Do you want me to stop the recovery to get the console messages? After
issuing the start command I did not capture the logs because I did not
notice a problem right away. Now I cannot see them anymore.
Is there anything I can do?
The database is 99GB with almost all the data in one table.
How long should the recovery process take?
From: Anjan Dave [mailto:adave(at)vantage(dot)com]
Sent: Friday, June 17, 2005 3:32 PM
To: Andrew Janian; pgsql-admin(at)postgresql(dot)org
Subject: RE: [ADMIN] startup subprocess hangs
Probably some kind of a hardware issue (you lost a path to the array for
e.g.), or someone may have issued a kill -9 to postmaster?
Open another terminal and check the system log (/var/log/messages) if
you are not redirecting logs anywhere.
Can you paste the console messages on the list so people can see what
exactly might be going on?
From: Andrew Janian [mailto:ajanian(at)scottrade(dot)com]
Sent: Friday, June 17, 2005 4:21 PM
Subject: [ADMIN] startup subprocess hangs
I am running Postgres 7.4.5 and I have recently run into a problem. My
database was running fine for days and then today in the middle of the
day it seems that the startup subprocess kicked in and started
recovering. I am not sure from what, and I am not sure how it started,
but nothing was allowed to connect and it has been running for quite
some time (30 minutes). The process said it was starting to recover,
but how long am I to wait?
Is there a way to get the status?
Can I just skip the recovery and get it going?
The environment is RH EL 3 2.4.21-4.ELsmp, with and EMC SAN for storage
and QLogix cards for fibre access.
pgsql-admin by date
|Next:||From: Simon Riggs||Date: 2005-06-17 21:37:19|
|Subject: Re: Indexes getting corrupted.|
|Previous:||From: Anjan Dave||Date: 2005-06-17 20:32:15|
|Subject: Re: startup subprocess hangs|