Discuss April 2015

discuss@lists.centos.org

168 participants
186 discussions

iptables marking and NAT issue
by Boris Epstein 03 Apr '15

03 Apr '15

Hello all, It appears that, for some reason I have thus far failed to understand when you use marking in iptables you then run into troubles if you attempt to do NAT (MAQUERADE). Let me describe this in more detail. We are attempting to use a network test environment named ATCD running it on a CentOS VM under VirtualBox. For more into on ATCD see: https://github.com/facebook/augmented-traffic-control The networking inside the VirtualBox environment is private so at some point before you get out of it you've got to have a NAT router - not necessarily on the same VM where the ATCD runs - which also is a router. Be that as it may, ATCD uses a combination of iptables marking and tc to degrade/control network transmission quality in accordance with your settings. And it seems to work just fine up until you reach the NATing router - at which point the transmission drops to very slow if not non-existent. An old article here makes a passing reference to a conflict between iptables marking and MASQ (NAT): http://tldp.org/HOWTO/Adv-Routing-HOWTO/lartc.netfilter.html Unfortunately, the link to another text supposedly detailing how to deal with this is dead. Has anybody encountered this? Any tips on how to fix this issue? Thanks. Boris.

1 0

mounted NFS does not show in df -h
by Tim Dunphy 03 Apr '15

03 Apr '15

Hey guys, This is kind of odd, so I wanted to do a sanity check. I mounted an NFS share like so: [root@web1:~] #mount -t nfs nfs1.jokefire.com:/home /mnt/home Seemed to go ok. Then I took a look at the output of df -h and didn't see it! [root@web1:~] #df -h Filesystem Size Used Avail Use% Mounted on /dev/vda 40G 24G 14G 64% / devtmpfs 996M 0 996M 0% /dev tmpfs 1001M 0 1001M 0% /dev/shm tmpfs 1001M 101M 901M 11% /run tmpfs 1001M 0 1001M 0% /sys/fs/cgroup s3fs 256T 0 256T 0% /backup/cassandradb s3fs 256T 0 256T 0% /backup/mysql nfs1.jokefire.com:/var/www 20G 3.1G 16G 17% /var/www Yet, when I do a df -h on the directory I mounted the NFS share on, I see that it's mounted via NFS as expected: [root@web1:~] #df -h /mnt/home Filesystem Size Used Avail Use% Mounted on nfs1.jokefire.com:/home 20G 3.1G 16G 17% /mnt/home So, what do you think could be happening? Why is it that I can't see the output I'm expecting just by going df -h??? Thanks!! Tim -- GPG me!! gpg --keyserver pool.sks-keyservers.net --recv-keys F186197B

2 1

can't mount an LVM volume inCentos 5.10
by Dave Stevens 03 Apr '15

03 Apr '15

I have a degraded raid array (originally raid-10, now only two drives) that contains an LVM volume. I can see in the appended text that the Xen domains are there but I don't see how to mount them. No doubt this is just ignorance on my part but I wonder if anyone would care to direct me? I want to be able to retrieve dom-0 and one of the dom-Us to do data recovery, the others are of no interest. Suggestions? Will rtfm if directed. Dave excerpt from failed mount attempts --------------- starting here ------------ root@Microknoppix:/home/knoppix# vgchange -ay bulkley Volume group "bulkley" not found root@Microknoppix:/home/knoppix# vgchange -ay /dev/VolGroup00/bulkley Invalid volume group name: VolGroup00/bulkley Run `vgchange --help' for more information. root@Microknoppix:/home/knoppix# vgimport -f VolGroup00 Volume group "VolGroup00" is not exported root@Microknoppix:/home/knoppix# vgchange -ay VolGroup00 8 logical volume(s) in volume group "VolGroup00" now active root@Microknoppix:/home/knoppix# lvscan ACTIVE '/dev/VolGroup00/Dom0' [40.00 GiB] inherit ACTIVE '/dev/VolGroup00/babine' [100.00 GiB] inherit ACTIVE '/dev/VolGroup00/centos-template' [100.00 GiB] inherit ACTIVE '/dev/VolGroup00/bulkley-old' [100.00 GiB] inherit ACTIVE '/dev/VolGroup00/ubuntu' [10.00 GiB] inherit ACTIVE '/dev/VolGroup00/morice' [200.00 GiB] inherit ACTIVE '/dev/VolGroup00/oldserver' [80.00 GiB] inherit ACTIVE '/dev/VolGroup00/bulkley' [100.00 GiB] inherit root@Microknoppix:/home/knoppix# mount /dev/mapper/VolGroup00 aa mount: special device /dev/mapper/VolGroup00 does not exist root@Microknoppix:/home/knoppix# mount /dev/mapper/bulkley aa mount: special device /dev/mapper/bulkley does not exist root@Microknoppix:/home/knoppix# mount /dev/mapper/dev/VolGroup00 aa mount: special device /dev/mapper/dev/VolGroup00 does not exist root@Microknoppix:/home/knoppix# mount /dev/mapper/dev/VolGroup00/bulkley aa mount: special device /dev/mapper/dev/VolGroup00/bulkley does not exist root@Microknoppix:/home/knoppix# mount /dev/mapper/dev/VolGroup00/LogVol00 aa mount: special device /dev/mapper/dev/VolGroup00/LogVol00 does not exist root@Microknoppix:/home/knoppix# mount /dev/mapper/dev/VolGroup00/LogVol00/bulkley aa mount: special device /dev/mapper/dev/VolGroup00/LogVol00/bulkley does not exist root@Microknoppix:/home/knoppix# mount /dev/VolGroup00/LogVol00/bulkley aa mount: special device /dev/VolGroup00/LogVol00/bulkley does not exist -- "As long as politics is the shadow cast on society by big business, the attenuation of the shadow will not change the substance." -- John Dewey

