Hestia on Ubuntu 20.04 LTS reboot server and not work

Removed yaml configs from / etc / netplan - they conflicted with the main network interface.

In the main network interface, I registered an additional IP address 185.100.65.33

in / run / systemd / resolv - registered the dns of the google server.

Removed the symlink /etc/resolv.conf and created it as a file with dns google servers.

Changed the config of the networking.script file

The bug seems to exists out of 2 problems:

User add ip via v-add-sys-ip / UI

This should not matter at first glance as both configured with /etc/network/interfaces (Debian ā€œStandardā€) and Netplan (Ubuntu ā€œstandardā€)

Then there is a check for Netplan
If Netplan is installed it will use that on

It checks if Netplan is installed but nut actively in use and add file / modify 60-hestia.yalm or when not installed add a line in /etc/network/interfaces

It has been introduced in:
https://github.com/hestiacp/hestiacp/commit/7f27d6c1a2a4a51ccfe729d721b3e3d0f61db76c#diff-aab7b19881bda15fa03a5909eb6bb1582829478f4286fd1119d1ac91383cbfd4 on 13 Jan 2019

For the second issue:
There seems to be issues with resolving when TS secondary ip is active I have no idea how to solve itā€¦

We just released 1.4.14 which helps to prevent such a behaviour in a future case, but I would like to catch up the current case.

@Al2333 in your case, you probaly ignored or forgot the following warning:

Due to that, you got an mix between systemd/interfaces and netplan config after youā€™ve added an additional ip. This caused into a network loss, which would have been prevented if you would either uninstall netplan or as suggested, completly switch over to it (error message above).

The new release contains the following change, which improves v-add-sys-ip to prevent such a crash - just in case there are other users which ignored the warning - also we do scan the current system if there is a double config present:

@glonin.kz for your system, please validate if youā€™ve got aswell a mixed configuration between netplan and/or systemd/interfaces. If yes, please clean that out and reboot the server, if all works well, upgrade to the latest hestia version as soon as you can. If then still issues shows up, please contact us again.

1 Like

I didnā€™t ignore it - as long as that file in the netplan folder existed - the server was not accessible from the network, each of my 14 servers with Hestia disappeared from the network exactly after the reboot, until the file in the netplan folder was deleted.

And until I deleted that file from the netplan folder - the servers did not appear on the network.
There was no action on my part, the panel itself did everything:

  1. fresh installation of the new panel 1.4.13;
  2. 1 day of work - at the end of the day a strong slowdown;
  3. Reboot;
  4. The server is no longer in the network.

This scenario with a small variant for all 14 servers with Hestia.

Please tell me the command to force Hestia to update via SSH (SSH on the servers is working, but the panel itself does not work on half of the servers, as nginx or apache do not start)

apt update && apt upgrade

No, these commands show that I have the latest version 1.4.13.
They donā€™t see version 1.4.14.

The commands from @jlguerrero are right, the repository is shipping 1.4.14, it has been tested a hour ago.

root@web01:~# apt-cache show hestia | head -n 22
Source: hestia
Package: hestia
Priority: optional
Version: 1.4.14
Section: admin
Maintainer: HestiaCP <[email protected]>
Homepage: https://www.hestiacp.com
Architecture: amd64
Depends: bash, awk, sed, acl, sysstat, setpriv | util-linux (>= 2.33), zstd, lsb-release
Description: hestia
 hestia is an open source hosting control panel.
 hestia has a clean and focused interface without the clutter.
 hestia has the latest of very innovative technologies.
 hestia is a fork from VestaCP, special thanks to vestacp.com and Serghey Rodin
Description-md5: ae2a35218bf144a1a16e0e3812f27380
Filename: pool/focal/main/h/hestia/hestia_1.4.14_amd64.deb
MD5sum: cfd5f6241b98c1bee1325d007d00bfe8
SHA1: e2610d757f39b9a006ce1ebbe84fd36c12fe051b
SHA256: 6d829c0b5b9d11549364cee170665009849754a7b6bac5d8a502064e9efd8766
SHA512: b67f8691ecd9509bfb5ecb4b9184fc297cef98da824c298c49695fc824ddfd6897735d711d455315cbebcaf88854b3bc6ddca4d7c2cb5ba2ac8388c26add21a6
Size: 2533640
root@web01:~# apt list --upgradable
Auflistung... Fertig
hestia/focal 1.4.14 amd64 [aktualisierbar von: 1.4.13]
N: Es gibt 26 zusƤtzliche Versionen. Bitte verwenden Sie die Option Ā»-aĀ«, um sie anzuzeigen.
root@web01:~# lsb_release -a
No LSB modules are available.
Distributor ID: Ubuntu
Description:    Ubuntu 20.04.3 LTS
Release:        20.04
Codename:       focal

Apparently it depends on the location - some servers do now see 1.4.14, but most do not yet, probably will be updated soon.

Should not be the case, all server should be up to date after running apt update once.

Several servers have already updated to 1.4.14 - the problem has not disappeared. Servers are empty, without sites, but you have to wait a minute for each action, the servers practically do not work.

Do I understand correctly that there is no way to restore them? And you have to delete everything?

1.4.14 does not contain a magic fix we only added a extra check to prevent Netplan used if /etc/network/interfaces is used otherwise. We didnā€™t like the idea to first delete all the additional ips and then re add them. It is a to big risk for our current usersā€¦ See:

Also we have added an check for existing users to be warn current users with this kind of network setup up that they need check their network settings.

During install we have a check and a warning that would show that this kind of setup does not work and you could not add an extra ip. The only thing ā€œHestiaā€ lacked was adding proper validation on v-add-sys-ip.

In the last 2,5 years when it was implemented nobody had complained about this issue and at randomly it seems to be an issue. I can not track why this method has been used and how it was implemented but the bug causing this has been fixed.

In 1.4.13 we have added a small wait in to iptables start up script to prevent any issues with iptables busy and the firewall not starting properly.

In our ā€œLicenseā€ (https://github.com/hestiacp/hestiacp/blob/main/LICENSE) You can se we donā€™t provide any warranty.

I think we have done more then enough and without any income out of this project I have nothing to say about it.

You misunderstood me. I am not demanding anything from you, I just asked if it is possible to fix running problematic servers now on the fly without deleting all data and formatting the server, or it is impossible and you have to delete everything and reinstall and install Hestia from scratch.

I am not very familiar with Ubuntu and other Linux systems, so I put your panel in automatic mode and did not touch anything, and then also not touch anything, and just add IP and domains according to the instructions, I do not do other things with the panel and I can not do any settings, because I am not familiar with Ubuntu.

Thatā€™s why I asked, if you can fix it, then I will try to ask you - how and what I do. If it canā€™t be fixed (without some serious Ubuntu knowledge and time spent to learn it) then of course I donā€™t need to play your panel anymore, Iā€™d better transfer everything to ISP and never have to face such sad situation again.

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.