Обсуждение: psql: FATAL: the database system is starting up

Hi,

We had to kill postmaster, and it's been ugly.   The last backup finished
successfully not last night, but the night before.  This is 161 GB database.
   How long should I expect this to take before it starts.   Developers are
coming into my office every 5 minutes asking for updates.   It's never taken
this long to come back.  We did have an awful lot of transactions last
night.    Is the server replaying everything

fhttp://by116fd.bay116.hotmail.msn.com/cgi-bin/compose?&curmbox=00000000%2d0000%2d0000%2d0000%2d000000000001&a=3b0ef010ec78a7ca825dcdc7953aaf4c22ff15917d66f2ef8a3ddf1fd5dbd659#
Sendrom the last succesful backup?   Is it hanging because it can't find
some WAL files that were archived?  It's been an hour and a half so far.
Would a reboot help?

~DjK





ep 21 10:07:05 p postgres[28301]: [2-3]  recovery.
Sep 21 10:07:05 p postgres[28301]: [3-1] < % %2006-09-21 10:07:05 EDT>LOG:
checkpoint record is at 14D/69FFBD8
Sep 21 10:07:05 p postgres[28301]: [4-1] < % %2006-09-21 10:07:05 EDT>LOG:
redo record is at 14D/696AC20; undo record is at 0/0; shutdown FALSE
Sep 21 10:07:05 p postgres[28301]: [5-1] < % %2006-09-21 10:07:05 EDT>LOG:
next transaction ID: 39405490; next OID: 1799955
Sep 21 10:07:05 p postgres[28301]: [6-1] < % %2006-09-21 10:07:05 EDT>LOG:
next MultiXactId: 5696; next MultiXactOffset: 11694
Sep 21 10:07:05 p postgres[28301]: [7-1] < % %2006-09-21 10:07:05 EDT>LOG:
database system was not properly shut down; automatic recovery in progress
Sep 21 10:07:05 p postgres[28301]: [8-1] < % %2006-09-21 10:07:05 EDT>LOG:
redo starts at 14D/696AC20
Sep 21 10:07:20 p postgres[28305]: [2-1] <[unknown] %[unknown] %2006-09-21
10:07:20 EDT>LOG:  invalid length of startup packet

"Mr. Dan" <bitsandbytes88@hotmail.com> writes:
> We had to kill postmaster, and it's been ugly.   The last backup finished
> successfully not last night, but the night before.  This is 161 GB database.
>    How long should I expect this to take before it starts.

Not longer than the time since the last checkpoint.  What does
pg_controldata show?  Is the recovery process actually doing any work?

            regards, tom lane

 Thanks everyone, it came back up a few minutes after I sent the SOS.

-----Original Message-----
From: pgsql-admin-owner@postgresql.org
[mailto:pgsql-admin-owner@postgresql.org] On Behalf Of Tom Lane
Sent: Thursday, September 21, 2006 11:58 AM
To: Mr. Dan
Cc: pgsql-admin@postgresql.org
Subject: Re: [ADMIN] psql: FATAL: the database system is starting up

"Mr. Dan" <bitsandbytes88@hotmail.com> writes:
> We had to kill postmaster, and it's been ugly.   The last backup finished
> successfully not last night, but the night before.  This is 161 GB
database.
>    How long should I expect this to take before it starts.

Not longer than the time since the last checkpoint.  What does
pg_controldata show?  Is the recovery process actually doing any work?

            regards, tom lane

---------------------------(end of broadcast)---------------------------
TIP 3: Have you checked our extensive FAQ?

               http://www.postgresql.org/docs/faq