Discuss February 2021

discuss@lists.centos.org

67 participants
48 discussions

Resizing a QEMU guest display
by Bill Gee 09 Feb '21

09 Feb '21

Hello everyone - I think I know the answer, but just in case ... I want to run it by the experts. I have a machines running CentOS 7. "uname -a" returns this: Linux practice7.billgee.local 3.10.0-1160.15.2.el7.x86_64 #1 SMP Wed Feb 3 15:06:38 UTC 2021 x86_64 x86 _64 x86_64 GNU/Linux This machine has an nVidia GS8400 and uses the nouveau driver. Its main job is to be a host for QEMU/KVM virtual machines. The QEMU packages on the host are: [bgee@practice7 ~]$ rpm -qa | grep qemu qemu-kvm-1.5.3-175.el7_9.3.x86_64 qemu-guest-agent-2.12.0-3.el7.x86_64 qemu-kvm-common-1.5.3-175.el7_9.3.x86_64 qemu-img-1.5.3-175.el7_9.3.x86_64 libvirt-daemon-driver-qemu-4.5.0-36.el7_9.3.x86_64 ipxe-roms-qemu-20180825-3.git133f4c.el7.noarch One of the guest VMs is Fedora 33. The guest comes up with a display resolution of 1024x768. I have not found a way to change it to something higher. If I go to the Settings - Display item in Fedora, there is a long list of resolutions. No matter what I choose, the system will switch to that resolution for about 1/2 second and then go back to 1024x768. This happens when using VNC to connect remotely and also when running on the host computer. Resizing the guest display window does nothing except make big black borders around the guest display. The guest video configuration in the virtual machine manager is set for "QXL". In my searching I found several suggestions to use "virtio" instead. When I make that change and start the guest, it immediately throws an error: "This QEMU does not support 'virtio' video device" I also tried "vmvga" and "cirrus". The cirrus setting does not resize and also loses coordination of the mouse pointer. The vmvga setting gives the same message about not being supported in this QEMU. I verified that the spice-vdagent package is installed on the guest VM. The virtual machine manager has "Resize guest with window" set to ON. Just to make life really interesting... Another guest VM on the same host is running CentOS7. That machine happily resizes! It is set for QXL in the virtual machine manager. I think I am just out of luck. I found some references which say that CentOS 7 does not have support for virtio displays. Is that right? Is there some way to resize the Fedora guest? I don't care about dynamic sizing by dragging a window border. If I can choose something and have it stick, I would be happy. Thanks! -- Bill Gee

2 5

Recommendation for 10 gigabit NICs on CentOS8
by Strahil Nikolov 09 Feb '21

09 Feb '21

Hi All, can you share what kind of old NICs do you use on CentOS 8 (Stream or not , it doesn't matter) without any issues? I was looking at ebay and I found some pretty old Mellanox "ConnectX" or "ConnectX-2" but I seriously doubt they will work on CentOS 8. Any proposals are also welcome. I don't care of the brand as long as it is PCIe and is supported by the vanilla kernel. Best Regards, Strahil Nikolov

4 3

firewalld - same source in different zones
by Łukasz Posadowski 08 Feb '21

08 Feb '21

Hi. I have a little trouble with firewalld. I'm trying to open some ports for monitoring server, but it's in the same network as "home" zone: Monitored host (192.168.111.60): lukasz @ strategie 17:52:19 ~ $ -> sudo firewall-cmd --get-active home sources: 192.168.111.0/24 (open ports 22, 80, 443) monitoring sources: 192.168.111.19 (open ports: 5666) public interfaces: ens18 (no open ports) --------------------------------------------------- Monitoring host (192.168.111.19): lukasz @ potemkin 17:57:25 ~ $ -> telnet strategie.ping.local 5666 Trying 192.168.111.60... telnet: connect to address 192.168.111.60: No route to host lukasz @ potemkin 17:57:26 ~ $ -> telnet strategie.ping.local 80 Trying 192.168.111.60... Connected to strategie.ping.local. Escape character is '^]'. ^] telnet> Connection closed. --------------------------------------------------- I think there are conflicting rules on a monitored host, that: - prevent access to 5666 from 192.168.111.0/24, - give access to 5666 from 192.168.111.19 and packets from potemkin are routed trough a home zone. I really would like to have dedicated "monitor" zone. Is there a way to give "monitor" zone more priority, than "home"? I may end with OpenVPN on potemkin and use 172.30.25.0/24 for monitoring, but, apart from encryption aspect, it seems a little excessive. Thank You. -- Łukasz Posadowski

