Search
388 results for “NetworkManager”
-
I digged an old Thinkpad T410 out of my pile of Thinkpads and decided to take it to the arcade as a general use laptop for the days I don't have anything more modern on me and something needs tweaking.
This specific machine had only 6 gigs of RAM, so I tried something new here. Normally I do these light but up-to-date setups with Debian and XFCE or IceWM, but since I've put Alpine into quite a few Docker containers, I wanted to see how it fares as a full installation.
For someone who has used Linux in the days when X configuration was done by hand, setting it up wasn't that hard, but there was also plenty of manual tweaking to get this thing behave somewhat like a modern era laptop would. I put XFCE on it, changed the networking to run under NetworkManager and so forth. Been quite a while when I've set something up from as scratch as this, but it was also a nice walk down the memory lane.
Anyway, the Thinkpad works great, Wifi and all. The whole installation with all the junk I need takes 1.4 gigabytes of disk space and 480 megabytes of RAM with full DE running. Pretty!
Isn't this thing glorious? ThinkLight and all!
-
I just tried multipath TCP on Linux and was surprised how easy it was. The idea was to be able to plug my phone into my Laptop and momentarily boost my download speeds. Because few content servers natively support MPTCP, I used a shadowsocks proxy to tunnel to a VPS. Other than that, all I had to was plug in my phone via USB, enable tethering and disable USB debugging. The necessary kernel options were already set and NetworkManager automatically configured the subflows.
#mptcp #networking -
Hier mal wieder ein mehr oder weniger sinnvoller Distro-Vergleich: Wie viel #RAM benötigen Distros direkt nach dem Start?
Es laufen: #sway, foot und htop (zum messen). Ich habe die Distros schon so optimiert, dass möglichst wenig Hintergrunddienste und überall möglichst die gleichen Dienste laufen (NetworkManager, CUPS, sshd)#Void Linux: 288M
#antiX 26 (mit dinit): 294M
#Debian 13: 341M
#openSuse #Tumbleweed: 355M
#FunOS (Ubuntu) 25.10: 395M
#Fedora 43: 415M -
#Linux Weekly Roundup for February 15th, 2026: #Mesa 26.0, #Ubuntu 24.04.4 LTS, #KDE Frameworks 6.23, #GNOME 49.4, #pearOS 26.2, #IPFire DBL, #KaOS 2026.02, #NetworkManager 1.56, #GNU Linux-libre 6.19, #OpenVPN 2.7, #Parrot 7.1, #Tails 7.4.2, GNOME 50 beta, #REMnux 8, #MythTV 36.0, GNU Binutils 2.46, #Vim 9.2, #GitHub Tray, and more https://9to5linux.com/9to5linux-weekly-roundup-february-15th-2026
-
It's always DNS, right? If someone experiencing some strange issues with ubuntu, maybe this toot is for you.
tl;dr: switch from systemd-resolved to resolvconf.
I thought, the saying from the beginning was just something from the "old days". No DNS Problems in 2024 anymore, right? But Ubuntu taught me different.
Ubuntu is using systemd-resolved since 20.04 (if I'm correct). But I was shocked, when I was looking at my uptime kuma Container on a Ubuntu 22.04 LTS Host. It was constantly failing. Sometimes 3 services at the same time, sometimes just 1 service a day. One Check suddenly failed. 60 seconds later, the next check, switched back to green again. But all fails had the same error message: "getaddrinfo ENOTFOUND domain.com". Doesn't matter if they were internal domains or external. Sometimes some of them just failed.
I thought it could be an old Firewall Applience that were running at like 120% system utilization and were serving DHCP and (with this) internal DNS. But no. Not even high latencies from that Firewall. Then I thought it might be AdGuard (in a Docker Container). So I switched to PiHole. But the problems were still the same.
