"Morris Goldstein" <morris.x.goldstein@gmail.com> writes:
> Two of the nodes have logs that look like this:
> 2007-09-22 07:06:05 [3060] LOG: could not open file
> "/var/lib/pgsql/data/pg_xlog/0000000000000000" (log file 0, segment
> 0): No such file or directory
> 2007-09-22 07:06:05 [3060] LOG: invalid primary checkpoint record
> 2007-09-22 07:06:05 [3060] LOG: could not open file
> "/var/lib/pgsql/data/pg_xlog/0000000000000000" (log file 0, segment
> 0): No such file or directory
> 2007-09-22 07:06:05 [3060] LOG: invalid secondary checkpoint record
> 2007-09-22 07:06:05 [3060] PANIC: could not locate a valid checkpoint record
This is not good. Is it really complaining about log file 0? I can
hardly credit that a database you care about wouldn't have used more log
space than that.
You might be able to recover, at least to the extent of getting some
data out of these DBs, by using pg_resetxlog. Read its man page
*carefully* beforehand.
http://www.postgresql.org/docs/7.4/static/app-pgresetxlog.html
> The other two look like this:
> FATAL: IDENT authentication failed for user "xxx"
This looks like user error, ie, trying to log in as a different Postgres
user than your Unix username is. You can hardly have been accustomed to
doing that if you were normally using IDENT authentication before.
If you want to switch to some other authentication method, read the docs
http://www.postgresql.org/docs/7.4/static/client-authentication.html
and edit pg_hba.conf.
BTW, 7.4.8 is pretty old; you really ought to be on something more
current. The latest release in that branch is 7.4.18 --- you are
missing two years worth of bug fixes.
regards, tom lane