2 2

Re: [CentOS] udev rules & USB devices - ignored at boot
by lejeczek 08 Feb '21

08 Feb '21

On 08/02/2021 08:10, Strahil Nikolov wrote: > Can you share your udev rules ? > > Best Regards, > Strahil Nikolov > > On Sun, Feb 7, 2021 at 15:33, lejeczek via CentOS > <centos(a)centos.org> wrote: > Hi guys, > > I have an Ethernet USB adapter for which udev executes my > custom rules but!... udev does it only at plug-in > event and > not! when the USB stays plugged in during a reboot, then > same rules are ignored (or some other problem results in > udev failure to do the same when device is plugged to a > running system). > My system is just an average box with a Ryzen on a B550 > platform. > Anybody sees this reproduce? > many thanks, L. > _______________________________________________ > CentOS mailing list > CentOS(a)centos.org <mailto:CentOS@centos.org> > https://lists.centos.org/mailman/listinfo/centos > <https://lists.centos.org/mailman/listinfo/centos> > simple rules: SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="00:1e:10:1f:00:00", ATTR{type}=="1", KERNEL=="eth*", NAME="lteusb" like I said - I get iface "lteusb" when I plug the USB so it works. If I leave that USB plugged in and reboot then no "lteusb". regards, L.

2 1

Challenging times in trying to access oracle Linux documentation
by Shamim Shahriar 07 Feb '21

07 Feb '21

Good afternoon everyone, hope you all are well. Apologies if this is considered unwelcome -- asking oracle question in centos group -- but a lot of centos Vs oracle is going on here so hoping this will not be shot down. I am unable to access any of the oracle documentation I need, as it keeps on asking for support identifier number, which, as far as I can tell, is only available if I purchase an expensive support subscription. Even obtaining a free oracle cloud account does not give you any access. I was trying to access oracle FreeIPA setup how to documentation, and even after many emails work oracle support over several weeks -- no resolution! They are not even clear as to what I need to do to obtain access -- all they are saying is I need a SI number!!! Has anyone experienced that or is it me having the issue? If I have to pay subscription, why not continue with RHEL subscription anyway? Would appreciate if someone could please confirm. Best regards Shamim

9 15

udev rules & USB devices - ignored at boot
by lejeczek 07 Feb '21

07 Feb '21

Hi guys, I have an Ethernet USB adapter for which udev executes my custom rules but!... udev does it only at plug-in event and not! when the USB stays plugged in during a reboot, then same rules are ignored (or some other problem results in udev failure to do the same when device is plugged to a running system). My system is just an average box with a Ryzen on a B550 platform. Anybody sees this reproduce? many thanks, L.

1 0

letsencrypt error
by Jerry Geis 06 Feb '21

06 Feb '21

Hi all - So I just ran into the changes lately from letsencrpt. certbot-auto is no longer available. I added this to httpd.conf <VirtualHost *:80> ServerName mydomain </VirtualHost> service httpd restart When I do "certbot -d mydomain" I get this : Domain: mydomain Type: unauthorized Detail: Invalid response from http://mydomain/.well-known/acme-challenge/i_fU1bFrQZzgfVI2FtWo8Ov0ITjplCcP… [97.107.162.8]: "<!DOCTYPE HTML PUBLIC \"-//IETF//DTD HTML 2.0//EN\">\n<html><head>\n<title>403 Forbidden</title>\n</head><body>\n<h1>Forbidden</h1>\n<p" What did I miss ? Thanks Jerry

4 8

msmilter doesn t start / Can't open PID file /run/MSMilter.pid
by Ralf Prengel 05 Feb '21