Then I turned on debug logs of systemd-resolved and found out that sometimes it was switching to the secondary DNS Server for whatever reason and just attaching the search domain to the following requests:
1. AAAA of demodomain.com
--> no answer (because only A were available)
2. A of demodomain.com
--> somehow failed, systemd-resolved switched to second DNS (debug log of systemd-resolved is hard to read, not sure why it somehow failed)
3. AAAA of demodomain.com.local
--> it just attached the searchdomain of the system to the domain which now resulting in errors from all following DNS ServerAfter another round of wrong requests it suddenly get back his head. But in the meantime, uptime kuma already failed.
The solution in my case: switch "back" to resolvconf package on Ubuntu. Which comes to at least one downside: it seems to not have an interface to netplan and/or networkmanager (which leads to manual creating and managing of resolv.conf, not via DHCP, bummer). But after I switched: Everything is working fine and without any problems since days.
"We" also have an open bug report since 3 years: https://github.com/systemd/systemd/issues/21123
It's not exactly the same issue, but I think the root cause is connected somehow: it seems to be a problem of IPv6.
But a) I need (or better: want) IPv6 in my case/that network and b) WTF? How can this be a good solution to turn off IPv6 (https://github.com/systemd/systemd/issues/21123#issuecomment-2028976737)? Not to mention that we still need a solution for Post-IPv4.By the way: If you still experiencing DNS issues inside Docker Container, maybe Alpine could be another issue: https://martinheinz.dev/blog/92
#systemdresolved #ubuntu #Ubuntu2404 #uptimekuma #dns #usg #unifi #ubiquiti #adguard #pihole #netplan #networkmanager #ipv4 #ipv6 #alpine #docker #glibc #musl
-
So, after I met problems with iwlwifi driver and my attempts to aggregate both em0 and wlan0 interfaces to the one lagg0 interface (https://mastodon.bsd.cafe/@evgandr/115985853500057386) — looks like I found a much (MUCH!) simpler solution :drgn_happy:
I wanted to automatically switch between wired and wireless networks when I plug-in (or disconnect) my Ethernet cable. First, because I was a newbie in the FreeBSD world, I tried to search for some kind of NetworkManager. Thankfully, I didn't find any NetworkManager clone ported to the FreeBSD. I found some tries to port NetworkManager from Linux to FreeBSD but all of them are failed (not surprised, lol).
Then, I finally started to read documentation :drgn_think_science: . In the section about advanced networking I read about aggregation interfaces. And somehow I managed to aggregate both of em0 and wlan0 to the one lagg0 interface and it works well.
But, looks like (see https://mstdn.social/@erikarn/115986265106931691) it is not the way how the lagg interfaces should work. It is not intended to use wireless interfaces in the aggregate interfaces — so my tricky setup stopped working in the FreeBSD 15.0.
BUT, since we have a beautiful devd daemon, which listens for various system events and able to execute actions when event is happened — I just wrote 23 lines of shell script to learn my laptop how to switch between interfaces when the Ethernet cable (dis)connects, lol. Solution is very simple:
First, we already have /etc/devd/dhclient.conf, which starts dhclient when some interface appeared in the system. I modified it, so it calls the sPeCiAL script, each time when em0, or wlan0, or ue0 interface appeared in the system, or when em0 is disappeared:
notify 0 {
match "system" "IFNET";
match "type" "LINK_UP";
media-type "ethernet";
action "/root/bin/unfuck_network.tcsh $subsystem ifup";
};notify 0 {
match "system" "IFNET";
match "type" "LINK_DOWN";
media-type "ethernet";
action "/root/bin/unfuck_network.tcsh $subsystem ifdown";
};notify 0 {
match "system" "IFNET";
match "type" "LINK_UP";
media-type "802.11";
action "/root/bin/unfuck_network.tcsh $subsystem";
};notify 0 {
match "system" "ETHERNET";
match "type" "IFATTACH";
match "subsystem" "ue0";
action "/root/bin/unfuck_network.tcsh ue0";
};Then, the main magic happens in the /root/bin/unfuck_network.tcsh:
— When Ethernet cable is connected — it destroys the wlan0 interface and starts dhclient for em0 to talk with DHCP server.
