YunoHost 11.0 (Bullseye) release / Sortie de YunoHost 11.0 (Bullseye)

@Aleks @eivind

I’m still stuck with postgresql migration (see above, problem with mobilizon).
Did anyone solve that issue?

Since mobilizon obviously is causing the problem, I deinstalled it. Sadly I still can not migrate postgresql.

Since there are other applications that are affected by that problem (mastodon, matrix, moodle, pixelfed, peertube, …) I’m now at the point that I want to delete (“DROP”) the mobilizon database by hand:

root@yunohost# su postgres
postgres@yunohost$ psql

postgres=# \l
        Name         |     Owner      | Encoding |   Collate   |    Ctype    |       Access privileges        
---------------------+----------------+----------+-------------+-------------+--------------------------------
 mobilizon           | mobilizon      | UTF8     | en_US.UTF-8 | en_US.UTF-8 | =Tc/mobilizon                 +
                     |                |          |             |             | mobilizon=C*T*c*/mobilizon

postgres=# DROP DATABASE mobilizon;

Is this a good idea, or should I avoid doing that?

1 Like

@chmeyer : I think some people managed to work around the issue by uninstalling Mobilizon then restoring a backup, which should effectively behind the scene restore the SQL DB on postgresql 13 instead of 11, somewhat making the migration that way

(Also I confirmed with some people familiar with postgis (a postgres extension for geographical data) that it’s a mess to upgrade postgresql version when that module is involved, though I don’t have any exact detail to back this up)

1 Like

I have a similar problem with Mobilizon. I have done like this. I kept a backup of mobilizon (done before). I remove the app (don’t forget remove also the folder /home/yunohost.app/mobilizon)

I delete all postgresql, also the folder /etc/postgresql/11.
Be sure no other postgresql cluster are running:
ps -C postgres this command will show if one is running.
then remove purge all postgresql, also folders

sudo apt-get --purge remove postgresql\*
rm -r /etc/postgresql/
rm -r /etc/postgresql-common/
rm -r /var/lib/postgresql/
userdel -r postgres
groupdel postgres

Then I restore my mobilizon app, and it works…
For info:

$ dpkg --list | grep postgis
ii  postgis                               3.2.2+dfsg-1.pgdg110+1                                                   amd64        Geographic objects support for PostgreSQL
ii  postgis-doc                           3.2.2+dfsg-1.pgdg110+1                                                   all          Geographic objects support for PostgreSQL -- documentation
ii  postgresql-13-postgis-3               3.1.1+dfsg-1                                                             amd64        Geographic objects support for PostgreSQL 13
ii  postgresql-13-postgis-3-scripts       3.1.1+dfsg-1                                                             all          Geographic objects support for PostgreSQL 13 -- SQL scripts

$ pg_lsclusters
Ver Cluster Port Status Owner    Data directory              Log file
13  main    5432 online postgres /var/lib/postgresql/13/main /var/log/postgresql/postgresql-13-main.log

My instance Mobilizon: mobilizon.linux07.fr
The only strange thing is that I have to reinitialize my password, I don’t know really why ?