05 Feb '21

Hallo, msmilter doesn t start. Has anyone a hint for me? I used this tutorial. /run exists and selinux is disabled. https://www.awsmonster.com/2020/03/how-to-install-mailscanner-stable-v521.h… Thanks for a hint. Ralf -- Unit msmilter.service has begun starting up. Feb 05 15:09:14 vader2 msmilter-init[1001]: Edit the file /etc/MailScann-- Unit msmilter.service has begun starting up. Feb 05 15:09:14 vader2 msmilter-init[1001]: Edit the file /etc/MailScanner/MailScanner.conf according to Feb 05 15:09:14 vader2 msmilter-init[1001]: your needs. When complete, edit /etc/MailScanner/defaults and set Feb 05 15:09:14 vader2 msmilter-init[1001]: the variable to enable MailScanner to run: Feb 05 15:09:14 vader2 msmilter-init[1001]: run_mailscanner=1 Feb 05 15:09:14 vader2 systemd[1]: (yet?) after start: No such file or directory Feb 05 15:09:14 vader2 systemd[1]: Failed to start LSB: MSMilter daemon. -- Subject: Unit msmilter.service has failed -- Defined-By: systemd -- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel er/MailScanner.conf according to Feb 05 15:09:14 vader2 msmilter-init[1001]: your needs. When complete, edit /etc/MailScanner/defaults and set Feb 05 15:09:14 vader2 msmilter-init[1001]: the variable to enable MailScanner to run: Feb 05 15:09:14 vader2 msmilter-init[1001]: run_mailscanner=1 Feb 05 15:09:14 vader2 systemd[1]: Can't open PID file /run/MSMilter.pid (yet?) after start: No such file or directory Feb 05 15:09:14 vader2 systemd[1]: Failed to start LSB: MSMilter daemon. -- Subject: Unit msmilter.service has failed -- Defined-By: systemd -- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel

2 1

Which bug tracker should I use to report things to CentOS ? (recap)
by Fabian Arrotin 05 Feb '21

05 Feb '21

With some moving parts within the CentOS Project, some people are now confused about where to report things, as https://bugs.centos.org isn't the only bugs/tickets tracker, so let's just have a quick recap so that people aren't confused anymore ? :) # CentOS Stream (all kind of tickets) see https://wiki.centos.org/FAQ/CentOSStream#Where_do_I_report_bugs.3F So basically https://bugzilla.redhat.com/enter_bug.cgi?product=Red%20Hat%20Enterprise%20… # CentOS Infra, CBS, SIGs, mirror, CI, etc see previous mail : https://lists.centos.org/pipermail/centos-devel/2020-August/055980.html # All the rest (CentOS Linux pkgs, etc) https://bugs.centos.org Hope that it helps , as we have people asking on irc the status of a ticket that was filed in wrong tracker (and so also not correct category as for example infra categories were removed/archived on https://bugs.centos.org) and not reviewed. -- Fabian Arrotin The CentOS Project | https://www.centos.org gpg key: 17F3B7A1 | twitter: @arrfab

1 0

Filesystem choice for BackupPC extrenal drive
by Kenneth Porter 05 Feb '21

05 Feb '21

I'm setting up a CentOS 7 box as a BackupPC 4 server to back up Windows boxes on my LAN. I'm using an external 1.5 TB USB drive for the "pool". BackupPC deduplicates by saving all files in a pool, a directory hiearchy with each file named for the checksum of the file, and the directories acting as a hash tree to reach each pool file. A backup for a specific workstation is a directory tree of checksums and metadata that point into the pool for the actual file data. Incremental backups are reverse deltas from periodic "filled" backups of all files. I'm using rsyncd to pull changed files from the workstations. I'm deciding which filesystem to use for my external drive. I'm thinking the main candidates are ext4 and xfs. What's the best filesystem for this application? <https://github.com/backuppc/backuppc> Repo for CentOS 7 users: <https://copr.fedorainfracloud.org/coprs/hobbes1069/BackupPC/>

3 2

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Discuss February 2021