2 1

CentOS mailing list for/in Vietnamese
by Nguyen Hung Vu 03 Apr '15

03 Apr '15

Hello all, Any Vietnamese are on this list? How about creating a dedicated mailing list, for Vietnamese, in Vietnamese[1] for the community? Looking the the list of existing mailing list, I see that there are some language-specific communities already there[2] Ref. [1] http://lists.centos.org/mailman/listinfo/centos-vi <http://lists.centos.org/mailman/listinfo/centos-vi> (not exists yet) [2] List of available mailing lists on CentOS: http://lists.centos.org/mailman/listinfo/

1 0

Kernel panic, CentOS 7.1503 fully updated, with executing gkrellm.
by Lamar Owen 03 Apr '15

03 Apr '15

Just a heads up, since I know gkrellm is in EPEL and not in the main CentOS repos. However, something fairly fundamental has changed, as prior to updating, gkrellm worked fine, but now every time I execute gkrellm the kernel panics. I'm going to triage on a different machine, as the panic corrupted at least one system library used by nautilus, but rpm -Va is your friend in these situations, and yum reinstall is its close kin. So I would prefer to duplicate on a burner machine, and I'll try to get a snapshot of the panic. If anyone else wants to try to reproduce, simply try installing the current gkrellm from EPEL on a fully updated CentOS 7 machine and see if it panics on you. I kindof hope it's local to my machine and its configuration (/home is LUKS, for instance), and it would be great if someone could verify that.

4 9

emailing plain text to exchange/outlook
by Les Mikesell 03 Apr '15

03 Apr '15

I know this isn't CentOS-specific, but it is probably a common problem - does anyone have a solution? If you mail something that is plain text from linux a recipient using outlook, it will remove line breaks more or less randomly. There is a way to tell outook to put them back as you read each message, but most people just think I sent it wrong. Is there something you can do to make a plain text list show up correctly short of converting it to html with <br>'s? -- Les Mikesell lesmikesell(a)gmail.com

3 2

Boot Failure [Solved]
by pro alias 02 Apr '15

02 Apr '15

I writ: "The latest kernel kernel-3.10.0.229.1.2.el7.x86_64 appears not to like our Supermicro X9SBAA-F board. It fails on boot with the last message before panic being apic_timer_interupt ...." <snipped rest> It turned out that only one X9SBAA box was affected. Deleting and reinstalling the kernel did not fix the issue. The problem turned out to be the old "initrd" issue in grub.conf. We have seen this from time to time on updates in the past. The line: initrd /initramfs-3.10.0-229.1.2.el7.x86_64.img was missing from grub.conf. When manually inserted, the new kernel worked swimmingly. I suspect that this is related to updating grub either immediately before or after a new kernel is installed. It's just enough of an annoyance that someone with spare time on their hands might want to look in it.

2 1

Boot Failure Lastest CEntos 7 Kernel on X9SBAA-F Board
by pro alias 02 Apr '15

02 Apr '15

The latest kernel kernel-3.10.0.229.1.2.el7.x86_64 appears not to like our Supermicro X9SBAA-F board. It fails on boot with the last message before panic being apic_timer_interupt . Previous kernels load very well. We erased and reinstalled the kernel, it verifies good as does everything else using "rpm- qaV". Anyone have any ideas?? Pretty hard to troubleshoot when you just get the last few lines on the screen.

2 1

OpenSSL vulnerability fix
by Venkateswara Rao Dokku 02 Apr '15

02 Apr '15

Hi All, I wanted to fix the openssl vulnerabilities (CVE-2014-3569, CVE-2014-3570, CVE-2014-3571, CVE-2014-3572) in my CentOS 5.5 and found out that 0.9.8zd has the fixes I am looking for (from the https://www.openssl.org/news/vulnerabilities.html link). But, When I tried to find the openssl-0.9.8zd rpm package, I did not find it in http://mirror.centos.org/centos/5/updates/x86_64/RPMS/. The latest that I could find was 0.9.8e-31-el5. Can you please help me on how can I find the rpm I am looking for or How can I fix the vulnerabilities.? Thanks for your help. -- Thanks & Regards, Venkateswara Rao Dokku.

3 4

low latency kernel?
by Les Mikesell 02 Apr '15

02 Apr '15

Someone recently posted on the x2go list that he had a problem with jerky videos playing remotely on Ubuntu, but solved it by installing a low latency kernel that was available as an alternative. That made me curious as to whether CentOS has an equivalent - or a way to build something similar. -- Les Mikesell lesmikesell(a)gmail.com

2 1

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 April 2015