[06:08:51]
<florent> Luc (he/him): Thanks for sharing that. Can you also paste the result of this commands, which prints `borg.timer`:
```
sudo systemctl cat borg.timer
```
It should be very few lines, this is just the timer.
[10:41:42]
<Michele Agostinelli> Anyone knows how to add more model/languages to a libretranslate installation? The script install_models.py freeze aftet launching
[11:22:23]
<tituspijean[m]> (I don't know about libretranslate specifically) How do you run it?
[11:39:51]
<someuser> I would also consider fixing the issue with the Python script. If it doesn't throw any kind of error, maybe its a timeout?
[11:40:27]
<someuser> But a timeout should throw an exception or log a trace too
[12:01:16]
<Michele Agostinelli> > <@tituspijean:matrix.org> (I don't know about libretranslate specifically) How do you run it?
Official yunohost package
[12:01:49]
<Michele Agostinelli> > <@someuser:matrix.woolh.at> But a timeout should throw an exception or log a trace too
Nope, the script remain running like downloading...
[12:50:34]
<tituspijean[m]> I definitely hope that you use the package, but how do you start that python script?
[12:51:33]
<Michele Agostinelli> yunohost app shell libretranslate
[13:06:56]
<tituspijean[m]> ok thank you! That's the way to do it. I'll investigate tonight on my own server.
[13:12:37]
<Tio> question, shouldnt `yunohost backup restore <archive_name> --app <appname>` work? I get
```
Traceback (most recent call last):
File "/usr/bin/yunohost", line 25, in <module>
import yunohost
ModuleNotFoundError: No module named 'yunohost'
```
[13:12:49]
<Aleks (he/him/il/lui)> 🤔
[13:13:01]
<Aleks (he/him/il/lui)> well does `yunohost --version` works ?
[13:13:36]
<Tio> nope... hm very weird
[13:14:38]
<Tio> ok closed the ssh connection and reconnected and works now
[13:14:43]
<Tio> i migrated yunohost to another server so who knows 😄
[13:14:55]
<Tio> good that is just that 😄
[13:31:06]
<Aleks (he/him/il/lui)> did something in particular happen before you ran into this situation ? What's the last major operation you did ? @_@
[14:07:49]
<Tio> i was restoring backups ...
[14:07:59]
<Tio> thats what I remember
[14:09:18]
<Tio> by the way if I triggered a restore from the YNH panel and it is running, is it ok to close the window? Like is the process always going to run regardless if you keep the window with the YNH panel open?
[14:09:51]
<Tio> I have a huge friendica backup restoring and it is taking ages for 80GB of data....and I need to go somewhere 😄 so I was wondering
[14:10:51]
<Aleks (he/him/il/lui)> yes
[14:11:21]
<Tio> ok so it is safe to close the window for example. great
[14:11:28]
<Tio> thx
[14:14:40]
<Aleks (he/him/il/lui)> yes, that was already the case before 12.1 though after you had no real way to check what was going on on the server and the UI would basically be frozen / blank, but now if you reopen the webadmin later it should catch up and you can see new log mesages and wether or not it's still going on
[14:55:53]
<miro5001> > <@Alekswag:matrix.org> yes, that was already the case before 12.1 though after you had no real way to check what was going on on the server and the UI would basically be frozen / blank, but now if you reopen the webadmin later it should catch up and you can see new log mesages and wether or not it's still going on
It would be great to have such improvement in ssh 😉
[15:04:04]
<Aleks (he/him/il/lui)> hmyeah i think somebody recently was suggesting we could have some script executed in bash_profile.sh-whatever (the script ran when you open a shell) to check the current command and we could point to the `tail -f /var/log...` command or have `yunohost log follow` or something)
[15:24:10]
<tituspijean[m]> To be frank, my experience so far is that the webadmin still freezes, but the pacman message states that maybe yunohost-api crashed. 😅 (I get this with "big" apps like Immich, Home Assistant, Grist)
[15:26:21]
<Aleks (he/him/il/lui)> then it should be repawted as a bug 😭
[15:26:34]
<Aleks (he/him/il/lui)> with ideally instructions to reproduce
[15:27:53]
<Aleks (he/him/il/lui)> but yeah the other day we were discussing that maybe sometime the API process is CPU-starved and there's no CPU left for the SSE thread, something along those line idk
[15:28:58]
<Aleks (he/him/il/lui)> i mean if the main yunohost-api process is doing CPU intenstive stuff ... but that'd be interesting to know if that also happens when building an app like the process eating the CPU be like `npm` or `python` etc
[15:32:26]
<miro5001> > <@Alekswag:matrix.org> with ideally instructions to reproduce
A slow hardware can replicate it.
