Beta-stage testing for Yunohost 11.0/Bullseye and Buster->Bullseye migration

I hit a small snag while installing on a test system on Proxmox, with smallish disk space (8 GB),

After acknowedging the warning and continuing, there was an error saying /etc/resolv.dnsmasq.conf did not exist (Proxmox issue, perhaps).

On retry, I did not get so far. I did touch /etc/resolv.dnsmasq.conf, but the postinstall crashed because the designated domain already got set up on the first attempt.

It’s a bit of an edge case, and certainly no show stopper, but if it is clear what the problem is than it might be worth repairing. Maybe I’ll try to find the responsible code on Github, it would be a nice intro :slight_smile:

@wbk : i think the issue you encountered is indeed an important bug related to the new defaultapp mechanism, configured from the domain config panel (it’s very technical, can’t really explain it with just a couple sentences). Anyway that’s fixed in dev branch but we didn’t yet push a new release.

1 Like

(From a ynhuser in the support chat room)

The configuration panel for automated DNS records has wrong text boxes types:

The last three should be text boxes, not list selection boxes.

Hi all,
I started a new installation, just one application :slight_smile:

  • Roundcube : one issue during installation :
    /usr/share/yunohost/helpers.d/php: ligne 408: / 2 / 20 : erreur de syntaxe : opérande attendu (le symbole erroné est « / 2 / 20 »)
    and when a start the application I have a BAD GATEWAY 502.

Thanks for your help if you have a solution.

Hi, I have exactly the same problem for 14 days, for now I have not found how to fix it… I am running under armbian linux-image-edge-arm64 kernel 5.15.y.

I re-installed Debian and Yunohost, same issue.
Next added Nextcloud+Nextcould&Mail
I removed Roundcube. I have a warning : no configuration file found!
And re-installed Roundcube: No issue
Roundcube work correctly!

Petit tour des soucis suite à la migration

Elle m’est toujours proposées et quand je tente de la lancer :
Exécution de la migration 0021_migrate_to_bullseye…

Démarrage de la migration vers Bullseye

La migration 0021_migrate_to_bullseye a échoué avec l’exception Command ‘wget -O- https://forge.yunohost.org/yunohost_bullseye.asc -q | apt-key add -qq -’ returned non-zero exit status 2. : annulation

Je suis bien passé sous Debian 11 mais de là plusieurs soucis

Dnsmasq est en panne, impossible à relancer
Fail2ban est en panne et idem

Impossible de faire les mises a jour de yunohost de la 4.4.0 vers la 11

Une partie des logs dans les tentatives de mises a jour système

2022-02-24 18:12:33,840: INFO - Mise à jour des paquets du système (non liés a YunoHost)...
2022-02-24 18:12:34,119: DEBUG - Running apt command :
DEBIAN_FRONTEND=noninteractive APT_LISTCHANGES_FRONTEND=none apt-get --fix-broken --show-upgraded --assume-yes --quiet -o=Dpkg::Use-Pty=0 -o Dpkg::Options::="--force-confold" -o Dpkg::Options::="--force-confmiss" -o Dpkg::Options::="--force-confdef" dist-upgrade
2022-02-24 18:12:34,174: INFO - + Lecture des listes de paquets…
2022-02-24 18:12:34,367: INFO - + Construction de l'arbre des dépendances…
2022-02-24 18:12:34,369: INFO - + Lecture des informations d'état…
2022-02-24 18:12:34,486: INFO - + Calcul de la mise à jour…
2022-02-24 18:12:34,590: INFO - + Le paquet suivant a été installé automatiquement et n'est plus nécessaire :
2022-02-24 18:12:34,591: INFO - +   libreadline7
2022-02-24 18:12:34,591: INFO - + Veuillez utiliser « apt autoremove » pour le supprimer.
2022-02-24 18:12:34,611: INFO - + Les paquets suivants seront ENLEVÉS :
2022-02-24 18:12:34,612: INFO - +   libgc1c2
2022-02-24 18:12:34,612: INFO - + Les NOUVEAUX paquets suivants seront installés :
2022-02-24 18:12:34,614: INFO - +   libgc1
2022-02-24 18:12:34,615: INFO - + Les paquets suivants ont été conservés :
2022-02-24 18:12:34,615: INFO - +   moulinette ssowat yunohost yunohost-admin
2022-02-24 18:12:34,616: INFO - + Les paquets suivants seront mis à jour :
2022-02-24 18:12:34,618: INFO - +   guile-2.2-libs
2022-02-24 18:12:34,633: INFO - + 1 mis à jour, 1 nouvellement installés, 1 à enlever et 4 non mis à jour.
2022-02-24 18:12:34,634: INFO - + Il est nécessaire de prendre 5 219 ko dans les archives.
2022-02-24 18:12:34,634: INFO - + Après cette opération, 32,8 ko d'espace disque supplémentaires seront utilisés.
2022-02-24 18:12:34,635: INFO - + Err :1 http://debian.mirrors.ovh.net/debian bullseye/main amd64 libgc1 amd64 1:8.0.4-3
2022-02-24 18:12:34,635: INFO - +   Erreur temporaire de résolution de « debian.mirrors.ovh.net »
2022-02-24 18:12:34,635: INFO - + Err :2 http://debian.mirrors.ovh.net/debian bullseye/main amd64 guile-2.2-libs amd64 2.2.7+1-6
2022-02-24 18:12:34,636: INFO - +   Erreur temporaire de résolution de « debian.mirrors.ovh.net »
2022-02-24 18:12:34,637: WARNING - E: Impossible de récupérer http://debian.mirrors.ovh.net/debian/pool/main/libg/libgc/libgc1_8.0.4-3_amd64.deb  Erreur temporaire de résolution de « debian.mirrors.ovh.net »
2022-02-24 18:12:34,637: WARNING - E: Impossible de récupérer http://debian.mirrors.ovh.net/debian/pool/main/g/guile-2.2/guile-2.2-libs_2.2.7%2b1-6_amd64.deb  Erreur temporaire de résolution de « debian.mirrors.ovh.net »
2022-02-24 18:12:34,638: WARNING - E: Impossible de récupérer certaines archives, peut-être devrez-vous lancer apt-get update ou essayer avec --fix-missing ?
2022-02-24 18:12:35,891: WARNING - Impossible de mettre à jour les paquets suivants : guile-2.2-libs

