syslog-ng's matching vs a perl-regex match
Hi all, I am using syslog-ng version 1.4.12 on red hat linux 7.1. I have been doing some matches to get certain events out of my logs and have certain programs react to them. I was wondering if it would be faster to pipe ALL output to a perl script and have it do its own regex matching. I'm not sure how fast syslog-ng's matching is and if I take a performance hit by using it. thanks -sv
On Thu, Oct 04, 2001 at 08:30:44PM -0400, Seth Vidal wrote:
Hi all, I am using syslog-ng version 1.4.12 on red hat linux 7.1. I have been doing some matches to get certain events out of my logs and have certain programs react to them. I was wondering if it would be faster to pipe ALL output to a perl script and have it do its own regex matching.
I'm not sure how fast syslog-ng's matching is and if I take a performance hit by using it.
syslog-ng is using a the regexp lib in libc. -- Bazsi PGP info: KeyID 9AF8D0A9 Fingerprint CD27 CFB0 802C 0944 9CFD 804E C82C 8EB1
--On Tuesday, October 09, 2001 11:32 AM +0200 Balazs Scheidler <bazsi@balabit.hu> wrote:
syslog-ng is using a the regexp lib in libc.
If it's using the POSIX interfaces, I recommend linking it against the PCRE (Perl Compatible Regular Expression) library. In my tests, it's the fastest, and supports the perl RE syntax (with one or two very obscure exceptions). If it's not using the POSIX interfaces, porting code from the old regexp APIs to the new POSIX API is trivial. -- Carson
participants (3)
-
Balazs Scheidler
-
Carson Gaspar
-
Seth Vidal