Discuss

discuss@lists.centos.org

34753 discussions

Re: [CentOS] evaluating backup systems: rsync
by Gordon Messmer 14 Jan '13

14 Jan '13

On 01/13/2013 02:59 AM, Reindl Harald wrote: >> There is no common >> >mechanism for making files and databases consistent and making a >> >snapshot for backups. Admins must do this on their own. If you aren't >> >actively taking steps to make your backups consistent, they aren't. > open-vm-tools 2012.12.26 changes: > * vmsync is not longer being compiled on newer (3.0+) Linux > kernels since they support FIFREEZE/FITHAW ioctls. That's only needed to make a filesystem consistent for a snapshot at a lower level (such as the raid controller or VM hypervisor). It doesn't integrate with higher level components to make files and databases consistent. So, like I said, Linux falls far short of Windows' backup infrastructure, and if you aren't actively making your files consistent for a backup, they won't be. All of the required components are present, but it's up to admins to do all of the integration work.

1 0

Upgrading Python with NumPy, SciPy and Mayavi on a CentOs 6.2 (Rocks 6.0) cluster
by Patrick Haley 13 Jan '13

13 Jan '13

Hi, We are looking for some guidance in installing an upgraded Python on our cluster. Our cluster was installed with Rocks 6.0, is running CentOs 6.2, and has python-2.6.6, gcc-4.4.6. We would like to install an upgraded version of Python along with the following modules NumPy Scipy (which will require a compatible version of the Atlas libraries) Mayavi (which will require a compatible version of vtk) We were wondering if anyone has experience in installing these packages, what are the best combinations of compatible versions, what other upgrades will be needed (e.g. will we need to upgrade the gnu compilers to install the latest versions) and where these should be installed (local disks, /share/apps, other?). A little more background, our students are currently working with a local installation using Python-2.7.1 numpy-1.5.1 scipy-0.9.0b1 Mayavi-3.4.0 ATLAS-3.8.3 vtk-5.4 We would like to install these packages in a more appropriate, cluster-wide area and upgrade these packages as appropriate. We are also aware of a Rocks python roll which contains python-2.7.2 and an upspecified version of NumPy (is it possible to find out which version, before installing the roll?). Thanks -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Pat Haley Email: phaley(a)mit.edu Center for Ocean Engineering Phone: (617) 253-6824 Dept. of Mechanical Engineering Fax: (617) 253-8125 MIT, Room 5-213 http://web.mit.edu/phaley/www/ 77 Massachusetts Avenue Cambridge, MA 02139-4301

1 0

gigantic memory leak in Clock Applet...
by fred smith 13 Jan '13

13 Jan '13

I've discovered recently that something on my Centos 5.8 box (up to date) is hogging a ton of RAM. so a little while ago I sat and watched top for a while. it showed (sorry, I didn't take screen shots or write this down, so the numbers are a bit rough) that out of 8 gigs of swap, around 2 1/2 was in use, and all the RAM (except for the little the kernel keeps for itself) was in use (it's got 4 gigs). this might not sound bad, but there's hardly ever anything big running on this box, it's just my home desktop machine used mostly for web browsing/music/email and similar. so, watching top run for a while I could eventually make out that something had "1.6g" flashing in the "RES" column. slowing the refresh a little I saw that it was "clock applet". so I killed the clock applet and restarted it, then clock applet showed "11m" in the "RES" column, and the unused RAM was suddenly like a gig and 3/4, or so, and the swap used slowly started dropping while the free ram began being used up, as it normally should. as I continue to watch it run (10-15 mins later) I can see that clock applet is now showing 14m in the RES column, so it's still growing. Is anyone else seeing the clock applet hogging (tons of tiny leaks, I assume) RAM needlessly? -- ---- Fred Smith -- fredex(a)fcshome.stoneham.ma.us ----------------------------- I can do all things through Christ who strengthens me. ------------------------------ Philippians 4:13 -------------------------------

