Discuss

discuss@lists.centos.org

34729 discussions

Version of curl does not support session resumption
by Joscha Knobloch 28 Oct '21

28 Oct '21

Hi, we have a CentOS7 Server running that pushes some Backups via FTP onto another server. We switches the FTP-Server for a new one that works very well. One new thing is that FTPS session resumption is now mandetory and cannot be turned off since it is an important security feature. Unfortunately the version of curl on the system does not yet support it. This is the version installed on the system: curl 7.29.0 (x86_64-redhat-linux-gnu) libcurl/7.29.0 NSS/3.53.1 zlib/1.2.7 libidn/1.28 libssh2/1.8.0 I have discussed this on the curl mailing list and they said the version is from 2013 and we should update to a new version of curl. Is curl going to be updated to support ftps session resumption in the near future? If not: What would be the best way to get a newer version onto the system? Kind Regards Joscha

3 2

grepmail
by Jon LaBadie 28 Oct '21

28 Oct '21

On my Fedora box I occasionally use grepmail. Recently I went to use it on my CentOS 7 mail server and found I had not installed it. Surprisingly (to me) I did not find it in the repositories. Grepmail is a perl script and uses several perl modules. I copied the executable from the Fedora box and installed a missing module from the CentOS 7 repos. Seems to work. However the CentOS perl modules are older versions than the executable uses on Fedora. I'd prefer the executable and modules were tested together. Have I missed grepmail in any CentOS 7 repos? -- Jon H. LaBadie jcu(a)labadie.us

2 1

IPv6 mailing list?
by Kenneth Porter 27 Oct '21

27 Oct '21

Can anyone recommend an end-user IPv6 mailing list? (A web forum would also be acceptable.) I've been looking at available lists and they all seem targeted at backbone players and ISPs. I'm looking for something where we can report and resolve problems with our ISPs. For example, I just got an AT&T business connection and the Edgemark fiber gateway doesn't provide RA or prefix delegation. It assumes the customer equipment is all leaf nodes that are statically-configured. It doesn't recognize RA from the customer, either. So I'm using ndppd (added to EPEL7 this morning!) to proxy neighbor announcements through my CentOS7 gateway/firewall. I have a backup/secondary C7 gateway and it got confused when the primary sent RA for the LAN-side subnet upstream to the common WAN link (via the radvd package) and the secondary added a default routing table entry pointing to the primary gateway instead of using its own statically-configured default gateway setting. That was a head-scratcher until I noticed my firewall logs on the main gateway showing dropped DNS packets from the secondary that should have been going to the ISP gateway. IPv6 DNS was failing on the secondary with timeouts (10 seconds!) and I couldn't figure out what was eating the packets. So I'm wondering how multiple gateways sharing a link are supposed to cooperate and inform each other without confusing each other about the desired topology.

2 2

anyone her using nexus 3 as a repo-mirror for centos
by Ralf Prengel 27 Oct '21

27 Oct '21

Hallo anyone her using nexus 3 as a repo-mirror for centos. My problem: yum update ends with error 404 and I ve no idea what is going wrong. The nexus repo is empty so it is clear why 404 appears but I thought that the nexus fetches the files immdeatly from outside. Thanks for hints Ralf

1 0

ifconfig not showing virtual IP when using NetworkManager
by Jerry Geis 26 Oct '21

26 Oct '21

When using NetworkManager instead of the ifcfg-eth1/eth0 files - I make a nmcli connection and modify "Wired connection 1" +ipv4.addr IP/22 kind of command - it works - which is great. It "adds" the virtual IP. However - when doing ifconfig that "virtual" IP does not show up. When using NetworkManager - to add the virtual IP - how do I get that to show using ifconfig also ? it used to show as eth1:0 Thanks, jerry

2 1

Intel Sunrise Point-LP HD Audio doesn't work past kernel 4.18-0-240.22.1
by Frank Cox 26 Oct '21

