Re: Issue in PG start
On 5/8/21 7:32 PM, sivapostg...@yahoo.com wrote: Yesterday's log file. Till we shut down the windows the message was: 2021-05-08 19:03:32.151 IST [10376] FATAL: the database system is starting up 2021-05-08 19:03:42.210 IST [7568] FATAL: the database system is starting up Comments in line below. Today morning log file, I tried to start the service [ after setting the startup type to Manual ] twice *** 2021-05-09 07:43:02.122 IST [16776] LOG: test message did not get through on socket for statistics collector Something is blocking the socket. Do you have anti-virus software on this machine? 2021-05-09 07:43:02.122 IST [16776] LOG: trying another address for the statistics collector 2021-05-09 07:43:02.624 IST [16776] LOG: test message did not get through on socket for statistics collector 2021-05-09 07:43:02.624 IST [16776] LOG: disabling statistics collector for lack of working socket 2021-05-09 07:43:02.624 IST [16776] WARNING: autovacuum not started because of misconfiguration 2021-05-09 07:43:02.624 IST [16776] HINT: Enable the "track_counts" option. You should take care of this by following the HINT. 2021-05-09 07:43:02.685 IST [17144] LOG: database system was interrupted; last known up at 2021-05-07 18:46:00 IST 2021-05-09 07:43:04.489 IST [17196] FATAL: the database system is starting up ... 2021-05-09 07:45:14.796 IST [17144] LOG: database system was not properly shut down; automatic recovery in progress 2021-05-09 07:45:14.864 IST [17144] LOG: redo starts at 2/88985C80 2021-05-09 07:45:14.866 IST [17144] LOG: invalid record length at 2/889872B8: wanted 24, got 0 2021-05-09 07:45:14.866 IST [17144] LOG: redo done at 2/88987280 2021-05-09 07:45:14.866 IST [17144] LOG: last completed transaction was at log time 2021-05-07 18:48:44.227574+05:30 2021-05-09 07:45:15.046 IST [16776] LOG: database system is ready to accept connections 2021-05-09 07:50:17.295 IST [17204] ERROR: canceling statement due to user request So did you or someone else do the above and the below manually or was this coming from the system? 2021-05-09 07:50:17.300 IST [16776] LOG: received fast shutdown request 2021-05-09 07:50:17.322 IST [4408] LOG: could not receive data from client: An existing connection was forcibly closed by the remote host. Are you connecting to database from outside the machine and if so is that from a local network or a remote one? 2021-05-09 07:50:17.412 IST [16776] LOG: aborting any active transactions 2021-05-09 07:50:17.419 IST [16776] LOG: background worker "logical replication launcher" (PID 17204) exited with exit code 1 2021-05-09 07:50:17.548 IST [11032] LOG: shutting down 2021-05-09 07:50:17.607 IST [16776] LOG: database system is shut down ** See more below. Log file after a re-start *** 2021-05-09 07:52:40.341 IST [13220] LOG: test message did not get through on socket for statistics collector 2021-05-09 07:52:40.342 IST [13220] LOG: trying another address for the statistics collector 2021-05-09 07:52:40.844 IST [13220] LOG: test message did not get through on socket for statistics collector 2021-05-09 07:52:40.844 IST [13220] LOG: disabling statistics collector for lack of working socket 2021-05-09 07:52:40.845 IST [13220] WARNING: autovacuum not started because of misconfiguration 2021-05-09 07:52:40.845 IST [13220] HINT: Enable the "track_counts" option. 2021-05-09 07:52:40.904 IST [13860] LOG: database system was shut down at 2021-05-09 07:50:17 IST 2021-05-09 07:52:41.122 IST [13220] LOG: database system is ready to accept connections 2021-05-09 07:55:23.628 IST [11188] LOG: using stale statistics instead of current ones because stats collector is not responding * Not sure what happened, it's working now. Any idea of what happened and what should we do to prevent re-occurance ? It's up but not really working. Both the statistics collector and the autovacuum processes are down. They are both important for the operation of the database. Happiness Always BKR Sivaprakash -- Adrian Klaver adrian.kla...@aklaver.com
Setting wal_keep_segments parameter
Hello , I want to set wal_keep_segments parameter to a value but what should be the process I follow? I have archiving process on all clusters is on but when a sudden process comes to postgres wal files will be bigger than i assume. I don't know what I should do. Could you please help me? Thanks.
Re: Setting wal_keep_segments parameter
rob stan writes: > I want to set wal_keep_segments parameter to a value but what should be the > process I follow? I have archiving process on all clusters is on but when a > sudden process comes to postgres wal files will be bigger than i assume. I > don't know what I should do. Could you please help me? Well, the thing about wal_keep_segments is that there is no guaranteed safe value. If you are trying to use it to protect streaming standby servers then the better modern solution is to have the standbys use replication slots, so the primary actually knows how much old WAL they need. On the other hand, if you're just concerned about archiving then I don't think you need it at all; the archiving mechanism doesn't need any help to ensure that WAL doesn't get recycled before it's archived. Or in short: I think the correct setting for wal_keep_segments is zero. There are generally better ways to do what it does. regards, tom lane
Re: Postgres upgrade 12 - issues with OIDs
On Sat, 8 May 2021 at 1:47 pm, Laurenz Albe wrote: > On Sat, 2021-05-08 at 13:37 +1000, Venkata B Nagothi wrote: > > We are thinking to upgrade to PG 11 instead so that we can avoid doing > ALTER TABLE.. SET WITHOUT OIDs. > > Does that makes sense ? Please advise if there are any gotchas ! > > It makes sense, but it means that you will have to face the same problem > later. > However, for upgrading from v11 with little down time you may be able to > use > logical replication. Yes, we will have a lot of time to deal with the OID problem later, good thing is we will be out of 9.5 with less trouble. Hopefully we will be on a better replication architecture soon which will make it much easier for us. > > Yours, > Laurenz Albe > -- > Cybertec | https://www.cybertec-postgresql.com > > -- Regards, Venkata B N Database Consultant
Re: Issue in PG start
On Sat, 2021-05-08 at 05:14 +, sivapostg...@yahoo.com wrote: > Yesterday [07th May] morning when we switched on the computer and > subsequently PGAdmin, we got the message following message > FATAL: the database system is starting up You will have to wait until startup is completed. Either you didn't shut down PostgreSQL cleanly, or some server process crashed. The log should tell you which of the two happened. Did you set "checkpoint_timeout" to a high value? That will make the startup process potentially take longer to recover. Yours, Laurenz Albe -- Cybertec | https://www.cybertec-postgresql.com