12 37

luks and aes-ni
by Markus Falb 13 Jan '13

13 Jan '13

Hi, Short version: If I had a CPU with the aes-ni [1] feature would luks use it? I know that Upstream Vendors Security Guide [2] says: ...snip The default cipher used for LUKS (refer to cryptsetup --help) is aes-cbc-essiv:sha256 (ESSIV - Encrypted Salt-Sector Initialization Vector). Note that the installation program, Anaconda, uses by default XTS mode (aes-xts-plain64) snap... I also found a notion in the forums that maybe only aes-cbc is using aes-ni [3] and that could mean that after a install aes-ni is not used at all. Does anyone know about this or has experiences? [1] http://en.wikipedia.org/wiki/AES_instruction_set [2] https://access.redhat.com/knowledge/docs/en-US/Red_Hat_Enterprise_Linux/6/h… [3] http://forum.centos.org/modules/newbb/viewtopic.php?topic_id=38226&forum=56… -- Kind Regards, Markus Falb

1 0

Re: [CentOS] wiping out data on a disk (no physical acess to the machine)
by Gordon Messmer 13 Jan '13

13 Jan '13

On 01/12/2013 04:35 PM, Reindl Harald wrote: > Usually if no service is running "dd if=/dev/zero of=/dev/your-sysdisk" > does crash more or less very late and if you destory the datadisks before > there is nearly zero chance to recover any data If I care enough to wipe the disks in a server, "usually" and "late" is not going to cut it. Any attempt by the kernel to read any filesystem is likely to cause a panic before wiping is complete. If you want to completely wipe a disk, you need the root filesystem to be somewhere else.

1 0

Why is localhost self-signed cert a CA cert?
by Robert Moskowitz 13 Jan '13

13 Jan '13

I am building a mail server on Centos 6.3 and working with OpenSSL to create a self-signed certificate for mail use. Along the line of learning the 'best' options to use for OpenSSL and dealing with the default SSL virtual host for Apache, I discovered that the localhost cert created (I believe) during firstboot has the X509v3 extensions set as a CA cert (eg basicConstraint CA:TRUE). I was once very involved in PKIX and legal issues on certificate policy. Having the localhost cert being a CA cert, thus allowed to sign other certs, MAY have legal implications in the USofA and EU. Why was this chosen? Why is not -extensions v3_req used in the certificate creation? Oh you can see this for yourself with: openssl x509 -in /etc/pki/certs/localhost.crt -text -nameopt multiline -noout|more

3 11

use of hard links and diff in rsync
by ken 12 Jan '13

12 Jan '13

On 01/11/2013 07:08 PM Reindl Harald wrote: > .... > different versions of the backup > daily, weekly, monthly Ah, yes, in that kind of scenario hard links would be useful. > unchanged files are replaced with hard-links > the destination files are virtually on the same place And so then could changes to a file be recorded in the daily version as a diff against the weekly?

3 2

CentOS-announce Digest, Vol 95, Issue 4
by centos-announce-request＠centos.org 12 Jan '13

12 Jan '13

Send CentOS-announce mailing list submissions to centos-announce(a)centos.org To subscribe or unsubscribe via the World Wide Web, visit http://lists.centos.org/mailman/listinfo/centos-announce or, via email, send a message with subject or body 'help' to centos-announce-request(a)centos.org You can reach the person managing the list at centos-announce-owner(a)centos.org When replying, please edit your Subject line so it is more specific than "Re: Contents of CentOS-announce digest..." Today's Topics: 1. CentOS 5.9 i386 and x86_64 Packages Available in the Continuous Release (CR) Repository (Johnny Hughes) ---------------------------------------------------------------------- Message: 1 Date: Fri, 11 Jan 2013 07:46:19 -0600 From: Johnny Hughes <johnny(a)centos.org> Subject: [CentOS-announce] CentOS 5.9 i386 and x86_64 Packages Available in the Continuous Release (CR) Repository To: CentOS-Announce <centos-announce(a)centos.org> Message-ID: <50F017AB.10406(a)centos.org> Content-Type: text/plain; charset="iso-8859-1" The CentOS-5.9 Packages for the i386 and x86_64 Architectures are now available via the Continuous Release (CR) Repository. The CR Announcements for CentOS-5.9 can be viewed here: http://lists.centos.org/pipermail/centos-cr-announce/2013-January/thread.ht… More information about the CR Repository and how to install it is available here: http://wiki.centos.org/AdditionalResources/Repositories/CR As explained in the above link, we are still doing advanced QA testing on CentOS-5.9 and there may be some minor changes before the final release. This "opt-in" CR Repo process allows earlier access to our best effort packages now while we complete the final CentOS-5.9 QA processes, spin the ISOs and create an installable tree. Thanks, Johnny Hughes