26 Oct '21

As stated in the subject line, the last Centos kernel where Intel Sunrise Point-LP HD Audio works is version 4.18-0-240.22.1 Booting the computer with any later kernel makes the audio device not show up and the only audio device is "dummy output". -- MELVILLE THEATRE ~ Real D 3D Digital Cinema ~ www.melvilletheatre.com

1 0

libxml2 update/packaging issues?
by Steve Meier 25 Oct '21

25 Oct '21

Hello all, last Friday (Oct 15) I encountered a weird issue relating to the libxml2* packages. I have a script which monitors the CentOS mirrors to find new packages. On Friday, these showed up: libxml2-2.9.1-6.el7.5.i686.rpm libxml2-2.9.1-6.el7.5.x86_64.rpm libxml2-devel-2.9.1-6.el7.5.i686.rpm libxml2-devel-2.9.1-6.el7.5.x86_64.rpm libxml2-python-2.9.1-6.el7.5.x86_64.rpm libxml2-static-2.9.1-6.el7.5.i686.rpm libxml2-static-2.9.1-6.el7.5.x86_64.rpm However, looking at Scientific Linux (https://scientificlinux.org/category/sl-errata/slsa-20213810-1/) and Red Hat Errata (https://access.redhat.com/errata/RHSA-2021:3810) it seems that CentOS is one release behind (el7.5 from CentOS vs. el7.6 for SL and RHEL). Looking at my systems which have libxml2 packages installed, it gets even weirder: $ rpm -q --queryformat '%{VERSION}-%{RELEASE}\t%{BUILDTIME}\t%{INSTALLTIME}\n' libxml2 2.9.1-6.el7.5 1601484919 1605352337 My system already has this exact release. $ rpm -qi libxml2 Name : libxml2 Version : 2.9.1 Release : 6.el7.5 Architecture: x86_64 Install Date: Sa 14 Nov 2020 12:12:17 CET Signature : RSA/SHA256, Mi 14 Okt 2020 20:55:29 CEST, Key ID 24c6a8a7f4a80eb5 Source RPM : libxml2-2.9.1-6.el7.5.src.rpm Build Date : Mi 30 Sep 2020 18:55:19 CEST The version which is now on the mirrors has a newer build date (Oct 14, 2021): $ rpm -qp --queryformat '%{VERSION}-%{RELEASE}\t%{BUILDTIME}\t%{INSTALLTIME}\n' http://mirror.centos.org/centos/7/updates/x86_64/Packages/libxml2-2.9.1-6.e… 2.9.1-6.el7.5 1634214499 (none) But its latest changelog entry does not match the latest changelog entry in the CentOS Git here: https://git.centos.org/rpms/libxml2/c/7219ebb7c53b80409f3f8fa3433d3ecf35665… So, to make a long story short: I believe there should be a el7.6 version of the libxml2* packages but instead we somehow have a repackaged el7.5 which is not up-to-date. Kind regards, Steve

3 2

Unexpected /etc/resolv.conf updates on CentOS 7
by Toralf Lund 22 Oct '21

22 Oct '21

Hi Does here anyone know exactly when NetworkManager creates or is supposed to create /etc/resolv.conf for a network connection? Is there a way I can control it, or alternatively, is there a good way to debug the functionality? I thought that there would simply be an update whenever a connection was established, and an addresses/network info was received (if using DHCP), and that the information would pretty much be left alone after that. However, I've lately found that a new file gets written every few hours even though there is no connection change (that I can detect), i.e. the same link is up all along. Does anyone have any ideas why that might happen? I thought at first that the updates could be linked to DHCP lease renewal, but on closer inspection, that does not seem to be the case. I don't see anything in the system log related to networking at the points where a new file gets written. I get the above behaviour for my home Wifi net. It seemed to start after I switched to a new router, but that might be coincidental. It's a problem for me because I'm also using "commercial" VPN software (not integrated with NetworkManager) that will create it's own resolv.conf file; it replaces data e.g. from Network Manager when VPN is enabled, and restores it on disable. If NetworkManager "refreshes" the information in the mean time, the DNS config for VPN is lost, and the link doesn't work as expected... This is on a CentOS 7 system with all the latest updates. - Toralf

