http://www.slony.info/bugzilla/show_bug.cgi?id=313

Christopher Browne <cbbrowne at ca.afilias.info> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|slony1-bugs at lists.slony.inf |ssinger at ca.afilias.info
                   |o                           |
             Status|NEW                         |ASSIGNED

--- Comment #1 from Christopher Browne <cbbrowne at ca.afilias.info> 2013-09-04 12:27:10 PDT ---
Points about this...

1.  It seems pretty reasonable to not bother griping too loudly about sequences
that are not being replicated onto a particular node.

It could be an error that they *aren't* being replicated, but that's "driver
error," rather than a bug in Slony itself.

2.  It would be nice if we could report some kind of DEBUG level message on
this, however, since this is being processed inside pl/pgsql, that may not be
particularly easy to do.

That makes me wonder if we shouldn't be exposing the list of sequences at the
slon process layer, but that's "wonder", not "wow, that sure is broken."

3.  When shall we fix this?

i.  Immediately, if not before...   Assuming that turning the ERROR into a
NO-OP is OK.

ii.  Defer 2.2.0 until it is fixed, that is, create another 2.2.0 beta
candidate. 

iii.  Fix in 2.2.1, which leaves exposure to this issue for a little while.

4.  How to repair it...

i.  Simply turn the ERROR in FUNCTION sequenceSetValue() into a NOOP

ii.  Add the set ID into the array structure that gets passed into
sl_log_script, and check against that.

Thus, sequenceSetValue() would:
  a) NOOP if the set isn't being subscribed to on the current node, or
  b) ERROR if it is, because clearly something is mussed up.

Note that changing to ii is somewhat messy, as it changes the structure of the
data being passed around.  Do we make sure that slon can cope with both old and
new form?

Recommending that we go with "just a NOOP", and do not delay 2.2.0.

-- 
Configure bugmail: http://www.slony.info/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
You are the assignee for the bug.