similar to: Applications staying open in Gnome

I'm in the process of rolling out new setups with dovecot on CentOS 5.2 and I notice that dovecot doesn't handle the brute-force attacks too nice. I reduced the limit a bit to some reasonable looking value: login_max_processes_count = 32 to stop them earlier and the number of processes stops at that figure when an attack happens. However, it stays at this count for hours although the

What's the best way to do this? Remove the yum group gnome (does this set default init to 3?), install the yum group KDE and then set default init to 5 again (as I assume installing KDE not during the setup will not automatically set the system to start the gui?). Kai -- Kai Sch?tzl, Berlin, Germany Get your web at Conactive Internet Services: http://www.conactive.com

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I've used denyhosts. If you do have an issue with fail2ban, it does pretty much the same thing. Andy - -------- Original Message -------- Subject: Re: [CentOS] fail2ban needs shorewall? Date: Wed, 23 Jul 2008 17:08:07 +0200 From: Kai Schaetzl <maillists at conactive.com> Reply-To: CentOS mailing list <centos at centos.org> To:

Is there a program on CentOS which will sort my passwd/group etc. files by username or uid/gid? On Suse I can do this with a sysconfig switch and Suseconfig, is there something similar on RH-based systems? Kai -- Kai Sch?tzl, Berlin, Germany Get your web at Conactive Internet Services: http://www.conactive.com

I thought I post this link http://monty-says.blogspot.com/2009/12/help-saving-mysql.html in case anyone isn't aware of this yet and wants to email the EC. Kai -- Kai Sch?tzl, Berlin, Germany Get your web at Conactive Internet Services: http://www.conactive.com

My update to 5.2 fails on one machine because of this error: Transaction Check Error: file /etc/init.d from install of chkconfig-1.3.30.1-2 conflicts with file from package AcronisAgentLinux-9.5.8076-1 file /etc/init.d from install of chkconfig-1.3.30.1-2 conflicts with file from package TrueImageAgent-9.5.8076-1 Obviously, both packages created (completely unnecessarily) the /etc/init.d

Using CentOS 5.1 with Xen 3.2. I'm trying to install a fully virtualized system for the first time and doing this with an old Suse 9.0 (this probably would run paravirtualized, maybe it won't run, anyway, because of hardware problems). I start with virt-manager and after choosing "fully virtualized" it presents me two options for install media: iso image and CD-ROM or DVD,

I have a local mirror of CentOS5 base and updates. Now with 5 having become 5.1 and in case I just want to keep on updating (at the moment) is it enough that I update the mirror of updates or do I also need an update of the base repo? Kai -- Kai Sch?tzl, Berlin, Germany Get your web at Conactive Internet Services: http://www.conactive.com

The new virtual machine manager in Centos 5.1 displays some weird behavior here and I assume it's because the UUID of one running VM is the same as of a VM that is shut down. I used the same config file template for creating all my test VMs, so several have the same UUID. Didn't produce any problems with xm. But the new VM manager in Gnome shows active and inactive VMs and identifies

I took over a custom firewall script from my older Suse machines to my Dom-Us and it works just fine. Doing the same for Dom-0 immediately killed all traffic for the VMs. As there was no need before I had been dropping everything on the FORWARD chain. After ACCEPTing all for FORWARD my VMs are happy again. What's best practice on Dom-0, what do you do? Can I restrict the forwarding, in

This is my first major version update after using a local repo. It seems that before I can "yum update" I have to fetch the complete new 5.2 base? Is this correct? When I run yum update it now fails with various strange dependency errors although all the updates are synced. When I change to using the original repo file it wants to download many many more and shows a lot of files that

I converted a physical maschine running CentOS 4.6 to a Xen VM on a CentOS 5 host. Almost everything was running when I decided to rsync /lib and /usr a while later in order to get updated files and get the one or two non-working programs to work. There are differences because of the xenU kernel I didn't think about. I forgot to exclude rpm database and /lib/modules. Because the synced

Darrell Dominey wrote on Thu, 10 Mar 2005 15:31:04 -0330: > make sure that your mountpoint does not have a .ogg extention with the > mp3 stream, otherwise, it will try to play using the ogg plugin, and > will stay at 0% buffer > Hi, thanks for the quick answer. We are serving the mp3s with .nsv endings since that is what the oddcast plugin seems to send by default when mp3 is

Hi Guys, OK, I figured out the problem, It would seem that Comcast spelling my DNS entry wrong would do it! I have done this a zillion times, I was totally stumped as to what I would be missing this time. Thanks everyone for their help. -Jason ----- Original Message ----- From: "Kai Schaetzl" <maillists at conactive.com> To: centos at centos.org Sent: Tuesday, December 22,

make sure that your mountpoint does not have a .ogg extention with the mp3 stream, otherwise, it will try to play using the ogg plugin, and will stay at 0% buffer On Thu, 10 Mar 2005 19:45:09 +0100, Kai Schaetzl <maillists@conactive.com> wrote: > I'm new to streaming or music hearing over the wire, I still prefer my > old-fashioned stereo equipment :-) > But a friend asked me

really, it should have no extention, like /stream/1001 for the mountpoint On Thu, 10 Mar 2005 20:51:29 +0100, Kai Schaetzl <maillists@conactive.com> wrote: > Darrell Dominey wrote on Thu, 10 Mar 2005 15:31:04 -0330: > > > make sure that your mountpoint does not have a .ogg extention with the > > mp3 stream, otherwise, it will try to play using the ogg plugin, and >

What does the plus sign after the blocks value exactly mean in the fdisk output? Some research reveals that it indicates that not all the blocks are included in the fdisk value. But what does this exactly mean? Kai -- Kai Sch?tzl, Berlin, Germany Get your web at Conactive Internet Services: http://www.conactive.com

The installed php on CentOS/RHEL is the CGI version. This means I should be able to get command line arguments with $GLOBALS[''argv''] or $_SERVER[''argv'']. That''s how the php documentation tells and how it works f.i. on Suse. However, on CentOS I get them only when I use the -n switch to php (no php.ini). The problem with -n is that php cannot find

Is anyone using successfully a specific rpm of Dovecot 1.0.something? There's nothing on dag and a few on rpmfind (for Fedora Core), but I don't know if they'd work. The one coming with Centos 4 (0.99.11) has quite a few bugs and I'd like to put an upgrade on top of it. Kai -- Kai Sch?tzl, Berlin, Germany Get your web at Conactive Internet Services: http://www.conactive.com

I've been getting over the last months several of these notices. Sometimes a few per day. What's the problem? Can't this be avoided? This is the mail system at host mail.centos.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