Re: End of hypocrisy ?
On Friday, August 8, 2014 5:40:01 AM UTC+5:30, Joel Rees wrote:
> On Fri, Aug 8, 2014 at 6:19 AM, AW <debian.list.tracker@1024bits.com> wrote:
> > On Thu, 7 Aug 2014 16:44:39 -0400
> > Tom H wrote:
> > > journalctl has output options:
> > > -o, --output=
> > > Controls the formatting of the journal entries that are shown. Takes
> > > one of the following options:
> > Seems fine to me after letting go of first impression of distrust in new
> > things...
> > However, I still like my pet idea of postgresql --- and SQL is much more fun
> > than journalctl statements...
> > So, back to the ranch I go...
> You do understand the chicken-and-egg nature of what you're asking for?
> You're needing to output logs to but up servers, but you have to boot
> a server as complex as anySQL server to get there. Where is anySQL
> going to log its progress as it boots up, not to mention that the
> system itself has to wait for anySQL to get up before it can do
> anything that might generate a log.
> (Having a separate machine be a log server can be useful, but even
> that can't take log messages when the network is not properly up.
> Which means there will be low-level log messages kept in a separate
> place, and sometimes high-level log messages waiting to be off-loaded
> to the log server.)
Yes circular definition is a problem.
And it is ubiquitous and central to our field:
http://blog.languager.org/2012/05/recursion-pervasive-in-cs.html
So... yes an SQL-based log system will need logs itself.
Should it (try to) to log itself?
Is the problem absent with text files?
If one wants to write a text-log one needs a text-file.
A text file exists on some file-system.
What of those messages that need to be logged before there are any filesystems
mounted?
This is not academic:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=698751
[Not sure of the status of the bug]
Reply to: