Discuss

discuss@lists.centos.org

5 participants
34752 discussions

28 Nov '05

Hi! Whew, I have googled around a lot w/this but can't quite seem to come up w/the right answer. This system works beautifully w/nothing wrong w/it. But my goal is to be able to test the raid system by just unplugging hda to mimic a faulty drive and have it just carry on and boot from hdc. md0 = hda1/hdc1 /boot (primary boot partitions on both drives) md1 = hda2/hdc2 / is it possible to just edit /boot/grub.conf??? I am really not sure of this Thx in advance. John Rose

2 1

kernel rpm with imq support
by mbneto 28 Nov '05

28 Nov '05

Hi, I was wondering if anyone mantains a Centos kernel* with IMQ patches applied. I need to use some features to control traffic (QoS) not available with the regular kernel Regards, mb

2 1

centos4.2 -raid 1 and grub
by Jerry Geis 27 Nov '05

27 Nov '05

I had this little shell script. Use at your own risk... Jerry --------- # To load a blank drive with the old partition information use the command: # sfdisk /dev/hda < /etc/silentm/raidinfo.partitions.hda # The following command will setup the passed argument to boot in case the main disk is faulty # $1 is either /dev/hdb or /dev/sdb grub_setup_scsi() { grub << EOF find /boot/grub/stage1 device (hd0) /dev/sdb root (hd0,0) setup (hd0) EOF } grub_setup_ide() { grub << EOF find /boot/grub/stage1 device (hd0) /dev/hdb root (hd0,0) setup (hd0) EOF } DISK_TYPE=`fdisk -l /dev/hda | grep "/dev/hda"` if [ "$DISK_TYPE" = "" ] then echo "Assuming SCSI drives" sfdisk -d /dev/sda > /etc/silentm/raidinfo.partitions.sda sfdisk -d /dev/sdb > /etc/silentm/raidinfo.partitions.sdb grub_setup_scsi else echo "Assuming IDE drives" sfdisk -d /dev/hda > /etc/silentm/raidinfo.partitions.hda sfdisk -d /dev/hdb > /etc/silentm/raidinfo.partitions.hdb grub_setup_ide fi

1 0

OT - redirection script help
by Thomas E Dukes 27 Nov '05

27 Nov '05

Hello, Sorry to ask an off topic question but I know there are people on this list that can help or point me in the right direction. I will be moving a website in a few days and want to have it redirected to the new url. I know how to do this but I would like to have it have a countdown on the page. I know everyone has seen this before, does anyone know how to set this up? I have looked for this on hotscripts.com but I may not be looking in the right category. Any help would be appreciated.

1 1

Evolution Filters
by Lee W 27 Nov '05

27 Nov '05

Hi, I have just switched from using Thunderbird to Evolution on Centos 4.2 for my IMAP email and seem to be having problems with the filters. The filters I have created a really simple, such as checking for the [CentOS] in the subject line of this lists postings. The filters don't however get applied automatically, I have to select all the message in the inbox and manually select "Apply Filters" at which point all the messages get filtered correctly. I have made sure that the "Apply filters to new message in INBOX on this Server" is selected but it doesn't make any difference. Can someone advise if Evolution is supposed to work this way (as I have come to expect from Thunderbird) or is it correct that I have to apply the filters manually for IMAP folders. Thanks in advance Lee

2 2

Start scripts ordering
by Phil Savoie 27 Nov '05

27 Nov '05

Hi All, Was wondering if you could help me out. At work we are running CentOS 4.2 as our mailserver/firewall. The script in question is /etc/rc3.d/S08iptables. As this starts before the network adapters are set up, it hoses our iptables script resulting in denying our internal hosts internet access. What I would like to do is mv the script to /etc/rc3.d/S11iptables so the the interfaces are up and then the iptables script kicks in. It is in this way that we find it works best for us. However on a reboot we found that the "system" moved the script back to it's original name of /etc/rc3.d/S08iptables. How would I "tell the system" to keep the script named as /etc/rc3.d/S11iptables so that it does not automatically rename it back to /etc/rc3.d/S08iptables? Thank you in advance, Phil

4 4

Quota and Squirrelmail
by Alain Reguera 26 Nov '05

26 Nov '05

Hello, I have CentOS with quotas installed and make the users read their mails through squirrelmail. So I've been applied quotas in 2 partition, one for home, and other for /var/spool/mail/ directory. the problem is when the quota is over, the inbox in squirrelmail is locked so the user can't delete their mails and free the space. And all is frozen for them. Don't know If the question is in the right place. But could you suggest something. Any thing would be very appreciated. Thanks for your time.

3 3

webmaster in mqueue
by John Hinton 26 Nov '05

26 Nov '05

Anybody finding a huge mailfile named webmaster in their mqueue? I'm wondering what/how this gets there? Just found one on one server that was over 2 gigs! Yuck!!! So far, this has been on CentOS 3 machines using sendmail. Best, John Hinton

1 1

kernel load 40-50%?
by Alberto 26 Nov '05

26 Nov '05

Is there any command to view what kernel is doing when I have a high kernel/system load??

5 5

Re: [CentOS] A minor beef
by James B. Byrne 26 Nov '05

26 Nov '05

On Fri, 25 Nov 2005 21:30:26 -0500, "Bryan J. Smith" <thebs413(a)earthlink.net> wrote: > You were paying $1,000/year for RHEL ES??? As I recall it was in the $700.00 USD range which amounted to over $1,000.00 CAD per year at that time. And that was with the discount that RH gave to early adopters of their new support programs when we were forced off of our previous plan with the depreciation of RH9 (at ~$60.00 USD per host per year). However, I might be confusing the renewal price with what we actually paid while on RH support. In any case, the final cost in the second and subsequent years was over the magic three figure limit. Perhaps if you have hundreds or thousands of machines then RH might offer attractive bulk discounts, but for companies with just a handful of servers the cost of RH update support was made unbearable. I see now that there is a Basic level of support available from RH, which I do not recall as being offered originally. This presently costs $349.00 USD or about $410.00 CAD per host per year and I suppose that this service is equivalent to what CentOS provides in effect. To create a local mirror for RH updates without paying the per host fee is, of course, a violation of the contracted support agreement. One is compelled to place each host on contract, go without support for some hosts, or pay a tremendous sum of money for the privilege of maintaining a local mirror. To put RH pricing in perspective, I just installed a 2.8Ghz P4 server with a DVD DL super-mode burner, 250Gb of disc and 2Gb of RAM for less than $600.00 CAD. As I wrote earlier CentOS, at ~$25.00 (CAD) per host, is a bargain in comparison to RH. The minuscule degree of delay in updates between RH and CentOS hardly bears serious consideration in the matter. I suppose that if you have a government regulator on your back about some security issue or other then one might find a certain legal solace in pointing to a subscribed service as evidence of due diligence. Then again, our federal government requires us to use an application written only for the now depreciated and unavailable Microsoft JVM and supports it only under MS-IE 5.5 running on a pre- XP version of MS-Windows. So, go figure. (It runs, so far, under a fully patched MS-IE 6, but that is not the supported platform.) Regards, Jim -- *** e-mail is not a secure channel *** mailto:byrnejb.<token>@harte-lyne.ca James B. Byrne Harte & Lyne Limited vox: +1 905 561 1241 9 Brockley Drive fax: +1 905 561 0757 Hamilton, Ontario <token> = hal Canada L8E 3C3

1 0

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Discuss