Hello, You might get more professional and accurate replies if you posted this problem to the postgres mailinglist. You don't seem to have a problem with the syslog-ng daemon.

I'm using Syslong-ng with Postgre SQL Database. I have a problem of queries running realy slow since the Database reached about 5 million records.

That's not really a lot, how do the tables look like and what's your query? If the query is composed of lots of joins or other algebraic operations that have non-linear runtime, you might have hit the point where it's getting expensive. But for that I'd need to see your DB design (ERM if you like) and the query.

Cleaning the DB and vacumdb didn't work and also other programs running on the same computer are running fine.

What is the time spent on? I/O, CPU, Memory pressure? Can you profile it a bit?

Does anyone have an idea ?

Not with that information, I don't. Regards, Roberto Nibali, ratz -- echo '[q]sa[ln0=aln256%Pln256/snlbx]sb3135071790101768542287578439snlbxq' | dc