Christopher Browne cbbrowne
Sat Oct 8 00:19:30 PDT 2005
Marc Munro <marc at bloodnok.com> writes:
> During a switchover test, we got the following error message:
>
> ERROR:  Slony-I: Node is uninitialized
>
> It happened because a table we were using for a heartbeat monitor
> erroneously had an extra slony trigger in place for a different (no
> longer active) slony cluster.  
>
> Totally strange circumstance, and totally my fault, but debug time would
> have been much reduced if the error message had told us which node in
> which cluster was not initialised.
>
> Is there any chance that this information could be added to the error
> message in a future release?

That's generated inside the stored procedures (slony1_func.c); they
aren't directly aware of what node they run on :-(, which may make
this a little challenging.

Could you report this as a feature request on the GBorg site?  That
would allow us to at least not lose track of it entirely...
-- 
let name="cbbrowne" and tld="ca.afilias.info" in String.concat "@" [name;tld];;
<http://dev6.int.libertyrms.com/>
Christopher Browne
(416) 673-4124 (land)


More information about the Slony1-general mailing list