From: Enrico Scholz <enrico.scholz@informatik.tu-chemnitz.de>
Subject: Re: [runit/svlogd] RFE: periodic log-rotation; filter-malfunction
Date: Thu, 02 Sep 2004 23:58:21 +0200 [thread overview]
Message-ID: <87llfse3k2.fsf@kosh.ultra.csn.tu-chemnitz.de> (raw)
In-Reply-To: <20040902194813.8878.qmail@892d5ec05c70b4.315fe32.mid.smarden.org> (Gerrit Pape's message of "Thu, 2 Sep 2004 19:48:11 +0000")
pape@smarden.org (Gerrit Pape) writes:
>> Would it be possible to add another option (e.g. 'a<seconds>') which
>> sends SIGALRM at <seconds> seconds after the last logrotation?
>
> I'll have to think about it, in the meantime you could setup a separate
> service doing this work, e.g.:
>
> /var/service/socklogrotate/run:
> #!/bin/sh
> runsvctrl alarm /var/service/socklog
> exec sleep <seconds>
I do not like this, because:
* it adds extra bloat to the system (it is bad enough that I have to spawn
a stunnel-instance per logrotation; an additional sleep+bash-process
would increase system-load significantly). Without looking at the code,
I guess that an implementation in svlogd would be much smaller because
the SIGALRM handler is already there.
(Things are getting worse because I run 20-30 svlogd instances per
machine in linux-vservers[1] (chroots, BSD-jail like))
* it requires additional software (svlogd is used without the other
runit tools here) and setup.
* it would logrotate every buffer inclusive those which should not be
touched (e.g. I have a local debug-buffer which I want to keep). 'a...'
could send a new signal (USR2??) which logrotates only the outdated
buffers (I know that it conflicts with my initial idea...).
>> Another issue: the '-/+...' filters seem to fail sometimes.
> ...
> This shouldn't happen. Can you please give an exact example of a line
> that ends up in the wrong log, copied and pasted, and also re-check that
> you post the exact contents of the config files? Do you use the -r or
> -R command line options to svlogd in socklog-ucspi-tcp/log/run?
Please see http://www.tu-chemnitz.de/~ensc/runit/badfilter.txt
When you need de-anonymoused values, please contact me privatly.
Enrico
Footnotes:
[1] http://linux-vserver.org
next prev parent reply other threads:[~2004-09-02 21:58 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-09-02 9:55 Enrico Scholz
2004-09-02 19:48 ` Gerrit Pape
2004-09-02 19:53 ` Charlie Brady
2004-09-02 20:12 ` Gerrit Pape
2004-09-02 21:58 ` Enrico Scholz [this message]
2004-09-03 11:08 ` Gerrit Pape
2004-09-11 9:56 ` Gerrit Pape
2004-09-03 14:54 ` Joan Picanyol
2004-09-03 15:15 ` Charlie Brady
2004-10-18 17:18 ` Gerrit Pape
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=87llfse3k2.fsf@kosh.ultra.csn.tu-chemnitz.de \
--to=enrico.scholz@informatik.tu-chemnitz.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).