[Pgcluster-general] problem/bug in all pgcluster versions?

[Markus Schmidleitner]

hello,

i'm testing pgcluster-1.9rc5.
i have following 2 node cluster setup ...

node1:          node2:
pgsql           pgsql
pgreplicate     pgreplicate

... and everything works fine so far.

i have a database with multiple schemas (namespaces).
i make a backup of this database using pg_dump on node1 and then i drop this database.
all ok so far.
then i restore the database using pg_restore on node1 with no error regarding to the restore.
but when i take a look at the database on node2 the public schema (namespace) doesn't contain any tables although there are tables on node1!!

i searched for the possible cause of this problem and found out that every "SET search_path = public"-query will be skipped from replication.
this is in all versions of pgcluster i have checked (started with 1.7)!
i commented out the expression "(strncmp(query,SKIP_QUERY_4,strlen(SKIP_QUERY_4)) == 0)" in src/backend/libpq/replicate.c and now everything works as expected!

is there any reason why not replicating such a command?
btw: i would also consider using strncasecmp rather than strncmp

regards,
markus schmidleitner

__________________________________________________________________________________
Dieses Mail wurde vom Infotech SecureMail Service ueberprueft und fuer sicher befunden.
Fuer weitere Informationen zu Infotech SecureMail Service waehlen Sie bitte: www.infotech.at

This email has been scanned by Infotech SecureMail Service and it has been classified as secure.
For more information on Infotech SecureMail direct your web browser to: www.infotech.at

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://pgfoundry.org/pipermail/pgcluster-general/attachments/20080328/ba4f61b9/attachment.html