Running postgresql 8.2.0 on Debian testing, Linux kernel 2.6.12, Dual
Xeon 2.80GHz motherboard,
gcc 4.0.3.
If I run a statement like this:
select * from [table] limit null;
it crashes the server.
Under 8.1.3, limit NULL was synonymous with limit ALL. Here, the
server terminates and restarts with a fatal error message:
<%:4170> 2006-12-21 10:38:10 ESTLOG: server process (PID 4495) was
terminated by signal 11
<%:4170> 2006-12-21 10:38:10 ESTLOG: terminating any other active
server processes
<postgres%tmc:4496> 2006-12-21 10:38:10 ESTWARNING: terminating
connection because of crash of another server process
<postgres%tmc:4496> 2006-12-21 10:38:10 ESTDETAIL: The postmaster
has commanded this server process to roll back the current
transaction and exit, because another server process exited
abnormally and possibly corrupted shared memory.
<postgres%tmc:4496> 2006-12-21 10:38:10 ESTHINT: In a moment you
should be able to reconnect to the database and repeat your command.
<postgres%tmc:4497> 2006-12-21 10:38:10 ESTWARNING: terminating
connection because of crash of another server process
<postgres%tmc:4497> 2006-12-21 10:38:10 ESTDETAIL: The postmaster
has commanded this server process to roll back the current
transaction and exit, because another server process exited
abnormally and possibly corrupted shared memory.
<postgres%tmc:4497> 2006-12-21 10:38:10 ESTHINT: In a moment you
should be able to reconnect to the database and repeat your command.
<norman%clndb:4499> 2006-12-21 10:38:10 ESTFATAL: the database
system is in recovery mode
<%:4170> 2006-12-21 10:38:10 ESTLOG: all server processes
terminated; reinitializing
---------------------------(end of broadcast)---------------------------
TIP 2: Don't 'kill -9' the postmaster