— When Ethernet cable is disconnected — it makes all to remove route using em0 from routing table (removes em0 interface completely, flush routing table, etc — somehow em0 still stays in the routing table if interface is not destroyed; btw system will create it anyway later, in some point) and recreates the wlan0 interface.
— When wlan0 device is created — it starts dhclient for it.Script contents (for tcsh):
#!/bin/tcshswitch ( $1 )
case "em0":
if ( $2 == "ifup" ) then
service netif quietstop wlan0
service dhclient quietstart em0
else if ( $2 == "ifdown" ) then
service dhclient quietstop em0
ifconfig em0 delete
route flush
service routing restart
service netif quietstart wlan0
endif
breaksw;
case "wlan0":
service dhclient quietstart wlan0
breaksw;
case "ue0":
service dhclient quietstart ue0
breaksw;
endsw -
Setup -> I have a #RaspberryPi #RaspberryPiZero2W running #linux #pios #RaspberryPiOS #headless
Aim: I want the pi to always offer a #hotspot #wifihotspot on #boot
based on the commandnmcli device wifi hotspot ssid pizerohotspot password my_password
Is this below a bad solution ?? -> I came up with this solution in the link below using a .service but fear it shows I am clueless? Is there a better way? Thanks!
https://www.loramesh.org/subpages/pi_install.html#pi_hotspot -
I have updated my Automatic AccessPoint script.
Network at home & AccessPoint when you are out.
"AccessPopup" on the #raspberrypi so that it now works on #archlinux arm for Pi4 as well as the #Ubuntu Pi Image and #PiOS Bookworm. Will probably work on other systems but not tested yet. Needs NetworkManager but i also have a different script for the older PiOS -
#Linux Weekly Roundup for August 10th, 2025: #Debian 13, #Ubuntu 24.04.3 LTS, #LinuxMint 22.2 beta, #Mesa 25.2, Linux 6.17 RC, #NVIDIA 580 beta, #KDE Plasma 6.4.4, #Calibre 8.8, #GStreamer 1.26.5, #Tails 7.0 RC, #Audacity 3.7.5, #PeaZip 10.6, PorteuX 2.2, #HandBrake 1.10, #Darktable 5.2.1, NetworkManager 1.54, KDE Frameworks 6.17, #Amarok 3.3.1, and more https://9to5linux.com/9to5linux-weekly-roundup-august-10th-2025
-
#Linux Weekly Roundup for November 23rd, 2025: #Blender 5.0, #KDE Plasma 6.5.3, #Debian Libre Live Images, #PHP 8.5, #GIMP 3.2 Release Candidate, #AlmaLinux OS 9.7, VKD3D-Proton 3.0, #openSUSE's Agama 18 installer, KDE #Slimbook VII laptop, Finnix 251, #Budgie 10.10 Developer Preview, #Calibre 8.15, #Git 2.52, #Wireshark 4.6.1, #NetworkManager 1.54.2, #OpenVPN 2.6.16, HPLIP 3.25.8, and more https://9to5linux.com/9to5linux-weekly-roundup-november-23rd-2025
-
Done. Now its time to send this #zfs mount to darkstar02.
Using #linux as desktop-workstation for #animation and #film has gotten harder the last years. All the awesome tools like #mergerfs , nfscached, zfs and kickstart are outweighed by gnomes lack of documentation, sneaking abandon of old linux/unix virtues like one tool for one job (systemd, NetworkManager), multi user. But still #linux makes things just possible.
-
I'm trying to learn to develop multiplayer games in Unity, using the project from Binary Lunar's excellent tutorial:
https://www.youtube.com/watch?v=HWPKlpeZUjM
If I simultaneously run a local build on my machine (which happens to be a Linux box) and run the game within the Unity editor, I can create a host in one and a client in the other. They play together happily.
