Introduction (and cfengine)
by David Douthitt
Hello!
I am a UNIX System Administrator from the midwest (Wisconsin), and have
worked with multiple open source and proprietary UNIX and Linux
environments. (I'll skip over the UNIX and non-Red Hat experiences
unless there is interest.) I've used Linux on non-Intel machines as
well as Intel machines (PARISC and PowerPC specifically). I've used BSD
systems as well, including on non-Intel architectures. I'm also
certified: RHCE, SCSA, Linux+, and LPI Level 1.
First time I used Linux was Slackware on a 386SX-16MHz machine with 4M
of RAM; my oldest Red Hat is Red Hat 2 on a Linux User Group CD I still
have around here. I also have 4.4BSD Lite and FreeBSD 2.2.6 and OpenBSD
2.6 on CDs.....
I had my own floppy-based Linux distribution (Oxygen) based on the Linux
Router Project. I have, in the past, also "built" my own Red Hat
Enterprise systems without the benefit of an installer, and also in some
cases from source RPMs alone (rather than binary RPMs).
As far as cfengine goes, I contribute sporadically to the cfengine
documentation, mostly proofreading and copyediting. I will be embarking
on a book about cfengine shortly (after I complete the GNU screen
book). I have run cfengine in a production environment and tend to be
rather a voracious supporter of cfengine :-)
I've not used puppet, but am a Ruby programmer and would like to learn
more about it - it was designed after using cfengine and was designed to
use the power of Ruby.
I'm also a long-time Korn shell scripter - I used ksh for scripting
before I used ksh as my shell! (I used csh then, but programmed in ksh.)
I've also programmed in Perl. I've used Perl since the MSDOS 4.0 port,
and Ruby since version 1.4.
I also tend to be something of a programmer and programming languages
nut; I've used FORTH (extensively and comprehensively), Smalltalk
(dabble), LISP (moderate), and BASIC, C, Pascal, C++, COBOL, SNOBOL, all
moderate to extensively. I contributed to the LCDd project briefly
(mostly revamping the system to use GNU getopts).
I'll be joining the documentation group at the same time. I'll also
post a cfengine review shortly.
--
David Douthitt
HP-UX, Unixware, Linux, FreeBSD
RHCE, SCSA, Linux+, LPIC-1
http://www.lulu.com/ssrat
17 years, 4 months
Introduction / Xen Stuff
by Michael DeHaan
Hi,
My name is Michael DeHaan, and I work in Red Hat's emerging technologies
group in Raleigh, NC. I joined this list because (A) I really want to
get more involved in the Fedora project beyond maintaining some extras
packages, (B) I'm hoping that I can contribute some on the Xen and
systems-management side of things, since I've heard infrastructure is
going to start to do more with virtual machines. I also have a good
deal of experience with other companies build systems (but not Red Hat's
or Fedora's) and possibly I might be able to contribute some there also,
or at least learn more about Fedora's setup. Mainly I'm interested in
systems management and virtualization stuff though.
Xen-wise, two things. I wrote and maintain a provisioning application
called cobbler (http://cobbler.et.redhat.com), which potentially could
be useful for provisioning Xen boxes in the infrastructure environment.
Maybe. If not, maybe it can be extended, and feedback is awesome. If
it's not useful, that's ok.
The other thing is that I'm looking at ways to deploy Xen in easy ways
for Fedora users, basically an open way of managing a /lot/ of Xen
machines with integrated configuration management, central interfaces,
and not a lot of setup or technical knowledge needed. That's all
pretty new, though if some of that would be useful for infrastructure as
it evolves, that would be pretty neat. Either way, I figure I can
learn from ways that Fedora is using Xen just the same.
Anyhow, hi.
--Michael
17 years, 4 months
Mailing List Troubles?
by David Douthitt
I sent my signup to the list using the *old-fashioned* way: I sent an
email to fedora-infrastructure-list-request(a)redhat.com (and to
fedora-documentation-list-request(a)redhat.com) with "subscribe" in the
subject and mail with a standard signature (marked by dash-dash-space on
a line).
Both replied twice (for whatever reason!).
I also seem to have sent off an HTML mail for my intro! Ack! I
expected it to ask (at which point I'd use text only) but natch....
--
David Douthitt
HP-UX, Unixware, Linux, FreeBSD
RHCE, SCSA, Linux+, LPIC-1
http://www.lulu.com/ssrat
17 years, 4 months
Introduction
by Ask Bjørn Hansen
Hi everyone!
I've written another mail to the list I'm about to send, but I
figured it'd be polite to introduce myself first. I think it's very
nice how most contributors to this list have been doing that.
I'm Ask Bjørn Hansen. My little consulting company develops neat
application for companies large and small (lately mostly larger ones)
and occasionally at conferences and to companies I give advice on how
to architect web applications to scale[1].
More relevant for the work you guys are doing here, then I've been
running many of the perl.org services for the Perl community since
1999 or so. We usually use RHEL (3 and 4), but lately a bit of
Fedora too in addition to a FreeBSD system here and there. I also
have a cabinet with Fedora servers for a startup I am, uh, starting.
We might change those to RHEL5 at some point, but more likely then
we'll get the Stateless Linux magic working instead and use that for
managing all the systems (virtual and physical).