[15:34:31]
<miro5001> > <@Alekswag:matrix.org> hmyeah i think somebody recently was suggesting we could have some script executed in bash_profile.sh-whatever (the script ran when you open a shell) to check the current command and we could point to the `tail -f /var/log...` command or have `yunohost log follow` or something)
https://github.com/Xfennec/progress
But written in C 😁
[16:23:00]
<Melchisedech> Hi, I use borg to backup my Yunohost installation. I tried to extract a very big backup (> 3TB) and after more than 36 hours, the process failed, resulting in an incomplete and unusable tar file. I used the commands provided in the app description ( https://github.com/YunoHost-Apps/borg_ynh/blob/master/doc/ADMIN.md ). I tried to see the logs in /var/log/borg but there were only logs from my daily automated backups. Do the manual commands generate any log ? If yes, does anyone know where they are located?
[16:48:10]
<Aleks (he/him/il/lui)> 3TB @_@
[16:52:26]
<Melchisedech> Yes, mainly multimedia files for my Jellyfin server.
[16:54:13]
<·☽•Nameless☆•777 · ±> You should definitely try with journalctl.
( To read the borg logs )
[16:54:27]
<Melchisedech> Also, a lot of images and videos from Nextcloud, synced from various phones
[16:54:59]
<Melchisedech> Thanks ·☽•Nameless☆•777 · ± i’ll try that right away.
[17:06:14]
<Tio> for such huge archives I recommend you first restore the app without the data with something like this
```
app=borg
PATH="/var/www/$app/venv/bin/:$PATH"
export BORG_PASSPHRASE="$(sudo yunohost app setting $app passphrase)"
export BORG_RSH="ssh -i /root/.ssh/id_${app}_ed25519 -oStrictHostKeyChecking=yes"
repository="$(sudo yunohost app setting $app repository)"
borg export-tar -e apps/nextcloud/backup/home/yunohost.app "$repository::auto_nextcloud-2025-09-15T00:25:34" /home/yunohost.backup/archives/auto_nextcloud-2025-09-15T00:25:34.tar
```
then restore the data
```
app=borg
PATH="/var/www/$app/venv/bin/:$PATH"
export BORG_PASSPHRASE="$(sudo yunohost app setting $app passphrase)"
export BORG_RSH="ssh -i /root/.ssh/id_${app}_ed25519 -oStrictHostKeyChecking=yes"
repository="$(sudo yunohost app setting $app repository)"
borg extract "$repository::auto_peertube-2025-09-14T01:12:00" apps/peertube/backup/home/yunohost.app/
mv apps/peertube/backup/home/yunohost.app/peertube ./
rm -r apps
```
of adapt the code for your app
[17:06:21]
<Tio> for such huge archives I recommend you first restore the app without the data with something like this
```
app=borg
PATH="/var/www/$app/venv/bin/:$PATH"
export BORG_PASSPHRASE="$(sudo yunohost app setting $app passphrase)"
export BORG_RSH="ssh -i /root/.ssh/id_${app}_ed25519 -oStrictHostKeyChecking=yes"
repository="$(sudo yunohost app setting $app repository)"
borg export-tar -e apps/nextcloud/backup/home/yunohost.app "$repository::auto_nextcloud-2025-09-15T00:25:34" /home/yunohost.backup/archives/auto_nextcloud-2025-09-15T00:25:34.tar
```
then restore the data
```
app=borg
PATH="/var/www/$app/venv/bin/:$PATH"
export BORG_PASSPHRASE="$(sudo yunohost app setting $app passphrase)"
export BORG_RSH="ssh -i /root/.ssh/id_${app}_ed25519 -oStrictHostKeyChecking=yes"
repository="$(sudo yunohost app setting $app repository)"
borg extract "$repository::auto_peertube-2025-09-14T01:12:00" apps/peertube/backup/home/yunohost.app/