I also made a MacOS build and downloaded it on a separate machine. I can run the game there and create a host.
Strangely, if I then try to create a client from the other machine (either the local Linux build or within the editor), I get this message:
[Netcode] Cannot start Host while an instance is already running
UnityEngine.Debug:LogWarning (object)
Unity.Netcode.NetworkLog:LogWarning (string) (at ./Library/PackageCache/[email protected]/Runtime/Logging/NetworkLog.cs:28)
Unity.Netcode.NetworkManager:CanStart (Unity.Netcode.NetworkManager/StartType) (at ./Library/PackageCache/[email protected]/Runtime/Core/NetworkManager.cs:697)
Unity.Netcode.NetworkManager:StartHost () (at ./Library/PackageCache/[email protected]/Runtime/Core/NetworkManager.cs:830)
NetworkUI/<>c:<Awake>b__4_0 () (at Assets/Scripts/NetworkUI.cs:20)
UnityEngine.EventSystems.EventSystem:Update () (at ./Library/PackageCache/[email protected]/Runtime/EventSystem/EventSystem.cs:530)This only happens if there is an instance running on the Mac!
Of course, the only time you would want to create a client is if there is another instance (a host) running. It's as if the game running on the Mac and the one running in the Unity editor think they're the same program, and therefore refuse to make a "second" network connection.
What can I do about this?
-
#BeagleBoneBlack Disaster Flash
The good news is, if you do it like that--and means the only thing you do iis the basics of bringing it up to scratch, do not uninstall anything or change network managers,--you're fine.
If you want to switch to NetworkManager, however, two changes, gotten from the OTBR Codelab, but honestly, use it either way and just skip the NetworkManager part if you're not using that. This is the perfect tutorial:
-
AerynOS Repository Update: 18th November 2025
Another week, another repository sync. This was a shorter week as we synced later in the week last week, however there was still a lot that we got landed!
Major Updates:
- KDE Plasma 6.5.3
- Mesa v25.3
- KDE Frameworks v6.20.0
- vscode-bin v1.106.0
- wine v10.19
- intel-microcode v20251111
- ibus 1.5.33
- freeglut v3.8.0
- qemu now supports aarch64/arm
- git 2.52.0
- discord 0.0.114Minor updates:
- qemu
- qpwgraph
- easyeffects
- cython
- ddcutil
- wine
- glycin
- sshfs
- kquickimageeditor
- gnu-coreutils-compat
- syncthingtray
- resources
- just
- graphviz
- libdrm
- networkmanager
- drm-info
- glib2
- nssNew packages:
- deepfilternet
- kubectl (https://kubernetes.io/docs/reference/kubectl/)
- hugo (https://gohugo.io/)
- zola (https://www.getzola.org/)
- gitkFixes:
- solaar didn't have access to uinput in a Wayland session and so rules that created virtual keypresses wouldn't work
- DLSS support with the Nvidia driver not working (hopefully fixed)Security updates:
- libarchive 3.8.3
- openvpn 2.6.16Git changelog:
As usual, there are many other minor updates not mentioned as the list would get too long. You can see the full changelog here:
Get Involved:
If you would like to engage with the team, join our new Zulip server at:
https://aerynos.zulipchat.com/join/fuqokhsomj5mzqj6akqaiqlr
To sponsor the project, we accept donations via Ko-Fi to directly support our project costs and ensure the long term viability of the project:
-
AerynOS Repository Update: 13th November 2025
This week's sync is the first sync to our new Netcup root server, and you may or may not notice increased download speeds as a result. In terms of packaging work, it's been a reasonably normal week. Internally, the team has been busy with some background boring admin stuff related to bringing down our cost base and improving internal coordination.
