1 |
Mike Frysinger wrote: |
2 |
|
3 |
>On Tuesday 27 September 2005 02:59 pm, Donnie Berkholz wrote: |
4 |
> |
5 |
> |
6 |
>>Ciaran McCreesh wrote: |
7 |
>>| Ok, for those who don't remember... |
8 |
>>| |
9 |
>>| In the good old days, syslog-ng was the default. Then, for one of the |
10 |
>>| releases, the x86 team accidentally included metalog instead of |
11 |
>>| syslog-ng on the distfiles CD, so the default was changed over to make |
12 |
>>| the x86 networkless install documentation work (at the expense of the |
13 |
>>| archs that got it right). So, it's not a case of moving away from |
14 |
>>| metalog at all, it's a case of undoing a temporary change. |
15 |
>> |
16 |
>>In the bad old days, metalog was the default. Then it got switched to |
17 |
>>syslog-ng because people got confused as hell by metalog's buffering, |
18 |
>>and there was no reason not to use a more capable logger that's |
19 |
>>compatible with the traditional sysklogd, minus the annoying feature of |
20 |
>>metalog. |
21 |
>> |
22 |
>> |
23 |
> |
24 |
>ok, i thought that's how *i* remembered it :) |
25 |
> |
26 |
> |
27 |
> |
28 |
>>FWIW, I think metalog's buffering disappeared a while back, at least by |
29 |
>>default. |
30 |
>> |
31 |
>> |
32 |
> |
33 |
>yes, default metalog config file is unbuffered (and has been for a while), and |
34 |
>the init script lets you switch between modes now on the fly so you dont have |
35 |
>to send the kill signals yourself. whether it's 'annoying' is certainly a |
36 |
>matter of opinion :P, but from a technical aspect, buffering makes a lot of |
37 |
>sense if the performance matters. |
38 |
>-mike |
39 |
> |
40 |
> |
41 |
I think syslog-ng is a better logger, except the default configuration |
42 |
is pretty icky. I would shoot for something better, even if it's in |
43 |
/usr/share. I heard hardened's config is quite nice :) At least |
44 |
provide an example of a config that sorts logging out into seperate |
45 |
files like most distro's do. |
46 |
-- |
47 |
gentoo-dev@g.o mailing list |