Discuss

discuss@lists.centos.org

34753 discussions

SysVinit guidance, please
by Robert 05 Sep '07

05 Sep '07

I have experienced a couple of power failures since installing CentOS 5, both while I was working on the system. The first time, several weeks ago, I thought there was some "feature" that I had never seen before but couldn't really put my finger on it. The power came back on after a few seconds and I never missed a beat. I also shrewdly forgot about the anomaly that I had almost noticed. I had a real power failure 2 days ago and tried to be more observant. What had gotten my attention earlier is that the once-familiar banner displayed on about everything except my room's ceiling was missing. After some red herrings and blind alleys (hey, you don't have to worry about mixed metaphors when you're an illiterate old coot!) I decided that the missing banners and the fact that the power was killed while a graphical page was displayed was what I saw. Anyhow, it seems that these extinct banners were a function of the /bin/shutdown routine in SysVinit because banners happen in CentOS 4.x whether the shutdown command is issued by fingers or by a UPS monitoring daemon. And I have since found that they're gone from CentOS 5, no matter how shutdown is invoked. After all of this, my question boils down to one of replacing the current SysVinit package, SysVinit-2.86-14.i386.rpm with the one from CentOS 4.x, SysVinit-2.85-34.3.i386.rpm Would that be courting disaster?

2 1

Hard-set network via ifup scripts...
by Ray Van Dolson 05 Sep '07

05 Sep '07

Anyone know off the top of their head the "right" way to hard set a link automatically to 100Mbps Full Duplex, etc using the ifup or ifcfg files? Don't see options in the scripts that might cover this, but maybe there's a more elegant way to do it than to do something in rc.local or hack up the ifup files myself... Ray

3 2

shrink LV with ext3 filesystem
by Thomas Antony 04 Sep '07

04 Sep '07

Hi. I want to make a snapshot from a logical volume and noticed that there must be enough free disk space in the volume group. Actually there is no free disk space left. How do i shrink online /var without losing any data or restore from backup? I do not have physical access to the server. Specs: Dell PE SC1430 with a 5/i RAID Controller one RAID 1 array from the Dell RAID controller 2 partitons (boot and LVM) 1 VG 3 LV (swap, /var (/dev/VolGroup00/LogVol02) and / formatted with ext3) /dev/mapper/VolGroup00-LogVol02 211G 39G 162G 20% /var Thank you in advance. Thomas Antony