Highlights:
- Cosmic Beta6
- KDE Gear 25.08.3
- Discord v0.0.114
- EasyEffects v8.0.0
- Firefox 145.0
- Fish shell 4.2.1
- Kernel 6.17.7
- Mesa v25.2.7
- Uutils-coreutils 0.4.0
- Zed v210.04New Apps:
- btop
- networkmanager-openvpn
- zulip-desktopFixes:
- font-noto-cjk: Fix and add sans-serif and serif versions
- nautilus: fix broken thumbnail generation
- waydroid: Add missing dnsmasq rundep
Other updates include, but are not limited to, containerd, curl, freerdp, golang, mullvad, nano, qbittorrent, runc, and the docker stack.Git changelog:
Support the project:
In our last blog post, we highlighted a renewed effort around donations. If you are able to support the project financially, it would be greatly appreciated if you could head over to our Ko-Fi page (https://ko-fi.com/aerynos) and chip in. Every little bit helps.
Thank you to everyone who has already signed up or chipped in!
As ever, if you want to get in touch or if you want to contribute via your skills, feel free to join us on Matrix (https://matrix.to/#/#aerynos:matrix.org) or comment on the GitHub Discussions post (https://github.com/orgs/AerynOS/discussions/108).
#Linux #FOSS #AerynOS #Distro #OpenSource #OSS #OS #DistroDevelopment
-
Can’t connect to public WiFi via Linux
I am using Fedora 42, and I am unable to access the captive portal of inOui TGV. I can connect to the WiFi with no problems, but I can’t get access to the Internet.
I have tried everything:
- Connecting to
192.168.1.1, tohttp://neverssl.com, tohttp://wifi.sncfvia browser - Running
sudo dhclient -rand rebooting the device - Running
systemctl restart NetworkManager - Also other stuff I tried in the past with other public WiFis. I am sure there is something wrong with my specific device configuration, it’s not this network.
Nothing works 😩
Can somebody who understands this stuff help me?
#help #mastoAiuto #Linux #LinuxNetwork #Network #CaptivePortal #WiFi #PublicWiFi #HTTP #HTTPS #troubleshoot #LinuxHelp #Fedora #Fedora42
- Connecting to
-
I'm running some tests on my old and trusted Raspberry Pi A+.
I've installed Raspbian - latest release, lite version. At the prompt, it's using 92 MB of RAM, mainly due to systemd and NetworkManager. As soon as you use it (even just for apt), it starts swapping and becomes almost unusable. It took 5 minutes just to install Python.I tried NetBSD (on the same memory card): 35 MB of RAM used (including Postfix!) and it's totally usable.
-
#AlmaLinux OS 9.6 Is Out as a Free Alternative to #RedHat #EnterpriseLinux 9.6
This release ships with GCC 11.5, LLVM 19.1.7, Go 1.23.6, Rust 1.84.1, Annobin 12.70, NetworkManager 1.52, QEMU 9.1, and other updates. https://9to5linux.com/almalinux-os-9-6-is-out-as-a-free-alternative-to-red-hat-enterprise-linux-9-6
#RHEL #RHEL9 #Linux -
I'm running some tests on my old and trusted Raspberry Pi A+.
I've installed Raspbian - latest release, lite version. At the prompt, it's using 92 MB of RAM, mainly due to systemd and NetworkManager. As soon as you use it (even just for apt), it starts swapping and becomes almost unusable. It took 5 minutes just to install Python.I tried NetBSD (on the same memory card): 35 MB of RAM used (including Postfix!) and it's totally usable.
-
I'm running some tests on my old and trusted Raspberry Pi A+.
I've installed Raspbian - latest release, lite version. At the prompt, it's using 92 MB of RAM, mainly due to systemd and NetworkManager. As soon as you use it (even just for apt), it starts swapping and becomes almost unusable. It took 5 minutes just to install Python.I tried NetBSD (on the same memory card): 35 MB of RAM used (including Postfix!) and it's totally usable.
-
I'm running some tests on my old and trusted Raspberry Pi A+.
