Discuss August 2012

discuss@lists.centos.org

190 participants
167 discussions

default gateway outside of the LAN
by Boris Epstein 20 Aug '12

20 Aug '12

Hello all, We have a somewhat unique setup whereby our default router is outside of the local network. Let us say our network is 192.168.10.0/255.255.255.0 So we have a route-eth0 file that looks something like this: 10.1.1.1 via 192.168.10.1 dev eth0 default 10.1.1.1 dev eth0 The last definition would simply not take - but it has to for the setup to work. And no, 192.168.10.1 does not operate as a full-fledge router due to our setup. When we attempt to add it manually we get the following error: RTNETLINK answers: No such process When we use ifup we get the following: Error: either "to" is a duplicate, or "10.1.1.1" is a garbage. A little discussion of this can be found here: http://www.centos.org/docs/5/html/5.1/Deployment_Guide/s1-networkscripts-st… and here: http://www.linuxquestions.org/questions/linux-networking-3/etc-init-d-netwo… It does not help us, however. Any tips much appreciated. Cheers, Boris.

4 5

better securing files on files servers
by Gregory Machin 20 Aug '12

20 Aug '12

Hi. I have a couple of questions : I have inherited a file server that provides Samba and NFS file shares. We use a combination on file system acls and posix permissions . I'm looking to better secure access to the files by trimming some of the permissions etc. 1) What services could break if I change the umask from 022 to 007 thus by default only user and group have rw access to the files and directories ignoring facls ? 2) I have found that using samba and NFS to share the same file system are not so grate as when some windows file permissions are set this cause the file permissions to be more open eg permissions for other get set to rwx, when this is not wanted as over NFS this grants everyone access to the files. Is there a workaround for this ? 3) are there any best practices for configuring file system acls, that will ensure better data security , other then the obvious limiting of write and execute permissions. Any other recommendations around using NFS and Samba together ? Thanks G

1 0

LVM overhead? Does it cripple I/O?
by Smithies, Russell 19 Aug '12

19 Aug '12

For a high-performance system (64-cores, 512GB RAM, 5TB local disk, 110TB NFS-mounted storage) is there any advantage of dropping lvm and mounting partitions directly? We're not planning on changing partition sizes, but if we did we'd probably do a full rebuild. Has anyone done performance testing to show that lvm isn't crippling I/O? Thanx, Russell ======================================================================= Attention: The information contained in this message and/or attachments from AgResearch Limited is intended only for the persons or entities to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipients is prohibited by AgResearch Limited. If you have received this message in error, please notify the sender immediately. =======================================================================

3 2

CentOS with sssd and samba4
by Steve Thompson 18 Aug '12

18 Aug '12

If you are authenticating your CentOS 6 systems using sssd from a samba4 DC using GSSAPI, I'd like to hear from you. I have been able to get it to work using only cleartext passwords in sssd.conf, and of course I'd prefer to use GSSAPI. Steve

1 1

DNS DoS attack
by Jussi Hirvi 17 Aug '12

17 Aug '12

Looks like one of my name servers (CentOS 5) gets a lot of malicious queries. The cpu load is constantly about 3 %. I put on stricter limits on who is allowed recursive queries, but this does not affect the CPU load. I also updated bind. I temporarily turned on querylog (command: rndc querylog), and noticed that I get over 200 queries like this per second: > Aug 17 07:41:38 mx2 named[6873]: client 205.145.64.200#53: query (cache) 'ripe.net/ANY/IN' denied > Aug 17 07:41:38 mx2 named[6873]: client 204.10.45.5#53: query (cache) 'ripe.net/ANY/IN' denied > Aug 17 07:41:38 mx2 named[6873]: client 78.40.35.212#53: query (cache) 'ripe.net/ANY/IN' denied > Aug 17 07:41:38 mx2 named[6873]: client 207.207.3.126#53: query (cache) 'ripe.net/ANY/IN' denied Are there any ways to mitigate this, or do I just have to wait? - Jussi

5 6

stupid bash question
by Craig White 17 Aug '12

17 Aug '12

the relevant snippet is... NAME="*.mov" cd $IN if test -n "$(find . -maxdepth 1 -name $NAME -print -quit)" and if there is one file in this directory - ie test.mov, this works fine but if there are two (or more) files in this directory - test.mov, test2.mov then I get an error... find: paths must precede expression So my code is evidently wrong. I just want a test for 1 or more files ending in .mov in the directory? Any one want to toss me a bone here? -- Craig White ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ craig.white(a)ttiltd.com 1.800.869.6908 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ www.ttiassessments.com Need help communicating between generations at work to achieve your desired success? Let us help!