4 10

CPAN error
by Mark Quitoriano 04 Sep '07

04 Sep '07

Hi all im not sure if this is the right mailing list but im using centos anyway. i run CPAN by perl -MCPAN -e shell and got through the first time configuration and after selecting the mirror sites i got this error and i can't get any modules to be installed. I haven't encountered this before and not very familiar with cpan. Constant subroutine __USE_POSIX undefined at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/features.ph line 8, <STDIN> line 33. Constant subroutine __USE_POSIX2 undefined at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/features.ph line 9, <STDIN> line 33. Constant subroutine __USE_POSIX199309 undefined at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/features.ph line 10, <STDIN> line 33. Constant subroutine __USE_POSIX199506 undefined at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/features.ph line 11, <STDIN> line 33. Constant subroutine __USE_XOPEN undefined at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/features.ph line 12, <STDIN> line 33. Constant subroutine __USE_XOPEN_EXTENDED undefined at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/features.ph line 13, <STDIN> line 33. Constant subroutine __USE_UNIX98 undefined at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/features.ph line 14, <STDIN> line 33. Constant subroutine __USE_LARGEFILE undefined at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/features.ph line 16, <STDIN> line 33. Constant subroutine __USE_LARGEFILE64 undefined at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/features.ph line 17, <STDIN> line 33. Constant subroutine __USE_FILE_OFFSET64 undefined at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/features.ph line 18, <STDIN> line 33. Constant subroutine __USE_BSD undefined at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/features.ph line 19, <STDIN> line 33. Constant subroutine __USE_SVID undefined at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/features.ph line 20, <STDIN> line 33. Constant subroutine __USE_MISC undefined at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/features.ph line 21, <STDIN> line 33. Constant subroutine __USE_GNU undefined at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/features.ph line 22, <STDIN> line 33. Constant subroutine __USE_REENTRANT undefined at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/features.ph line 23, <STDIN> line 33. Constant subroutine _POSIX_SOURCE undefined at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/features.ph line 48, <STDIN> line 33. Constant subroutine _POSIX_C_SOURCE undefined at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/features.ph line 50, <STDIN> line 33. Constant subroutine _XOPEN_SOURCE undefined at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/features.ph line 52, <STDIN> line 33. Constant subroutine _XOPEN_SOURCE_EXTENDED undefined at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/features.ph line 54, <STDIN> line 33. Constant subroutine _LARGEFILE64_SOURCE undefined at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/features.ph line 56, <STDIN> line 33. Constant subroutine _LARGEFILE_SOURCE undefined at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/features.ph line 98, <STDIN> line 33. Constant subroutine __USE_ISOC99 undefined at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/features.ph line 102, <STDIN> line 33. Constant subroutine __GNU_LIBRARY__ undefined at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/features.ph line 147, <STDIN> line 33. cpan shell -- CPAN exploration and modules installation (v1.7602) ReadLine support enabled Can't ioctl TIOCGETP: Invalid argument Consider installing Term::ReadKey from CPAN site nearby at http://www.perl.com/CPAN Or use perl -MCPAN -e shell to reach CPAN. Falling back to 'stty'. If you do not want to see this warning, set PERL_READLINE_NOWARN in your environment.

3 3

recompiling httpd
by Matthias Leopold 04 Sep '07

04 Sep '07

hi, i'm trying to recompile the httpd rpm on a x86_64 system. all i've changed in the spec file is the "--with-suexec-docroot" configure option. the error message i get from the "rpmbuild -ba SPECS/httpd.spec" command is "ln: creating symbolic link `/var/tmp/httpd-root/etc/httpd/logs' to `../../usr/var/log/httpd': No such file or directory". so what do i do now? thx matthias -- Mit freundlichen Grüssen Matthias Leopold System & Network Administration Streams Telecommunications GmbH Universitaetsstrasse 10/7, 1090 Vienna, Austria tel: +43 1 40159113 fax: +43 1 40159300 ------------------------------------------------

3 7

NIS/Samba update
by Scott Ehrlich 04 Sep '07

04 Sep '07

My test setup consists of Enterprise 5 server running NIS and Samba, and dual-boot XP and CentOS 5 clients. On my test setup, I'm able to create new samba accounts and change samba passwords with no problems. NIS still gives me problems, though. I am unable to change passwords no matter what. Also, ypwhich on the client returns localhost. I have reviewed everything, including configs of other networks that are live, and cannot figure out where the problem is. I believe that once I can password changed effortlessly on the NIS side, I can implement this on the live system. Until NIS account passwords can be changed at-will, this setup won't go to production. So, what might I be missing? Why can't I change passwords? Also, what would cause the client, when issued ypcat passwd, to not show the newer accounts create by the server, even after a /var/yp make has been issued? Both client and server have been rebooted many times. Thanks. Scott

