Re: RFC about socket/cores numbers and HT info
by Matthew J. Roth
Gianluca Cecchi wrote:
> b) Do you think there is any way to enumerate the number of physical
> sockets of a system based only on values in cpuinfo and not using
> dmidecode command?
Gianluca,
There is a Red Hat Knowledgebase article titled "How do I determine if my x86-compatible Intel system is multi-processor, multi-core or supports hyperthreading?" that I think you will find useful. It states:
To determine whether a system is multi-processor, multi-core, has
hyperthreading or supports a combination of the three, look at the
physical id, siblings, core id and cpu cores values in /proc/cpuinfo
on Red Hat Enterprise Linux 4 or on Red Hat Enterprise Linux 5 while
running a non-Xen kernel.
It provides several examples that you could use to write and test your script.
Here's the link: http://kbase.redhat.com/faq/docs/DOC-7715
Regards,
Matthew Roth
InterMedia Marketing Solutions
Software Engineer and Systems Developer
14 years, 4 months
24 hours format in gdm
by François Patte
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Bonjour,
On f12, gdm-greter display hours in 12 hours format: how to change it in
24 hours format?
Thank you.
- --
François Patte
UFR de mathématiques et informatique
Université Paris Descartes
45, rue des Saints Pères
F-75270 Paris Cedex 06
Tél. +33 (0)1 4286 2145
http://www.math-info.univ-paris5.fr/~patte
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org/
iEYEARECAAYFAkthrn0ACgkQdE6C2dhV2JUQdQCfduqqJU0Gq0WAdL1Dn97Huu61
uMsAn1CvlDi4WRhMIAAjrFs9EHuZTKsn
=L2Ps
-----END PGP SIGNATURE-----
14 years, 4 months
Recompile PHP package
by Rodrigo B Brasil
Hello!
I have the
--
Rodrigo Bezerra Brasil
I.M: rodrigobrasil[at]live[dot]com
Tel: (91) 8115.5998
Belém, PA, Brazil
'"Software Livre" é uma questão de liberdade, não de preço. Para entender o
conceito, você deve pensar em "liberdade de expressão", não em "cerveja
grátis".'
14 years, 4 months
Demand for LUKS password blocks boot
by Robert Nichols
In my Fedora 12 system, if an external LUKS encrypted volume happens to be
connected to the system while it is booting, the boot sequence blocks
waiting for the password. This means that my system cannot boot unattended
(perhaps following a power failure) if an encrypted external drive happens
to be connected. This is completely unacceptable. How can I get the
system to ignore this unnecessary external volume during boot? This could
be any random device, so I can't set up a rule to ignore specific device
identifiers.
--
Bob Nichols "NOSPAM" is really part of my email address.
Do NOT delete it.
14 years, 4 months
Httpd
by jarmo
Hi
I'm still suffering a bit httpd service mallfunction...
[root@oh1mrr LimeWire]# service httpd start
Käynnistetään httpd: [ OK ]
Error_log says:
[Thu Jan 28 19:10:10 2010] [notice] core dump file size limit raised to
4294967295 bytes
[Thu Jan 28 19:10:10 2010] [notice] suEXEC mechanism enabled (wrapper:
/usr/sbin/suexec)
[Thu Jan 28 19:10:10 2010] [notice] Digest: generating secret for digest
authentication ...
[Thu Jan 28 19:10:10 2010] [notice] Digest: done
[Thu Jan 28 19:10:11 2010] [error] python_init: Python version mismatch,
expected '2.6', found '2.6.2'.
[Thu Jan 28 19:10:11 2010] [error] python_init: Python executable found
'/usr/bin/python'.
[Thu Jan 28 19:10:11 2010] [error] python_init: Python path being used
'/usr/lib/python26.zip:/usr/lib/python2.6/:/usr/lib/python2.6/plat-
linux2:/usr/lib/python2.6/lib-tk:/usr/lib/python2.6/lib-
old:/usr/lib/python2.6/lib-dynload'.