8 11

Software RAID woes due to GRUB
by SilverTip257 16 Aug '12

16 Aug '12

Hello, I recently cloned an old install from one server chassis to another via rsync. The old setup only had one disk and the new one is set up with two disks in a softraid. I managed to solve the problem and thought I'd present it here to benefit others. attempting to add sda3 to md2 says: mdadm: hot add failed for /dev/sda3: Invalid argument dmesg says: md: could not bd_claim sda3. md: error, md_import_device() returned -16 Drive is not mounted or in use by the OS. SMART error logs do not indicate any disk failures. I can boot up to a rescue CD and the array assembles just fine. Go figure the old install used partition labels to determine the boot device. So GRUB grabbed the first disk /dev/sda, on which root was label "/" or device /dev/sda3. So GRUB kept /dev/sda3 busy so that mdadm couldn't add it to the array. Changing the root= device kernel param in grub.conf to the raid device md2 and re-rolling the initial ramdisk cleared up the problem. It boots fine and assembles a healthy softraid now. Maybe it's a stupid item to overlook, but it certainly gave me a head scratcher for a bit. Have a laugh ... I certainly did when things were up and running properly again. ---~~.~~--- Mike // SilverTip257 //

1 0

libfget ported to CentOS 6.2?
by drichards＠globalcerts.net 16 Aug '12

16 Aug '12

Hi, I have used the library calls of an open source library called libfget, which I had running on Fedora 8. I am now trying to build it and run it on CentOS 6.2, but there are some problems. The original maintainer of that package is not currently available and I tried to post to that list but no one answered. I wonder if anyone here has needed to use this and had success in building and running it. There are bugs in the socket programming (getsock* calls and select()). I replaced gethostbyname_r with getaddrinfo, but I can't figure out why a select() call is failing. I am using the lastest version: fget-1.3.3.tar.gz (libfget.so). Thanks, Dirce

1 0

Re: [CentOS] slowness with new kernel+nouveau?
by Marko Vojinovic 16 Aug '12

16 Aug '12

On Thursday, 16. August 2012. 16.34.01 Sorin Srbu wrote: > > -----Original Message----- > > From: centos-bounces(a)centos.org [mailto:centos-bounces@centos.org] On > > Behalf Of m.roth(a)5-cent.us > > Sent: den 16 augusti 2012 16:13 > > To: CentOS mailing list > > Subject: Re: [CentOS] slowness with new kernel+nouveau? > > > > >> I hear rumors that kmod-nvidia will be coming in RHEL soon, and so here > > >> to CentOS. > > > > > > Any source for the rumors? > > > > > > It'd be great though! > > > > It's in fc17. No it isn't. As long as nVidia drivers are closed source and proprietary, they will never be in Fedora. I wouldn't know about RHEL, though, they might decide whatever they want to do. However, some main Nouveau developers are employed by RH, so I wouldn't bet on RHEL importing nVidia drivers. HTH, :-) Marko

1 0

Re: [CentOS] slowness with new kernel+nouveau?
by m.roth＠5-cent.us 16 Aug '12

16 Aug '12

Sorin Srbu wrote: >> -----Original Message----- >> From: centos-bounces(a)centos.org [mailto:centos-bounces@centos.org] On >> Behalf Of m.roth(a)5-cent.us >> Sent: den 16 augusti 2012 15:46 >> To: CentOS mailing list >> Subject: Re: [CentOS] slowness with new kernel+nouveau? >> >> I hear rumors that kmod-nvidia will be coming in RHEL soon, and so here >> to CentOS. > > Any source for the rumors? > > It'd be great though! It's in fc17. mark

3 2

← Newer
1
...
5
6
7
8
9
10
11
...
17
Older →

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 August 2012