Thread: PG_TERMINATE_BACKEND not working.
I am facing problem i.e. connections after execution completed are residing in pg_stat_activity and pg_stat_database. but when i am trying to kill them manually using pg_terminate_backend (All IDLE connections are getting killed but, others like declare, select etc.) are not getting killed getting result false. Kindly help to resolve this issue. -- View this message in context: http://postgresql.1045698.n5.nabble.com/PG-TERMINATE-BACKEND-not-working-tp5732379.html Sent from the PostgreSQL - general mailing list archive at Nabble.com.
On 11/16/2012 02:34 PM, Harry wrote: > I am facing problem i.e. connections after execution completed are residing > in pg_stat_activity and pg_stat_database. > but when i am trying to kill them manually using pg_terminate_backend (All > IDLE connections are getting killed but, others like declare, select etc.) > are not getting killed getting result false. > Kindly help to resolve this issue. More detail is needed. Pg version? OS and version? Exact command output? What the queries that're not terminated are doing? Why you're terminating them / what problem you're trying to solve? -- Craig Ringer http://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Training & Services
On Sun, Nov 18, 2012 at 10:25 PM, Craig Ringer <craig@2ndquadrant.com> wrote: > On 11/16/2012 02:34 PM, Harry wrote: >> I am facing problem i.e. connections after execution completed are residing >> in pg_stat_activity and pg_stat_database. >> but when i am trying to kill them manually using pg_terminate_backend (All >> IDLE connections are getting killed but, others like declare, select etc.) >> are not getting killed getting result false. >> Kindly help to resolve this issue. > > More detail is needed. > > Pg version? > > OS and version? > > Exact command output? > > What the queries that're not terminated are doing? > > Why you're terminating them / what problem you're trying to solve? And: Are you logged in as superuser when you try to kill them, and if not, are the other sessions from your own user? ChrisA
Chris Yes, i am logging in as Super User also, other side same user's connection(Superuser) i am trying to kill. -- View this message in context: http://postgresql.1045698.n5.nabble.com/PG-TERMINATE-BACKEND-not-working-tp5732379p5732694.html Sent from the PostgreSQL - general mailing list archive at Nabble.com.
n Mon, Nov 19, 2012 at 10:52 AM, Harry <shirlekar.harshal@gmail.com> wrote:
Chris Yes, i am logging in as Super User also, other side same user's
connection(Superuser) i am trying to kill.
Can you give a try, by picking the IP's of the processes which are not getting killed from pg_stat_activity(client_addr column) and make REJECT entries in pg_hba.conf file as below:
host all all <client_addr IP> reject
and reload the cluster using pg_ctl -D $PGDATA reload
Now try killing the process
pg_ctl kill TERM <Process PID>
Regards,
Raghavendra
EnterpriseDB Corporation
Hi Harry,
As you reported "pg_terminate_backend(pid)" returning "f" when executing. Please confirm are u getting any error like below .
postgres=# select pg_terminate_backend(342033);
WARNING: PID 342033 is not a PostgreSQL server process
pg_terminate_backend
----------------------
f
(1 row)
Not sure, what kind of error your are getting while executing it. So, requesting you to share your postgresql version, command what you have executed and it's result. However, pg_terminate_backend() is responsible to send "SIGTERM" to it's OS process. From there onwards, OS will take care about the given process termination.
Can you give a try, by picking the IP's of the processes which are not getting killed from pg_stat_activity(client_addr column) and make REJECT entries in pg_hba.conf file as below:host all all <client_addr IP> rejectand reload the cluster using pg_ctl -D $PGDATA reloadNow try killing the processpg_ctl kill TERM <Process PID>
Regards,
Then same problem with you,and any idea? -- View this message in context: http://postgresql.1045698.n5.nabble.com/PG-TERMINATE-BACKEND-not-working-tp5732379p5812909.html Sent from the PostgreSQL - general mailing list archive at Nabble.com.