1 0

Re: [CentOS] Building Dovecot CentOS 5 RPMs with custom LDAP packages
by Nikolaos Milas 12 Jan '13

12 Jan '13

On 11/1/2013 1:44 μμ, fakessh @ wrote: > I do not know the only thing I can tell you that laziness is a value in > the computer You are partly right. However, IMHO building dovecot using standard openldap-devel on CentOS 5 means that, even though the final RPM works, it will use *ancient* openldap v2.3 libraries. If we manage to use custom openldap libaries (like those from LTB Openldap builds) then Dovecot can be built against modern/current Openldap v2.4 libraries. That may provide significant improvements to the final RPM, in terms of LDAP usage. In any case, I found the solution (based on the documentation: http://wiki2.dovecot.org/CompilingSource) I added the following lines (marked below with ++) in your spec file: %build export PATH=/usr/kerberos/bin:$PATH #required for fdpass.c line 125,190: dereferencing type-punned pointer will break strict-aliasing rules ++ export CPPFLAGS="${CPPFLAGS} -I/usr/local/openldap/include" ++ export LDFLAGS="${LDFLAGS} -L/usr/local/openldap/lib64" export CFLAGS="$RPM_OPT_FLAGS -fno-strict-aliasing" %configure INSTALL_DATA="install -c -p -m644" \ --docdir=%{_docdir}/%{name}-%{version} \ --disable-static \ --disable-rpath \ --with-nss \ --with-shadow \ --with-pam \ --with-gssapi=plugin \ --with-ldap=plugin \ ... So, this produces a dovecot package based on Openldap 2.4 libraries(follows excerpt from the rpmbuild output): ... Requires(interp): /bin/sh /bin/sh /bin/sh /bin/sh Requires(rpmlib): rpmlib(CompressedFileNames) <= 3.0.4-1 rpmlib(PayloadFilesHavePrefix) <= 4.0-1 Requires(pre): /bin/sh shadow-utils Requires(post): /bin/sh chkconfig shadow-utils Requires(preun): /bin/sh chkconfig initscripts shadow-utils Requires(postun): /bin/sh initscripts Requires: /bin/bash /bin/sh config(dovecot) = 1:2.1.13-3.centme initscripts ... libldap-2.4.so.2()(64bit) ... ... It was initially using: libldap-2.3.so.0()(64bit) Regards, Nick

1 1

Re: [CentOS] grub issues on reinstall. [resolved-ish]
by zGreenfelder 12 Jan '13

12 Jan '13

>> running find /grub/grub.conf from the grub shell when the machine >> tries to boot returns (hd0,0) >> and /dev/sdi1 is where /boot mounts. >> > That's *very* odd. You don't have an HBA or something else - flash key? > anything? What's on this system that it sees stuff that high? > > Directly related: look at your BIOS, and see what drives it sees, and > what's bootable, in what order. Perhaps you can move up the boot drive? > <snip> > So I went back to this tonight and fought, fought and fought some more. there was no HBA sort of thing, just a little home machine, the BIOS could see all the drives and were all in the order I expected, but I never got anywhere with trying to write to the MBRs of any of the drives. I could go into recovery mode and get the installed system to come up that way, utilities like grub-install seemed to hang in attempting to write to the devices. I ended up just bailing, took a USB key installed it (it showed up as /dev/sda) and installed grub to the MBR there so now I can boot. Thanks for the suggestions. -- Even the Magic 8 ball has an opinion on email clients: Outlook not so good.

1 0

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Discuss