1 Like

hi, it is a good idea to backup all clusters before with

 <strong>```
pg_dumpall > outputfile
```</strong>

to not loose your databases of other apps.

2 Likes

Thanks a lot @Aleks @rodinux @emo ,

okay, I wanted to be cautious, so I first tried simply DROPing the DATABASE mobilizon. It worked like a charm and # LC_ALL=C pg_upgradecluster -m upgrade 11 main went with

Success. Please check that the upgraded cluster works. If it does,
you can remove the old cluster with
    pg_dropcluster 11 main

Ver Cluster Port Status Owner    Data directory              Log file
11  main    5433 down   postgres /var/lib/postgresql/11/main /var/log/postgresql/postgresql-11-main.log
Ver Cluster Port Status Owner    Data directory              Log file
13  main    5432 online postgres /var/lib/postgresql/13/main /var/log/postgresql/postgresql-13-main.log

But after that there still was an outstanding migration in the WebGUI. So I started migration again, again with “success” in the GUI.

Now I see this:

# systemctl status postgresql.service 
● postgresql.service - PostgreSQL RDBMS
     Loaded: loaded (/lib/systemd/system/postgresql.service; enabled; vendor preset: enabled)
     Active: active (exited) since Fri 2022-08-26 20:48:16 CEST; 1min 5s ago
    Process: 2971915 ExecStart=/bin/true (code=exited, status=0/SUCCESS)
   Main PID: 2971915 (code=exited, status=0/SUCCESS)
        CPU: 2ms

Aug 26 20:48:16 yunohost systemd[1]: Starting PostgreSQL RDBMS...
Aug 26 20:48:16 yunohost systemd[1]: Finished PostgreSQL RDBMS.


# pg_lsclusters
Ver Cluster Port Status Owner    Data directory              Log file
13  main    5433 down   postgres /var/lib/postgresql/13/main /var/log/postgresql/postgresql-13-main.log


# dpkg --list | grep postgis
ii  postgis                                   3.1.1+dfsg-1                                       amd64        Geographic objects support for PostgreSQL
ii  postgis-doc                               3.1.1+dfsg-1                                       all          Geographic objects support for PostgreSQL -- documentation
ii  postgresql-11-postgis-2.5                 2.5.1+dfsg-1                                       amd64        Geographic objects support for PostgreSQL 11
ii  postgresql-11-postgis-2.5-scripts         2.5.1+dfsg-1                                       all          Geographic objects support for PostgreSQL 11 -- SQL scripts
ii  postgresql-13-postgis-3                   3.1.1+dfsg-1                                       amd64        Geographic objects support for PostgreSQL 13
ii  postgresql-13-postgis-3-scripts           3.1.1+dfsg-1                                       all          Geographic objects support for PostgreSQL 13 -- SQL scripts


# less /var/log/postgresql/postgresql-13-main.log
#
(is an empty file)

In the GUI I see that service postgresql is inactive since more than 52 years, but it says it is activated on boot.
So I see that Port 5433 (should be 5432) has status down and I wonder if I killed postgresql 13 by executing the migration a second time.

Can please anyone give a hint how to fix that?
Is it save to purge the dpkg postgresql-11 packages ?

I told you I have similar problem, the solution for me was remove mobilizon but keep a backup, remove purge all postgresql,
then restore the backup of the app mobilizon, the restoration will create postgresql and upload the good packages… I think it was the solution… Also mobilizon was the only app which use postgresql for me… Now mobilizon works and there is only postgresql 13 working, I think the restoration have found the postgresql-13-postgis-3 and postgresql-13-postgis-3-scripts automatically.

1 Like

Thank you @rodinux ,
finally I’m now running YNH 11 with pq 13, too.

But since I used other apps that needs postgresql, it didn’t go that easy.
I did, what you told me, but that crashed my Invidious, Mastodon, Moodle, Peertube, Pixelfed, Pleroma and Synapse.
So after some fiddeling I ended with uninstalling all of them, purging postgresql* and then restoring the apps from backup. Well, it took me some hours but now everything is running fine again.

Thank you.

2 Likes

Hello,

la migration d’un yuno (sur Odroid N2+ armbian) 4.4.2.12 vers yuno 11 a échoué.

Ci-contre le log : https://paste.yunohost.org/raw/vikotavito

Merci d’avance bcp pour votre aide éventuelle.

Tu es sur que c’est vraiment le log complet fourni par Yunohost et pas juste les métadonnées ?

Bonsoir Aleks

Alors, je pense que c’est le log complet, i.e. je clique bien sur le bouton vert et ça donne : https://paste.yunohost.org/raw/inaqikasib

Mais en CLI ça donne un paste plus complet…désolé…https://paste.yunohost.org/raw/levidugane

Mouarf ouai effectivement y’a un truc chelou … je comprends pas pourquoi depuis la webadmin le log serait vide …

Bon par rapport au soucis de la migration: ça semble être une variante de libc6 : Breaks libgcc-8-dev mais différent, et sur Armbian … Il y a maintenant un mécanisme de tentative d’autopatch du truc mais codé seulement pour les RPi car je pensais que c’était les seuls à avoir le problème.

Bref, on peut tenter de l’appliquer manuellement:

sudo su # Pour devenir root si tu es loggé en admin

cp /var/lib/dpkg/status /root/dpkg_status.bkp

# This removes the dependency to build-essential from $app-ynh-deps
perl -i~ -0777 -pe 's/(Package: .*-ynh-deps\n(.+:.+\n)+Depends:.*)(build-essential, ?)(.*)/$1$4/g' /var/lib/dpkg/status

apt remove build-essential

LC_ALL=C DEBIAN_FRONTEND=noninteractive APT_LISTCHANGES_FRONTEND=none apt autoremove --assume-yes

