[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Preliminary log matching support (the third)
From: |
Jan-Henrik Haukeland |
Subject: |
Re: Preliminary log matching support (the third) |
Date: |
Wed, 3 Aug 2005 18:15:05 +0200 |
On 3. aug. 2005, at 17.37, Martin Pala wrote:
Looks good :)
Agree, but run the doc through a spell-checker before checkin :)
Can you send it to cvs? ... here is my +1 ;)
Mine too +1 _when_ the remaining issues are solved. Important! are
there some way to optimize the matching process? Using 6 sec parsing
400 lines are to much! It may screwup the monit cycle. And at least
400 lines are to be expected when a couple of log-files are parsed on
a normally stressed system. As you know from Perl or Pyton, using a
badly written regular expressions can impact performance
considerably. So it's an idea to look at the regex also and see if
the problem lays there as well..
Btw. regarding the regex library - i think that the system library
(standard libc) is sufficient (it is conforming to POSIX).
I'm not sure whether there is some advantage when we will bundle
the optional regex library with monit?
I'm not sure, but there may be some (old) systems that provide the
regex library outside of libc. Anyway I agree that we should *not*
bundle a separate regex package with monit. Configure will find any
external regex lib and if it's not on the system we have fall-back
options to standard string match. Together this should be enough.
--
Jan-Henrik Haukeland
Mobil +47 97141255
smime.p7s
Description: S/MIME cryptographic signature