Pour DNSMasq j’ai ça

févr. 24 16:08:05 dnsmasq[17052]: Usage: /etc/init.d/dnsmasq {start|stop|restart|force-reload|dump-stats|status}
févr. 24 16:08:05 systemd[1]: dnsmasq.service: Control process exited, code=exited, status=3/NOTIMPLEMENTED
févr. 24 16:08:05 systemd[1]: dnsmasq.service: Failed with result 'exit-code'.
févr. 24 16:08:05 systemd[1]: Failed to start dnsmasq - A lightweight DHCP and caching DNS server.
févr. 24 16:15:10 systemd[1]: Starting dnsmasq - A lightweight DHCP and caching DNS server...
févr. 24 16:15:10 dnsmasq[18960]: Usage: /etc/init.d/dnsmasq {start|stop|restart|force-reload|dump-stats|status}
févr. 24 16:15:10 systemd[1]: dnsmasq.service: Control process exited, code=exited, status=3/NOTIMPLEMENTED
févr. 24 16:15:10 systemd[1]: dnsmasq.service: Failed with result 'exit-code'.
févr. 24 16:15:10 systemd[1]: Failed to start dnsmasq - A lightweight DHCP and caching DNS server.
févr. 24 16:21:42 systemd[1]: Starting dnsmasq - A lightweight DHCP and caching DNS server...
févr. 24 16:21:42 dnsmasq[19892]: Usage: /etc/init.d/dnsmasq {start|stop|restart|force-reload|dump-stats|status}
févr. 24 16:21:42 systemd[1]: dnsmasq.service: Control process exited, code=exited, status=3/NOTIMPLEMENTED
févr. 24 16:21:42 systemd[1]: dnsmasq.service: Failed with result 'exit-code'.
févr. 24 16:21:42 systemd[1]: Failed to start dnsmasq - A lightweight DHCP and caching DNS server.

Le diagnostic me dit La résolution du nom de domaine semble être cassée sur votre serveur, ce qui semble lié au fait que /etc/resolv.conf ne pointe pas vers 127.0.0.1 .
Pourtant, le fichier est bon, j’ai suivi des procédures indiqués dans d’autres topics du forum pour regenerer le fichier.

Bref je suis un peu bloqué (bon, mes sites hébergé marchent bien en soit, par contre plus d’emails :smiley: )

Es-ce que c’est utile que j’ouvre des topics vis a vis de mes bugs ?

Tu peux les lister ici même :slight_smile:

3 post au dessus

@Neru here are some commands to help us to understand exactly your issues:

lsb_release -a
cat /etc/resolv.conf
systemctl restart dnsmasq
systemctl status dnsmasq
journalctl -n50 -u dnsmasq
dpkg -l | grep yunohost
dpkg -l | grep linux

After that, you probably could temporarily change the /etc/resolv.conf by using the nameserver you want:

cp /etc/resolv.conf /etc/resolv.conf.bak
echo "nameserver 89.234.141.66" > /etc/resolv.conf

Check it pings correctly

ping wikipedia.org

And retry to add the bullseye key:

wget -O- https://forge.yunohost.org/yunohost_bullseye.asc 1 -q | apt-key add -qq -
echo $?