I've installed Raspbian - latest release, lite version. At the prompt, it's using 92 MB of RAM, mainly due to systemd and NetworkManager. As soon as you use it (even just for apt), it starts swapping and becomes almost unusable. It took 5 minutes just to install Python.I tried NetBSD (on the same memory card): 35 MB of RAM used (including Postfix!) and it's totally usable.
-
I'm running some tests on my old and trusted Raspberry Pi A+.
I've installed Raspbian - latest release, lite version. At the prompt, it's using 92 MB of RAM, mainly due to systemd and NetworkManager. As soon as you use it (even just for apt), it starts swapping and becomes almost unusable. It took 5 minutes just to install Python.I tried NetBSD (on the same memory card): 35 MB of RAM used (including Postfix!) and it's totally usable.
-
How do I set up a captive portal (website login) WiFi on Linux?
I use wpa_supplicant. I am willing to move to a different daemon.I do not need the webpage to auto-open, I know the address.
Nothing obvious and web searches are failing me.
There's a few messy shell scrips for NetworkManager, but I don't see why wpa_supplicant would be unable to do it? -
It's always DNS, right? If someone experiencing some strange issues with ubuntu, maybe this toot is for you.
tl;dr: switch from systemd-resolved to resolvconf.
I thought, the saying from the beginning was just something from the "old days". No DNS Problems in 2024 anymore, right? But Ubuntu taught me different.
Ubuntu is using systemd-resolved since 20.04 (if I'm correct). But I was shocked, when I was looking at my uptime kuma Container on a Ubuntu 22.04 LTS Host. It was constantly failing. Sometimes 3 services at the same time, sometimes just 1 service a day. One Check suddenly failed. 60 seconds later, the next check, switched back to green again. But all fails had the same error message: "getaddrinfo ENOTFOUND domain.com". Doesn't matter if they were internal domains or external. Sometimes some of them just failed.
I thought it could be an old Firewall Applience that were running at like 120% system utilization and were serving DHCP and (with this) internal DNS. But no. Not even high latencies from that Firewall. Then I thought it might be AdGuard (in a Docker Container). So I switched to PiHole. But the problems were still the same.
Then I turned on debug logs of systemd-resolved and found out that sometimes it was switching to the secondary DNS Server for whatever reason and just attaching the search domain to the following requests:
1. AAAA of demodomain.com
--> no answer (because only A were available)
2. A of demodomain.com
--> somehow failed, systemd-resolved switched to second DNS (debug log of systemd-resolved is hard to read, not sure why it somehow failed)
3. AAAA of demodomain.com.local
--> it just attached the searchdomain of the system to the domain which now resulting in errors from all following DNS ServerAfter another round of wrong requests it suddenly get back his head. But in the meantime, uptime kuma already failed.
The solution in my case: switch "back" to resolvconf package on Ubuntu. Which comes to at least one downside: it seems to not have an interface to netplan and/or networkmanager (which leads to manual creating and managing of resolv.conf, not via DHCP, bummer). But after I switched: Everything is working fine and without any problems since days.
"We" also have an open bug report which since 3 years: https://github.com/systemd/systemd/issues/21123
It's not exactly the same issue, but I think the root cause is connected somehow: it seems to be a problem of IPv6.
But a) I need (or better: want) IPv6 in my case/that network and b) WTF? How can this be a good solution to turn off IPv6 (https://github.com/systemd/systemd/issues/21123#issuecomment-2028976737)? Not to mention that we still need a solution for Post-IPv4.By the way: If you still experiencing DNS issues inside Docker Container, maybe Alpine could be another issue: https://martinheinz.dev/blog/92
#systemdresolved #ubuntu #Ubuntu2404 #uptimekuma #dns #usg #unifi #ubiquiti #adguard #pihole #netplan #networkmanager #ipv4 #ipv6 #alpine #docker #glibc #musl
-
It's always DNS, right? If someone experiencing some strange issues with ubuntu, maybe this toot is for you.
tl;dr: switch from systemd-resolved to resolvconf.
