i have a situation on my sandbox:
pg_xlog directory of cluster was gone at stopped postgres. it was on another disk
which has broken. in this situation postgres falling to .core at startup with messages:
LOG: could not open file "pg_xlog/000000030000000100000058" (log file 1, segment 88): No such file or directory
LOG: invalid primary checkpoint record
before that happened postgres was shutdown normally.
i so understand that database is consistent, no? why we couldn't start?
pgsql-admin by date
|Next:||From: Tom Lane||Date: 2008-01-23 22:48:41|
|Subject: Re: lost /pg_xlog |
|Previous:||From: Tom Lane||Date: 2008-01-23 20:56:58|
|Subject: Re: no entry on pg_database for a live database |