1 0

Performance Issues
by Dan Dansereau 04 Sep '07

04 Sep '07

My apology for cross posting We have a DELL6850 with 8Gbytes of memory, four 3.2Ghz CPU's , perc 4 raid controller, with fourteen 300Gbyte 10Krpm disk on a powervault 220s, And a powervault 124T LTO-3 tape systems on a separate 160Mbyte/sec adaptec SCSI card. The disks are configured as two 2Tbyte raid 0 partitions using the perc 4 hardware. The problem is - reading from the disk, and writing to the tape is pathetically slow. The specs say I should get around 80Mbyte/second - in reality - is about 500 Kbytes/second Writing just to tape, using /dev/zero as input and dd to write, I can get up to 60Mybte/second, indicating that the tape drive and scsi card is functional from a hardware perspective. Since I can read and write data to the disks - I assume that they are functional. The files are 80Mybtes for the minimum size that I am writing to tape. DELL Support hides behind the fact that CENTOS is not a supported OS, But RED HAT is - so switch to it, so DELL can get RED HAT Support involved. And the fact that Rocks is a modification to an unsupported OS.... I have tried using various block size in tar from 256 to 8192 - with minimal success. Does anybody have any experience on setting the OS/Kernel parameters to utilize the hardware at something more then a snails pace? Or Suggestions on what I could test and/or try? Or Is there a standard disk / tape IO test package that could test the speed/performance of the disks and /or tapes? Thanks for your help Dan A. Dansereau

5 5

Re: [CentOS] RAID + LVM Addition to CentOS 5 Install
by Nick Webb 04 Sep '07

04 Sep '07

On 9/4/07, Nick Webb <webbn(a)acm.org> wrote: > On 9/4/07, Shad L. Lords <slords(a)lordsfam.net> wrote: > > > This worked fine, but on reboot I get a ton of errors from LVM saying > > > that volume with id xxxx-xxxx-xxxx... was not found and the system > > > automatically reboots. This seems to happen for all volumes, not just > > > the ones I changed. This error even happens for a separate volume > > > group (called 'extended') that is on a separate set of disks and was > > > existing prior to the CentOS 5 install. > > > > > > Any idea on some step I missed? I know things are still fine on the > > > disks, as when I boot with the CentOS DVD with the 'linux rescue' > > > option all RAID & LVM volumes are available for use. So from this it > > > seems I need to update some CentOS config file? > > > > Did you re-create the initrd after adding the new raid group? Not 100% > > about this but my guess would be that the initrd isn't starting md2 and thus > > can't find the pv on that raid device. > > > > -Shad > Hi Shad, That solved it! Thanks so much for the help. Awesome group on here, BTW, 4 responses within a hour or two of my post, and a solution! Nick

1 0

RAID + LVM Addition to CentOS 5 Install
by Nick Webb 04 Sep '07

04 Sep '07

Hi All, I have what I believe to be a pretty basic LVM & RAID setup on my CentOS 5 machine: Raid Partitions: /dev/sda1,sdb1 /dev/sda2,sdb2 /dev/sda3,sdb3 During the install I created a RAID 1 volume md0 out of sda1,sdb1 for the boot partition and then added sda2,sdb2 to a separate RAID 1 volume as well (md1). I then setup md1 as a LVM physical volume for volume group 'system'. I left the sda3,sdb3 partitions available for future use. Next I created swap, /, /usr, /var, etc. logical volumes in the system volume group and continued with this install as normal. Everything went fine. I was able to use the system, reboot, etc., without problems. I then discovered that I needed more space in my /var volume than was available in the system volume group. So, I created another RAID device, /dev/md2 (using sda3,sdb3), and created an LVM physical volume on top of that. Finally, I extended the system physical volume to contain this new physical volume and expanded the size of the /var volume. This worked fine, but on reboot I get a ton of errors from LVM saying that volume with id xxxx-xxxx-xxxx... was not found and the system automatically reboots. This seems to happen for all volumes, not just the ones I changed. This error even happens for a separate volume group (called 'extended') that is on a separate set of disks and was existing prior to the CentOS 5 install. Any idea on some step I missed? I know things are still fine on the disks, as when I boot with the CentOS DVD with the 'linux rescue' option all RAID & LVM volumes are available for use. So from this it seems I need to update some CentOS config file? Here are some config files: http://pastebin.com/m6d5075dc Thanks! Nick

4 5

Using HTTP proxy for yum
by Scott Moseman 04 Sep '07

04 Sep '07

Is there a way to use an HTTP proxy (with a user/pass) with yum? Or at least a way to pass a user/pass through yum? I have a situation with a CentOS server behind a web filter appliance. Thanks, Scott

4 9

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