I thought, the saying from the beginning was just something from the "old days". No DNS Problems in 2024 anymore, right? But Ubuntu taught me different.
Ubuntu is using systemd-resolved since 20.04 (if I'm correct). But I was shocked, when I was looking at my uptime kuma Container on a Ubuntu 22.04 LTS Host. It was constantly failing. Sometimes 3 services at the same time, sometimes just 1 service a day. One Check suddenly failed. 60 seconds later, the next check, switched back to green again. But all fails had the same error message: "getaddrinfo ENOTFOUND domain.com". Doesn't matter if they were internal domains or external. Sometimes some of them just failed.
I thought it could be an old Firewall Applience that were running at like 120% system utilization and were serving DHCP and (with this) internal DNS. But no. Not even high latencies from that Firewall. Then I thought it might be AdGuard (in a Docker Container). So I switched to PiHole. But the problems were still the same.
Then I turned on debug logs of systemd-resolved and found out that sometimes it was switching to the secondary DNS Server for whatever reason and just attaching the search domain to the following requests:
1. AAAA of demodomain.com
--> no answer (because only A were available)
2. A of demodomain.com
--> somehow failed, systemd-resolved switched to second DNS (debug log of systemd-resolved is hard to read, not sure why it somehow failed)
3. AAAA of demodomain.com.local
--> it just attached the searchdomain of the system to the domain which now resulting in errors from all following DNS ServerAfter another round of wrong requests it suddenly get back his head. But in the meantime, uptime kuma already failed.
The solution in my case: switch "back" to resolvconf package on Ubuntu. Which comes to at least one downside: it seems to not have an interface to netplan and/or networkmanager (which leads to manual creating and managing of resolv.conf, not via DHCP, bummer). But after I switched: Everything is working fine and without any problems since days.
"We" also have an open bug report since 3 years: https://github.com/systemd/systemd/issues/21123
It's not exactly the same issue, but I think the root cause is connected somehow: it seems to be a problem of IPv6.
But a) I need (or better: want) IPv6 in my case/that network and b) WTF? How can this be a good solution to turn off IPv6 (https://github.com/systemd/systemd/issues/21123#issuecomment-2028976737)? Not to mention that we still need a solution for Post-IPv4.By the way: If you still experiencing DNS issues inside Docker Container, maybe Alpine could be another issue: https://martinheinz.dev/blog/92
#systemdresolved #ubuntu #Ubuntu2404 #uptimekuma #dns #usg #unifi #ubiquiti #adguard #pihole #netplan #networkmanager #ipv4 #ipv6 #alpine #docker #glibc #musl
-
It's always DNS, right? If someone experiencing some strange issues with ubuntu, maybe this toot is for you.
tl;dr: switch from systemd-resolved to resolvconf.
I thought, the saying from the beginning was just something from the "old days". No DNS Problems in 2024 anymore, right? But Ubuntu taught me different.
Ubuntu is using systemd-resolved since 20.04 (if I'm correct). But I was shocked, when I was looking at my uptime kuma Container on a Ubuntu 22.04 LTS Host. It was constantly failing. Sometimes 3 services at the same time, sometimes just 1 service a day. One Check suddenly failed. 60 seconds later, the next check, switched back to green again. But all fails had the same error message: "getaddrinfo ENOTFOUND domain.com". Doesn't matter if they were internal domains or external. Sometimes some of them just failed.
I thought it could be an old Firewall Applience that were running at like 120% system utilization and were serving DHCP and (with this) internal DNS. But no. Not even high latencies from that Firewall. Then I thought it might be AdGuard (in a Docker Container). So I switched to PiHole. But the problems were still the same.
