http://www.slony.info/bugzilla/show_bug.cgi?id=196

           Summary: snapshot id values appear to travel backwards in
                    sl_event
           Product: Slony-I
           Version: devel
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: low
         Component: slon
        AssignedTo: slony1-bugs at lists.slony.info
        ReportedBy: ssinger at ca.afilias.info
                CC: slony1-bugs at lists.slony.info
   Estimated Hours: 0.0


^
test1=# select ev_origin,ev_seqno,ev_type,ev_timestamp,ev_snapshot FROM
_disorder_replica.sl_event where ev_seqno>=5000000050  AND
ev_seqno<=5000000054;
 ev_origin |  ev_seqno  | ev_type |        ev_timestamp        |   ev_snapshot  
-----------+------------+---------+----------------------------+------------------
         1 | 5000000050 | SYNC    | 2011-02-15 17:50:27.10242  |
3719843:3719843:
         1 | 5000000051 | SYNC    | 2011-02-15 17:50:27.180308 |
3719845:3719845:
         1 | 5000000052 | SYNC    | 2011-02-15 17:50:28.505496 |
3719843:3719843:
         1 | 5000000053 | SYNC    | 2011-02-15 17:50:33.125079 |
3719917:3719917:
         1 | 5000000054 | SYNC    | 2011-02-15 17:50:35.159158 |
3719921:3719921:
(5 rows)


1,5000000050  and 1,5000000052 have the same event snapshot.  1,5000000051 is a
later snapshot than 1,5000000052. 

This resulted in a slon trying to insert the same row into a table twice on the
remote node.


This was seen in 2.0.6+ with the SlonKilling.js test from the disorder test
suite.

The problem this bug is concerned with is that the exclusive lock on sl_event
that we take in slony_funcs.c comes too late.  THe snapshot was already
allocated.  We need to get that exclusive lock before the snapshot is
allocated. 

The solution proposed is to modify slonik to get the exclsuive lock before
calling any stored functions (since calling the stored functions creates the
snapshot).

-- 
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.