I joined this list to absorb any tips and ideas for running a cluster
of RH/Fedora boxes -- and ideally to offer my help, but of course I
haven't had any time for that.
- ask
[1] http://develooper.com/talks/
--
http://develooper.com/ - http://askask.com/
17 years, 4 months
Load test 1 - Sat Dec 23rd
by Mike McGrath
So the load test for the proxy servers is scheduled for this Saturday,
Dec 23rd at 10:00 a.m. (GMT). It will involve at most, proxy1,
proxy2, app1, and app2. Though will most likely not use both app
servers. One of the proxy servers will be used as a client so none of
the other RH systems (including the balancer for now) should be
affected.
Depending on the results of this test we will proceed with another
test at a later date. The test should take no longer than 2 hours but
may go 3.
Stacy, if this doesn't work for you let me know, this test sounds
pretty sane and shouldn't have any impact on anything other than
fedora servers. It'll also happen during non-peak hours which is
nice.
-Mike
17 years, 4 months
TurboGears and version control
by Mike McGrath
So many of us have started using TurboGears for Fedora infrastructure
projects. My question is how should we be version controlling the
source? Do we check in the entire project or just the related kid
templates, configs, controller files, etc. When you create a project
with one version of TurboGears does it work with other versions? How
do we prevent mismatches?
-Mike
17 years, 4 months
gary_lerhaupt@dell.com Re: Undelivered Mail Returned to Sender
by Warren Togami
Matt, is this person no longer at Dell? Who do we reassign the package
ownership to?
Warren Togami
wtogami(a)redhat.com
Mail Delivery System wrote:
> This is the Postfix program at host buildsys.fedoraproject.org.
>
> I'm sorry to have to inform you that your message could not
> be delivered to one or more recipients. It's attached below.
>
> For further assistance, please send mail to <postmaster>
>
> If you do so, please include this problem report. You can
> delete your own text from the attached returned message.
>
> The Postfix program
>
> <gary_lerhaupt(a)dell.com>: host smtp.ins.dell.com[143.166.83.183] said: 501
> #5.1.1 bad address gary_lerhaupt(a)dell.com (in reply to RCPT TO command)
>
>
> ------------------------------------------------------------------------
>
> Reporting-MTA: dns; buildsys.fedoraproject.org
> X-Postfix-Queue-ID: AEB04152140
> X-Postfix-Sender: rfc822; buildsys(a)fedoraproject.org
> Arrival-Date: Fri, 15 Dec 2006 16:32:59 -0500 (EST)
>
> Final-Recipient: rfc822; gary_lerhaupt(a)dell.com
> Action: failed
> Status: 5.0.0
> Diagnostic-Code: X-Postfix; host smtp.ins.dell.com[143.166.83.183] said: 501
> #5.1.1 bad address gary_lerhaupt(a)dell.com (in reply to RCPT TO command)
>
>
> ------------------------------------------------------------------------
>
> Subject:
> Broken dependencies in Fedora Extras - 2006-12-15
> From:
> Fedora Extras repoclosure <buildsys(a)fedoraproject.org>
> Date:
> Fri, 15 Dec 2006 21:32:59 -0000
> To:
> gary_lerhaupt(a)dell.com, bugs.michael(a)gmx.net
>
> To:
> gary_lerhaupt(a)dell.com, bugs.michael(a)gmx.net
>
>
> This is an automated mail created by an experimental script.
> Your following packages in the repository contain broken dependencies:
>
> package: dkms - 2.0.13-1.fc6.noarch from fedora-extras-development-ppc
> unresolved deps:
> kernel-devel
>
17 years, 4 months
Web server torturing
by Ahmed Kamal
Hi,
Due to the web server hit faced with FC6 release, some actions are being
taken to minimize the chance of facing such issues again. One of the steps
is to stress test our web-server infrastructure to measure our current and
future capabilities. I'd like to run some tests on fp.o web-server, please
let me know your thoughts/comments. Here are some details.
Test Targets:
1- Measure our bare (no caching) maximum serving rate
2- Measure our cached serving rate, to assess the implemented caching
efficiency
3- Gather numbers like (When do we get CPU saturated, RAM requirements ..).
Possibly draw graphs (everyone thinks graphs are cool), the numbers should
help us base future calculations on a solid basis
4- Future: Possibly implement a mechanism to cap the maximum connected
clients to a specific server, to the maximum it can handle gracefully, to
avoid killing a server
Test Plan:
1- A script was written which uses apache's ab tool to stress the server.
Script will run on the web-server host.
2- The script fires a total number of connections equal to ten times the
maximum concurrency rate (to get good average, and avoid transients)
3- The concurrency rate is sweeped between 10 and 400 (my 1G-RAM machine
swaps at about 100 connections)? any suggestions?
4- All ab output is recorded for future analysis
5- A monitoring thread is fired before ab is launched. The monitoring uses
"top" to record load/cpu/ram/process information in log files as well
6- Tests are repeated with "ab -k" for enabling the HTTP keep-alive option.
Not sure if this is needed, or if it will make much difference! comments?
7- Tests are done once with caching enabled and one more time without
caching
Please let me know your thoughts about the testing setup, should we be
recording more data? should we be stressing the server in a different way,
should we be testing some apache config options ... etc ?
Thanks
17 years, 4 months