[Thu Jan 28 19:10:11 2010] [notice] mod_python: Creating 4 session mutexes
based on 150 max processes and 0 max threads.
[Thu Jan 28 19:10:11 2010] [notice] mod_python: using mutex_directory /tmp
[Thu Jan 28 19:10:11 2010] [notice] Apache/2.2.14 (Unix) DAV/2 PHP/5.3.1
mod_python/3.3.1 Python/2.6.2 mod_ssl/2.2.14 OpenSSL/1.0.0-fips-beta4
mod_perl/2.0.4 Perl/v5.10.0 configured -- resuming normal operations
[root@oh1mrr LimeWire]# service httpd stop
Suljetaan httpd: [VIRHE ] <--- That
means, closing httpd: [failure]
Error_log says nothing at this point, I have to kill pid of httpd manually, if
want to stop it....
Any ideas?
Jarmo
14 years, 4 months
RFC about socket/cores numbers and HT info
by Gianluca Cecchi
Hello,
after digging around different systems (Fedora, RHEL) and mailing
lists suggestions I arrived at this script about identifying on a
system:
- number of sockets
- number of cores per socket
- number of logical processors (so computational units) seen by OS
- info about presence/enablement of HT if present
I would like to share and ask 2 things:
a) what about the script accuracy?
Can you verify with your hw setup/config?
b) Do you think there is any way to enumerate the number of physical
sockets of a system based only on values in cpuinfo and not using
dmidecode command? I have a system (not under my control) that
collects nmon values only for many servers and I would like to be able
to get the number of sockets ion them based on this info...
In nmon output I find, other than the cpuinfo info, only something like
AAA,cpus,4,4
But it seems quite static where the two numerical values are identical.
Thanks in advance,
Gianluca
14 years, 4 months
how to switch the lcd panel off when lid is closed
by François Patte
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Bonjour,
Everything in the subject, except that I want this to be when nobody is
logged on the computer.
(When someone is logged on he can choose this with gnome-power-manager)
Thanks
- --
François Patte
UFR de mathématiques et informatique
Université Paris Descartes
45, rue des Saints Pères
F-75270 Paris Cedex 06
Tél. +33 (0)1 4286 2145
http://www.math-info.univ-paris5.fr/~patte
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org/
iEYEARECAAYFAkth0w0ACgkQdE6C2dhV2JWYewCfVTCHBNfmTcc9lQP54NXMhfdD
2KUAn3Syn2n320RBA0r80NXnfRPDEaLX
=mId2
-----END PGP SIGNATURE-----
14 years, 4 months
F12: Latest Two Kernels Won't Boot
by Oliver Ruebenacker
Hello,
When a few weeks ago the new auto-updated kernel would not boot,
used an older kernel, waiting for the next update. This was today, and
now the two latest kernels don't work. Booting stops after the Fedora
logo fills up white, after which it says (copying manually from
screen, whitespaces probably not correct):
===
ERROR: asr: wrong # of devices in RAID set "asr_BOOT " [1/2] on /dev/sda
No root device found
Boot has failed, sleeping forever
===
The third oldest kernel apparently boots fine. Interestingly, Google
finds nothing for "ERROR: asr: wrong".
I am attaching grub.conf.
Thanks!
Take care
Oliver
--
Oliver Ruebenacker, Computational Cell Biologist
Systems Biology Linker at Virtual Cell (http://vcell.org/sybil)
Turning Knowledge Data into Models
Center for Cell Analysis and Modeling
http://www.oliver.curiousworld.org
14 years, 4 months
Boot Issues:
by Jeff Maxwell
After installing the latest upgrades including Kernel updates, I have
the following message when rebooting:
Fatal could not
load /lib/modules/2.6.31.12-174.2.3.fc12.i686/modules.dep
No root device found
boot has failed, sleeping forever
I am able to boot using the prior version of the Kernel.
Thanks.
14 years, 4 months