monitoring. But now our databases are 
too active to use it.

So we're looking for some similar way, but restricted to a period. For 
example "get checksum of all the entries that have been added/updated 
yesterday"...

We would like to do in a "generic way" without having to use some 
eventual "date" field in the tables.

I know it should be not directly related to bucardo, but may be some of 
you may have ideas/answers on this.

Regards,

-- 
Laurent GARCES
LRD www.l-rd.fr
+33 (0)4 67 92 52 56

Le 08/01/2017 à 14:47, Greg Sabino Mullane a écrit :
> On Sat, Jan 07, 2017 at 04:47:21PM -0800, Rosser Schwarz wrote:
>> If I understand your problem correctly, the fix is pretty simple. It sounds
>> like the data the third host is missing should exist on (at least one of)
>> the other two, is this correct?
>>
>> If so, and you can determine which rows are missing on the third host
> Rosser is correct, but if you have no way of telling which rows have changed
> since the creation of your backup image, you may need to simply "touch"
> *all* the rows in stages. At that point, it may be better to enter the
> delta rows directly rather than letting the bucardo triggers do it by
> touching the rows. A bit complicated, but you can get help on #bucardo
> at irc.freenode.net or here.
>



More information about the Bucardo-general mailing list