Hi,
I’m using PostgreSQL 7.1.2 on Sun Solaries 8.
My Server was kept for overnight test and I observed that on my server hostmachine I could not switch user (su) to 'postgres' to see the postgres pid. It looked as if the postgres process was tied up some how.
The server pids were available but postgress was not. This looked like a potential database crash although there are no postgres core files or logs that give clues as to what may have happned
The zipped server logs display lots of error trying to locate the postgres PID
The snapshot of the Postgres log is like this:
ERROR: cannot extend trap: No space left on device.
Check free disk space.
ERROR: cannot extend trap: No space left on device.
Check free disk space.
ERROR: cannot extend trap: No space left on device.
Check free disk space.
ERROR: cannot extend trap: No space left on device.
That was clear from the logs that disk is full. But is there anyway that I can check this programatically? i.e. when I call ‘PgDatabase::Exec()’ to execute query from C++ API, I get error code reflecting ‘low on the disk’? I know I can get corresponding error message in C++ from Postgres by
‘errorMessage()’ method. I want specific error code for this.
Pls. help ASAP.
Thanks in advance,
Anagha Joshi