apt remove gcc-8 libgcc-8-dev

(Peut-être qu’un des apt remove ne va pas marcher, dans ce cas là on creuse)

35 posts were split to a new topic: Bullseye migration stuck on “The following packages have been kept back”

Hello,
I’ve launched the migration from YNH4.4 to YNH 11 on a lxc container running on proxmox 7.2-7
This ynh instance only runs nextcloud & onlyoffice
no problem in the migration logs but after a reboot, nextcloud is unable to work, because redis-server is unable to start.
Here are the logs hastebin
Thanks for your help

@Benance : yeah some people reported similar issues with Proxmox:

Aug 30 14:12:06 systemd[597]: redis-server.service: Failed at step NAMESPACE spawning /usr/bin/redis-server: Permission denied
Aug 30 14:12:06 systemd[1]: redis-server.service: Main process exited, code=exited, status=226/NAMESPACE

the fix is to enable the “Nesting” stuff for your LXC (not sure how to do this in Proxmox exactly tho). Edit: In LXD this is called security.nesting=true

1 Like

All right
Thanks for your help.
I’ve toggled the nesting stuff in the lxc options and after a new reboot, the redis-server (and nextcloud by the way) is running again.
Thanks again!
2022-08-30 14_52_31-virtu - Proxmox Virtual Environment

1 Like

Hello @Aleks Merci de ta réponse,

les deux apt remove n’ont pas marché

# apt remove build-essential                                                                                 :( 100 22-08-30 - 22:09:29
Lecture des listes de paquets... Fait
Construction de l'arbre des dépendances       
Lecture des informations d'état... Fait
Vous pouvez lancer « apt --fix-broken install » pour corriger ces problèmes.
Les paquets suivants contiennent des dépendances non satisfaites :
 armbian-config : Dépend: build-essential mais ne sera pas installé
 binutils : Dépend: binutils-common (= 2.31.1-16) mais 2.35.2-2 devra être installé
            Dépend: libbinutils (= 2.31.1-16) mais 2.35.2-2 devra être installé
            Dépend: binutils-aarch64-linux-gnu (= 2.31.1-16) mais 2.35.2-2 devra être installé
 gcc-8 : Dépend: libgcc-8-dev (= 8.3.0-6) mais il n'est pas installable
 libc-dev-bin : Dépend: libc6 (< 2.29) mais 2.31-13+deb11u3 devra être installé
 libc6 : Casse: libc6:armhf (!= 2.31-13+deb11u3) mais 2.28-10+deb10u1 devra être installé
 libc6:armhf : Casse: libc6 (!= 2.28-10+deb10u1) mais 2.31-13+deb11u3 devra être installé
 libc6-dev : Dépend: libc6 (= 2.28-10+deb10u1) mais 2.31-13+deb11u3 devra être installé
 libstdc++-8-dev : Dépend: libgcc-8-dev (= 8.3.0-6) mais il n'est pas installable
 mailutils : Dépend: mailutils-common (= 1:3.5-4) mais 1:3.10-3 devra être installé
             Dépend: libmailutils5 mais il n'est pas installable
 openssh-sftp-server : Dépend: openssh-client (= 1:7.9p1-10+deb10u2)
 unscd : Dépend: libc6 (< 2.29) mais 2.31-13+deb11u3 devra être installé
E: Dépendances non satisfaites. Essayez « apt --fix-broken install » sans paquet


# apt remove gcc-8 libgcc-8-dev                                                                              :( 100 22-08-30 - 22:10:29
Lecture des listes de paquets... Fait
Construction de l'arbre des dépendances       
Lecture des informations d'état... Fait
Le paquet « libgcc-8-dev » n'est pas installé, et ne peut donc être supprimé
Vous pouvez lancer « apt --fix-broken install » pour corriger ces problèmes.
Les paquets suivants contiennent des dépendances non satisfaites :
 binutils : Dépend: binutils-common (= 2.31.1-16) mais 2.35.2-2 devra être installé
            Dépend: libbinutils (= 2.31.1-16) mais 2.35.2-2 devra être installé
            Dépend: binutils-aarch64-linux-gnu (= 2.31.1-16) mais 2.35.2-2 devra être installé
 g++ : Dépend: gcc-8 (>= 8.3.0-1~) mais ne sera pas installé
 g++-8 : Dépend: gcc-8 (= 8.3.0-6) mais ne sera pas installé
 gcc : Dépend: gcc-8 (>= 8.3.0-1~) mais ne sera pas installé
 libc-dev-bin : Dépend: libc6 (< 2.29) mais 2.31-13+deb11u3 devra être installé
 libc6 : Casse: libc6:armhf (!= 2.31-13+deb11u3) mais 2.28-10+deb10u1 devra être installé
 libc6:armhf : Casse: libc6 (!= 2.28-10+deb10u1) mais 2.31-13+deb11u3 devra être installé
 libc6-dev : Dépend: libc6 (= 2.28-10+deb10u1) mais 2.31-13+deb11u3 devra être installé
 libstdc++-8-dev : Dépend: libgcc-8-dev (= 8.3.0-6) mais il n'est pas installable
 mailutils : Dépend: mailutils-common (= 1:3.5-4) mais 1:3.10-3 devra être installé
             Dépend: libmailutils5 mais il n'est pas installable
 openssh-sftp-server : Dépend: openssh-client (= 1:7.9p1-10+deb10u2)
 unscd : Dépend: libc6 (< 2.29) mais 2.31-13+deb11u3 devra être installé
E: Dépendances non satisfaites. Essayez « apt --fix-broken install » sans paquet

Hi @Benance , could you please make a tuto for installing yunohost in proxmox. I’ve not seen any thread describing it in the forum.
Thanks

1 Like

BIG THANKS FOR THIS!!!

BEST REGARDS :heart:

Hi @jarod5001
No need tuto to achieve this.
Here is what I’ve done:
Install a debian 10 lxc (available from the lxc templates in proxmox)
Install yunohost on top of this existing debian container
Enjoy!

That’s all

2 Likes

Migration failed due to low disk space: https://paste.yunohost.org/raw/zelucasotu

I had 40GiB of free space, but the PG11 database is around 60GiB (mainly thanks to Matrix).

I suggest checking beforehand that there is more free space than the current DB if it needs double the space at some point:

[[ $(du -s /var/lib/postgresql/11 |cut -f 1) -lt $(df /var/lib/postgresql/ |tail -n 1 | tr -s " " |cut -f 4 -d " ") ]] && echo OK || echo Not enough disk space