Then I turned on debug logs of systemd-resolved and found out that sometimes it was switching to the secondary DNS Server for whatever reason and just attaching the search domain to the following requests:
1. AAAA of demodomain.com
--> no answer (because only A were available)
2. A of demodomain.com
--> somehow failed, systemd-resolved switched to second DNS (debug log of systemd-resolved is hard to read, not sure why it somehow failed)
3. AAAA of demodomain.com.local
--> it just attached the searchdomain of the system to the domain which now resulting in errors from all following DNS ServerAfter another round of wrong requests it suddenly get back his head. But in the meantime, uptime kuma already failed.
The solution in my case: switch "back" to resolvconf package on Ubuntu. Which comes to at least one downside: it seems to not have an interface to netplan and/or networkmanager (which leads to manual creating and managing of resolv.conf, not via DHCP, bummer). But after I switched: Everything is working fine and without any problems since days.
"We" also have an open bug report which since 3 years: https://github.com/systemd/systemd/issues/21123
It's not exactly the same issue, but I think the root cause is connected somehow: it seems to be a problem of IPv6.
But a) I need (or better: want) IPv6 in my case/that network and b) WTF? How can this be a good solution to turn off IPv6 (https://github.com/systemd/systemd/issues/21123#issuecomment-2028976737)? Not to mention that we still need a solution for Post-IPv4.By the way: If you still experiencing DNS issues inside Docker Container, maybe Alpine could be another issue: https://martinheinz.dev/blog/92
#systemdresolved #ubuntu #Ubuntu2404 #uptimekuma #dns #usg #unifi #ubiquiti #adguard #pihole #netplan #networkmanager #ipv4 #ipv6 #alpine #docker #glibc #musl
-
Setup -> I have a #RaspberryPi #RaspberryPiZero2W running #linux #pios #RaspberryPiOS #headless
Aim: I want the pi to always offer a #hotspot #wifihotspot on #boot
based on the commandnmcli device wifi hotspot ssid pizerohotspot password my_password
Is this below a bad solution ?? -> I came up with this solution in the link below using a .service but fear it shows I am clueless? Is there a better way? Thanks!
https://www.loramesh.org/subpages/pi_install.html#pi_hotspot -
Setup -> I have a #RaspberryPi #RaspberryPiZero2W running #linux #pios #RaspberryPiOS #headless
Aim: I want the pi to always offer a #hotspot #wifihotspot on #boot
based on the commandnmcli device wifi hotspot ssid pizerohotspot password my_password
Is this below a bad solution ?? -> I came up with this solution in the link below using a .service but fear it shows I am clueless? Is there a better way? Thanks!
https://www.loramesh.org/subpages/pi_install.html#pi_hotspot -
Setup -> I have a #RaspberryPi #RaspberryPiZero2W running #linux #pios #RaspberryPiOS #headless
Aim: I want the pi to always offer a #hotspot #wifihotspot on #boot
based on the commandnmcli device wifi hotspot ssid pizerohotspot password my_password
Is this below a bad solution ?? -> I came up with this solution in the link below using a .service but fear it shows I am clueless? Is there a better way? Thanks!
https://www.loramesh.org/subpages/pi_install.html#pi_hotspot -
Can’t connect to public WiFi via Linux
I am using Fedora 42, and I am unable to access the captive portal of inOui TGV. I can connect to the WiFi with no problems, but I can’t get access to the Internet.
I have tried everything:
- Connecting to
192.168.1.1, tohttp://neverssl.com, tohttp://wifi.sncfvia browser - Running
sudo dhclient -rand rebooting the device - Running
systemctl restart NetworkManager - Also other stuff I tried in the past with other public WiFis. I am sure there is something wrong with my specific device configuration, it’s not this network.
Nothing works 😩
Can somebody who understands this stuff help me?
#help #mastoAiuto #Linux #LinuxNetwork #Network #CaptivePortal #WiFi #PublicWiFi #HTTP #HTTPS #troubleshoot #LinuxHelp #Fedora #Fedora42
- Connecting to