After that you should retry to migrate.

And finally, check DNSmasq is up again and replace the temporary resolv.conf:

systemctl status dnsmasq
cp /etc/resolv.conf.bak /etc/resolv.conf 
1 Like

Ok, ça m’as pas mal aidé, j’ai pu finir la migration.

Mais j’ai toujours quelques soucis dans les diagnostic.
Maintenant, ça me dit : La résolution du nom de domaine semble interrompue pour une raison quelconque … Un pare-feu bloque-t-il les requêtes DNS ?

Le journal de DNSMasq me dit : hastebin - esevohasuh (yunohost.org)

Et j’ai cet énorme log quand je tente de le démarrer en SSH
hastebin - ivagivahed (yunohost.org)

Bref je ne sais pas trop quoi faire.

Mes tentatives de force et de regen conf ne fonctionnent pas (j’ai épluché les topics du forum sur dns masq, resolv conf)

Your issue seems bind to the fact you have a lot of network interfaces.
could you run

ip link
cat /etc/dnsmasq.conf
ls /etc/dnsmasq.d

It’s so weird, did you think the migration cause this ?

IP link

1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group default qlen 1
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: bond0: <BROADCAST,MULTICAST,MASTER> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
    link/ether 4e:20:b7:57:d8:9f brd ff:ff:ff:ff:ff:ff
3: dummy0: <BROADCAST,NOARP> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
    link/ether 06:5a:88:cf:e0:40 brd ff:ff:ff:ff:ff:ff
4: ifb0: <BROADCAST,NOARP> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 32
    link/ether 82:da:ce:78:72:2f brd ff:ff:ff:ff:ff:ff
5: ifb1: <BROADCAST,NOARP> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 32
    link/ether f2:87:1e:a5:2d:f0 brd ff:ff:ff:ff:ff:ff
6: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP mode DEFAULT group default qlen 1000
    link/ether 00:25:90:23:12:b8 brd ff:ff:ff:ff:ff:ff
7: eth1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
    link/ether 00:25:90:23:12:b9 brd ff:ff:ff:ff:ff:ff
8: teql0: <NOARP> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 100
    link/void
9: tunl0@NONE: <NOARP> mtu 1480 qdisc noop state DOWN mode DEFAULT group default qlen 1
    link/ipip 0.0.0.0 brd 0.0.0.0
10: gre0@NONE: <NOARP> mtu 1476 qdisc noop state DOWN mode DEFAULT group default qlen 1
    link/gre 0.0.0.0 brd 0.0.0.0
11: gretap0@NONE: <BROADCAST,MULTICAST> mtu 1462 qdisc noop state DOWN mode DEFAULT group default qlen 1000
    link/ether 00:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff
12: sit0@NONE: <NOARP> mtu 1480 qdisc noop state DOWN mode DEFAULT group default qlen 1
    link/sit 0.0.0.0 brd 0.0.0.0
13: ip6tnl0@NONE: <NOARP> mtu 1452 qdisc noop state DOWN mode DEFAULT group default qlen 1
    link/tunnel6 :: brd ::

cat /etc/dnsmasq.conf

domain-needed
expand-hosts
localise-queries



interface=lo

resolv-file=/etc/resolv.dnsmasq.conf

ls /etc/dnsmasq.d


cloud.karma-kusala.com   karma-kusala.com       neru.io  portail.karma-kusala.com  seedsmag.com                 www.neru.io
hadratrancefestival.net  mail.karma-kusala.com  ohao.io  README                    www.hadratrancefestival.net  www.ohao.io

Do we have an expected release date? I have to migrate my yunohost server by 4 april (scaleway is cleaning some of their old bays) should backuping on my current up to date stable yunohost and deploying them to beta yunohost work?

Do we have an expected release date?

I believe Aleks and other devs have said this in other threads, “don’t wait”. They don’t have expected release dates. They will release it when they feel it is ‘done’.

My gut tells me that debian 10 to 11 is a major change, so I wouldn’t expect backups to work. Have a backup of all your data along with the yunohost backups. Then if restore backup fails, you have all the data so you can do a fresh install and drop your data in there.

Backups between the last 4.x version and 11.x should theoretically work though (we have some tests validating this). It may depends on super technical details though, or the fact that some apps (such as mailman2 and ffsync) can’t work on Bullseye because there’s no Python2 anymore, for example.

3 Likes

Hi!

First great work with Yunohost and the Bullseye update :slight_smile:
I’m trying to restore my system from a Borg backup on a fresh installation of Armbian_21.08.1_Rockpro64_bullseye_current_5.10.60

It is been setup on a RockPro64 and I’m facing some errors during the restore, some minor and a critical one regarding users.
When I try to list them using sudo yunohost user list output is users:
Also when I use sudo slapcat I don’t see there any of the users from my previous system.