4 6

IPv6 routing preference with multiple default gateways
by Kenneth Porter 17 Oct '21

17 Oct '21

I've got a primary and backup CentOS 7 gateway, each with two interfaces, connected to my LAN and my fiber gateway. The default gateway (the fiber box) is set explicitly in ifcfg-eno2 with IPV6_DEFAULTGW. Using "ip -6 route show" I see two defaults, the static one and the one advertised from my other CentOS7 box via the internal interface (eno1), both with metric 1024 and preference medium. default via 2001:1890:1837:5b00::1 dev eno2 metric 1024 pref medium default via fe80::100 dev eno1 proto ra metric 1024 expires 293sec hoplimit 64 pref medium How do I set the IPv6 router advertisements on the LAN side to advertise a higher metric or lower preference so there's never ambiguity that the fiber box has precedence? I can't find a setting in the man page for radvd to set the metric or preference for the LAN route that's announced. Additionally, I think I need to announce a route on the WAN interface (eno2) of lower precedence than the static route so the fiber box knows that the two CentOS7 boxes are proxies for my LAN, but the two boxes don't try to route to each other? This is for C8 but it's useful for the concepts, like that a lower metric has a higher precedence: <https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux/8/ht…> Some radvd examples that lack an example of how to set metric or preference. <https://tldp.org/HOWTO/Linux+IPv6-HOWTO/ch22s04.html>

1 1

livestatus.o addon on Nagios 4.4.6 on CentOS Stream 8 OS
by Kaushal Shriyan 15 Oct '21

15 Oct '21

Hi, I am running Nagios Core 4.4.6 ( https://www.nagios.org/downloads/nagios-core/thanks/?product_download=nagio…) on CentOS Stream release 8. Please refer to the below details about nagios and thruk web interface https://www.thruk.org/ #rpm -qa | grep labs-consol labs-consol-stable-1.4-1.noarch #rpm -qa | grep thruk thruk-base-2.44.3-11112.1.x86_64 thruk-2.44.3-11112.1.x86_64 thruk-plugin-reporting-2.44.3-11112.1.x86_64 libthruk-2.44.2-0.x86_64 #rpm -qa | grep epel epel-release-8-11.el8.noarch epel-next-release-8-11.el8.noarch # #/usr/local/nagios/bin/nagios -v /usr/local/nagios/etc/nagios.cfg Nagios Core 4.4.6 Copyright (c) 2009-present Nagios Core Development Team and Community Contributors Copyright (c) 1999-2009 Ethan Galstad Last Modified: 2020-04-28 License: GPL Website: https://www.nagios.org Reading configuration data... Read main config file okay... Read object config files okay... Running pre-flight check on configuration data... Checking objects... Checked 16 services. Checked 2 hosts. Checked 1 host groups. Checked 0 service groups. Checked 1 contacts. Checked 1 contact groups. Checked 24 commands. Checked 5 time periods. Checked 0 host escalations. Checked 0 service escalations. Checking for circular paths... Checked 2 hosts Checked 0 service dependencies Checked 0 host dependencies Checked 5 timeperiods Checking global event handlers... Checking obsessive compulsive processor commands... Checking misc settings... Total Warnings: 0 Total Errors: 0 Things look okay - No serious problems were detected during the pre-flight check I am following https://www.thruk.org/documentation/install.html to install Livestatus addon for Nagios which provides real time status data. Is there a way to enable the below entry in /usr/local/nagios/etc/nagios.cfg file. I did search using yum and dnf package manager on CentOS Stream release 8 but unable to locate livestatus.o addon or install it. broker_module=/opt/local/livestatus/livestatus.o /tmp/live.sock Any help will be highly appreciable. Please guide/suggest. Thanks in advance. Best Regards, Kaushal

2 2

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Discuss