Обсуждение: Segmentation fault postgres 9.6

Mariel Cherkassky <mariel.cherkassky@gmail.com> writes:
> I'm trying to insert a few values into a foreign table via sqllite_fdw and
> my postgresql database crushes :
> ...
> any idea how can I investigate if the problem in the wrapper or in the
> database ?

Get a stack trace:

https://wiki.postgresql.org/wiki/Generating_a_stack_trace_of_a_PostgreSQL_backend

If the trace shows that the failure occurred in code of the wrapper,
I'd definitely blame the wrapper.  If sqllite_fdw is nowhere to be seen
anywhere in the trace, then it's more debatable.

            regards, tom lane

Hi Mariel,

it seems kernel shared memory parameter issue which blocking or creating a new process, which requires for data purpose. 

Check your kernel shared memory parameters.

On Mon, Sep 17, 2018 at 5:46 PM Mariel Cherkassky <mariel.cherkassky@gmail.com> wrote:

Hi,

I'm trying to insert a few values into a foreign table via sqllite_fdw and my postgresql database crushes : 

2018-09-17 12:09:54 UTC  43444  LOG:  server process (PID 43831) was terminated by signal 11: Segmentation fault

2018-09-17 12:09:54 UTC  43444  DETAIL:  Failed process was running: insert into vip_foreign values (2,2,2);

2018-09-17 12:09:54 UTC  43444  LOG:  terminating any other active server processes

2018-09-17 12:09:54 UTC  43451  WARNING:  terminating connection because of crash of another server process

2018-09-17 12:09:54 UTC  43451  DETAIL:  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.

2018-09-17 12:09:54 UTC  43451  HINT:  In a moment you should be able to reconnect to the database and repeat your command.

2018-09-17 12:09:54 UTC  43444  LOG:  archiver process (PID 43452) exited with exit code 1

2018-09-17 12:09:54 UTC  43444  LOG:  all server processes terminated; reinitializing

2018-09-17 12:09:54 UTC  43840  LOG:  database system was interrupted; last known up at 2018-09-17 12:08:35 UTC

2018-09-17 12:09:56 UTC  43840  LOG:  database system was not properly shut down; automatic recovery in progress

2018-09-17 12:09:56 UTC  43840  LOG:  redo starts at 0/6000098

2018-09-17 12:09:56 UTC  43840  LOG:  invalid record length at 0/60000D0: wanted 24, got 0

2018-09-17 12:09:56 UTC  43840  LOG:  redo done at 0/6000098

2018-09-17 12:09:56 UTC  43840  LOG:  checkpoint starting: end-of-recovery immediate

2018-09-17 12:09:56 UTC  43840  LOG:  checkpoint complete: wrote 0 buffers (0.0%); 0 transaction log file(s) added, 0 removed, 0 recycled; write=0.006 s, sync=0.000 s, total=0.030 s; sync files=0, longest=0.000 s, average=0.000 s; distance=0 kB, estimate=0 kB

2018-09-17 12:09:56 UTC  43840  LOG:  MultiXact member wraparound protections are now enabled

2018-09-17 12:09:56 UTC  43444  LOG:  database system is ready to accept connections

2018-09-17 12:09:56 UTC  43844  LOG:  autovacuum launcher started

any idea how can I investigate if the problem in the wrapper or in the database ?

I do have the same problem with a simple select max() from foreign_table where...

Syslog said the the mysql fdw lib segfaulted.

I've tried to compile the mysql_fdw sources ; but still same problem.



----- Mail original -----
De: "Mariel Cherkassky" <mariel.cherkassky@gmail.com>
À: pgsql-admin@lists.postgresql.org
Envoyé: Lundi 17 Septembre 2018 14:15:44
Objet: Segmentation fault postgres 9.6




Hi,
I'm trying to insert a few values into a foreign table via sqllite_fdw and my postgresql database crushes :

2018-09-17 12:09:54 UTC 43444 LOG: server process (PID 43831) was terminated by signal 11: Segmentation fault
2018-09-17 12:09:54 UTC 43444 DETAIL: Failed process was running: insert into vip_foreign values (2,2,2);
2018-09-17 12:09:54 UTC 43444 LOG: terminating any other active server processes
2018-09-17 12:09:54 UTC 43451 WARNING: terminating connection because of crash of another server process
2018-09-17 12:09:54 UTC 43451 DETAIL: The postmaster has commanded this server process to roll back the current
transactionand exit, because another server process exited abnormally and possibly corrupted shared memory.  
2018-09-17 12:09:54 UTC 43451 HINT: In a moment you should be able to reconnect to the database and repeat your
command. 
2018-09-17 12:09:54 UTC 43444 LOG: archiver process (PID 43452) exited with exit code 1
2018-09-17 12:09:54 UTC 43444 LOG: all server processes terminated; reinitializing
2018-09-17 12:09:54 UTC 43840 LOG: database system was interrupted; last known up at 2018-09-17 12:08:35 UTC
2018-09-17 12:09:56 UTC 43840 LOG: database system was not properly shut down; automatic recovery in progress
2018-09-17 12:09:56 UTC 43840 LOG: redo starts at 0/6000098
2018-09-17 12:09:56 UTC 43840 LOG: invalid record length at 0/60000D0: wanted 24, got 0
2018-09-17 12:09:56 UTC 43840 LOG: redo done at 0/6000098
2018-09-17 12:09:56 UTC 43840 LOG: checkpoint starting: end-of-recovery immediate
2018-09-17 12:09:56 UTC 43840 LOG: checkpoint complete: wrote 0 buffers (0.0%); 0 transaction log file(s) added, 0
removed,0 recycled; write=0.006 s, sync=0.000 s, total=0.030 s; sync files=0, longest=0.000 s, average=0.000 s;
distance=0kB, estimate=0 kB  
2018-09-17 12:09:56 UTC 43840 LOG: MultiXact member wraparound protections are now enabled
2018-09-17 12:09:56 UTC 43444 LOG: database system is ready to accept connections
2018-09-17 12:09:56 UTC 43844 LOG: autovacuum launcher started






any idea how can I investigate if the problem in the wrapper or in the database ?