I’ll also paste the output from them yunohost backup restore _auto_conf --system (as root of course)

Info: The configuration file ‘/etc/fail2ban/jail.conf’ is now managed by YunoHost (category fail2ban). **Success!** Configuration updated for 'fail2ban'
Warning: setfacl: /var/log/nginx: No such file or directory
Error: Could not run script: /usr/share/yunohost/hooks/conf_regen/01-yunohost
Warning: nginx: [alert] could not open error log file: open() “/var/log/nginx/error.log” failed (2: No such file or directory)
Warning: 2022/03/03 09:27:19 [info] 6010#6010: Using 131072KiB of shared memory for nchan in /etc/nginx/nginx.conf:61
Warning: nginx: the configuration file /etc/nginx/nginx.conf syntax is ok
Warning: 2022/03/03 09:27:19 [emerg] 6010#6010: open() “/var/log/nginx/access.log” failed (2: No such file or directory)
Warning: nginx: configuration file /etc/nginx/nginx.conf test failed
Error: Could not run script: /usr/share/yunohost/hooks/conf_regen/15-nginx
Warning: chown: cannot access ‘/var/log/redis’: No such file or directory
Error: Could not run script: /usr/share/yunohost/hooks/conf_regen/36-redis
Warning: Job for fail2ban.service failed.
Warning: See “systemctl status fail2ban.service” and “journalctl -xe” for details.
Error: Could not run script: /usr/share/yunohost/hooks/conf_regen/52-fail2ban
Success! YunoHost is now configured
Warning: The post-install completed! To finalize your setup, please consider:
– adding a first user through the ‘Users’ section of the webadmin (or ‘yunohost user create ’ in command-line);
– diagnose potential issues through the ‘Diagnosis’ section of the webadmin (or ‘yunohost diagnosis run’ in command-line);
– reading the ‘Finalizing your setup’ and ‘Getting to know YunoHost’ parts in the admin documentation: Administration guide | Yunohost Documentation.
Warning: 62208a10 Entry (cn=mail.main,ou=permission,dc=yunohost,dc=org): object class ‘permissionYnh’ requires attribute ‘authHeader’
Warning: slapadd: dn=“cn=mail.main,ou=permission,dc=yunohost,dc=org” (line=294): (65) object class ‘permissionYnh’ requires attribute ‘authHeader’
Error: Could not run script: /usr/share/yunohost/hooks/restore/05-conf_ldap
Error: Could not restore the ‘conf_ldap’ system part
Info: The operation ‘Restore system from a backup archive’ could not be completed. Please share the full log of this operation using the command ‘yunohost log share 20220303-092744-backup_restore_system’ to get help
Warning: The configuration file ‘/etc/cron.d/yunohost-diagnosis’ has been manually modified and will not be updated
Success! Configuration updated for ‘ssh’
Warning: The configuration file ‘/etc/nginx/conf.d/dentsh.host.conf’ has been manually modified and will not be updated
Success! Configuration updated for ‘dnsmasq’
Success! Configuration updated for ‘fail2ban’
Warning: setfacl: /var/log/nginx: No such file or directory
Error: Could not run script: /usr/share/yunohost/hooks/conf_regen/01-yunohost
Warning: chown: cannot access ‘/var/log/redis’: No such file or directory
Error: Could not run script: /usr/share/yunohost/hooks/conf_regen/36-redis
Warning: Job for fail2ban.service failed.
Warning: See “systemctl status fail2ban.service” and “journalctl -xe” for details.
Error: Could not run script: /usr/share/yunohost/hooks/conf_regen/52-fail2ban
Success! Restoration completed
apps:
system:
conf_ldap: Error
conf_manually_modified_files: Success
conf_ynh_certs: Success
conf_ynh_settings: Success

The errors regarding nginx, fail2ban, redis and yunohost I think I solved them already doing sudo yunohost tools regen-conf -f and creating the missing files/folders manually.

Any clues?

Thanks for your hard work and your help :smiley:

Hi again :slight_smile:

So after some debugging and digging, I found out that in the backup of my old system the file dc=yunohost-dc=org.ldif was missing some value in some entries, specifically on the mail.main and xmpp.main permissions.
In my old system those appear to be still wrong for some reason that I don’t know.
In the logs that I shared, you can see the error message related to this

So in order to solve it I had to add these values (by editing the ldif file) to those mail.main and xmpp.main entries:

mail.main

isProtected: TRUE
label: E-mail
showTile: FALSE
authHeader: FALSE

xmpp.main

isProtected: TRUE
label: XMPP
showTile: FALSE
authHeader: FALSE
1 Like

Bon, je m’en sors toujours pas avec ce que j’ai mis dans ma dernière réponse :frowning: