rsyslog

[Will Yonker]

> I demonstrated several simple and practical commands that have
real-life
> use cases, where similar filtering would not be as
simple or direct with
> bare rsyslog.  The addition of journald
was not arbitrary, and I was
> offering genuinely helpful advice,
not trolling.
> 
> If my tone at the beginning suggested
otherwise, it was because I've seen
> threads like this turn into
"I like the old thing fine so Fedora should
> not
>
use the new thing" many, many times.  The discussion is never
productive
> and rarely based on any technical technical argument
.
> 
> As to your question, there are probably answers in
various mailing list
> archives.  Fedora editions that wanted to
have rsyslog do, others do not.
> Consensus on the user support
mailing list is not a prerequisite for
> change.  If you want to
participate in an effective and useful way, I'm
> happy to help
you find that path, but please don't lessen the value of
> this
> list with opinion rants and overtly negative attitudes.

I do think your tone was a bit negative to my somewhat playful response
about losing my house.  So I
don’t think I was “lessening the value of this
list”.  I can figure out
how to use journalctl and will if that is the direction we must move
in.  However, someone must
speak for the trees.  Change by
its very nature is disruptive and by removing something like
/var/log/messages we create a butterfly effect.  Are you one of the developers that
made this change? Â Did you take
into account all the history we have had with that file?  If you are not one of the
developers, why argue the point anyway?  My hope is to catch the eye of the
decision makers and give them a point of view that perhaps they
didn’t have when making this decision.  Or is this the wrong place for such
a hope?
It is true. 
You did demonstrate possible ways to use journalctl and I
appreciate that.  I'm not
against using it.  However I do
also have the responsibility to teach the junior admins at my job how to
get things done on our systems. 
If we need to change, it takes not just learning the new way of
doing it, we have to change our documentation.  In the case of the messages file,
we have to change a wide array of scripts that were developed over years
(sometimes by people no longer with us) to use a new process.  
Why does this matter on Fedora?  Well, isn't Fedora the building
blocks for Redhat Enterprise Linux? 
If we change it here, won't it eventually make it there?  Then it is no longer just an issue
for the Fedora crowd right?  I
use Fedora because we use Redhat. 
Maybe that link isn't as pronounced for everyone else here but
those are _my_ reasons for wanting to moderate the changes.
Anyway, sorry for my reaction there but I think open
discussions as to the reasons for the changes are good and necessary.  It could even be that we find out a
change will cause too much disruption from an open discussion just like
this one.  ^.^
Consensus on the user support mailing list may not be
a prerequisite for change but it should have a strong bearing on said
change.
---
Will Y.

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.fedoraproject.org/pipermail/users/attachments/20150305/1f0ca170/attachment.html>