Re: pgsql: Fix a crash in logical replication

2019-01-30 Thread Peter Eisentraut
On 30/01/2019 10:54, Alvaro Herrera wrote: > On 2019-Jan-30, Peter Eisentraut wrote: > >> Fix a crash in logical replication >> >> The bug was that determining which columns are part of the replica >> identity index using RelationGetIndexAttrBitmap() would run >> eval_const_expressions() on index

pgsql: Fix a crash in logical replication

2019-01-30 Thread Peter Eisentraut
Fix a crash in logical replication The bug was that determining which columns are part of the replica identity index using RelationGetIndexAttrBitmap() would run eval_const_expressions() on index expressions and predicates across all indexes of the table, which in turn might require a snapshot, bu

pgsql: Fix a crash in logical replication

2019-01-30 Thread Peter Eisentraut
Fix a crash in logical replication The bug was that determining which columns are part of the replica identity index using RelationGetIndexAttrBitmap() would run eval_const_expressions() on index expressions and predicates across all indexes of the table, which in turn might require a snapshot, bu

pgsql: Fix a crash in logical replication

2019-01-30 Thread Peter Eisentraut
Fix a crash in logical replication The bug was that determining which columns are part of the replica identity index using RelationGetIndexAttrBitmap() would run eval_const_expressions() on index expressions and predicates across all indexes of the table, which in turn might require a snapshot, bu

pgsql: Fix a crash in logical replication

2019-01-30 Thread Peter Eisentraut
Fix a crash in logical replication The bug was that determining which columns are part of the replica identity index using RelationGetIndexAttrBitmap() would run eval_const_expressions() on index expressions and predicates across all indexes of the table, which in turn might require a snapshot, bu

pgsql: Fix a crash in logical replication

2019-01-30 Thread Peter Eisentraut
Fix a crash in logical replication The bug was that determining which columns are part of the replica identity index using RelationGetIndexAttrBitmap() would run eval_const_expressions() on index expressions and predicates across all indexes of the table, which in turn might require a snapshot, bu

Re: pgsql: Fix a crash in logical replication

2019-01-30 Thread Alvaro Herrera
On 2019-Jan-30, Peter Eisentraut wrote: > Fix a crash in logical replication > > The bug was that determining which columns are part of the replica > identity index using RelationGetIndexAttrBitmap() would run > eval_const_expressions() on index expressions and predicates across > all indexes of

pgsql: Fix a crash in logical replication

2019-01-30 Thread Peter Eisentraut
Fix a crash in logical replication The bug was that determining which columns are part of the replica identity index using RelationGetIndexAttrBitmap() would run eval_const_expressions() on index expressions and predicates across all indexes of the table, which in turn might require a snapshot, bu