mv apps/peertube/backup/home/yunohost.app/peertube ./
rm -r apps
```
ofc adapt the code for your app
[17:11:38]
<Melchisedech> Thanks Tio I’ll try that. ·☽•Nameless☆•777 · ± , I’ve got nothing about the manual borg commands i launched in journalctl. Maybe because I didn’t use sudo but sudo su
[17:14:26]
<·☽•Nameless☆•777 · ±> > <@melchisedech:matrix.org> Thanks Tio I’ll try that. ·☽•Nameless☆•777 · ± , I’ve got nothing about the manual borg commands i launched in journalctl. Maybe because I didn’t use sudo but sudo su
Perhaps you should try a command such as `` journalctl -u borg.service ``
[17:17:17]
<Melchisedech> I’m more the journalctl -b | grep borg kind of guy.
[17:20:39]
<n4ta> Howdy everyone! Though this issue isn't strictly a Yunohost thing, I figured I'd ask here since maybe someone would know. I just finished making a tunnel (-ish) with the Wireguard-Client Yunohost package, and it works flawlessly! However, now on android, it seems like `.local` tld-having domains don't work. From my research, this is due to Android just not supporting these, due to a lack of mDNS (?). So I wanted to ask this chat what is the best course of action:
- Is there perhaps a private IPv4 associated to the `.local` domain that I could use instead?
- Is there a way I could get Android to recognize these URLs?
- Is there, perhaps, somewhere in the Wireguard DNS settings where I could even maybe fix this?
All help is more than appreciated, hope everyone has a nice day! :]
[17:26:48]
<·☽•Nameless☆•777 · ±> > <@melchisedech:matrix.org> I’m more the journalctl -b | grep borg kind of guy.
comme je n'utilise pas borg, je viens de jeter un oeil à la doc de borg et j'ai l'impression qu'il ne faut pas se servir de journalctl mais de cette commande
`` borg create repo::archive myfiles 2>> logfile ``
[17:26:49]
<·☽•Nameless☆•777 · ±> > https://borgbackup.readthedocs.io/en/stable/usage/general.html#logging
[17:44:00]
<Luc (he/him)> @florent:matrix.skeptikon.fr
sudo systemctl cat borg.timer
# /etc/systemd/system/borg.timer
[Unit]
Description=Run backup borg regularly
[Timer]
OnCalendar=Daily
[Install]
WantedBy=timers.target
[18:29:37]
<ljf> I agree with you, your method is described on the docs https://github.com/YunoHost-Apps/borg_ynh/blob/master/doc/ADMIN.md#restoring-the-sourceconfig-of-the-app-and-its-data-separately
[18:41:01]
<n4ta> Oh, I guess that's one way to view it. Thanks, I'm going to research on that topic and see what I can find!
[19:13:20]
<Melchisedech> Thanks all, I will try that next weekend. ljf Tio I did not separate the app and the data, because this is specifically the data backup I’m trying to restore (my main HDD has some bad sectors).
[19:16:58]
<Melchisedech> But I guess I can try to restore it one big folder at a time.
[19:18:41]
<·☽•Nameless☆•777 · ±> If I understand you correctly, I would say that it is because the mDNS protocol cannot be transported by Wireguard*.
[19:22:54]
<n4ta> Update: No update... Every single place I find points to mDNS not being possible to be transmitted over (just) WireGuard and all claimed solutions didn't help either... Basically, everything points to having to install some alternative software. Thank you guys anyways!
[19:47:06]
<tituspijean[m]> @n4ta:matrix.org yes, unfortunately the mDNS needs a network that allows broadcasts, which is not possible with wireguard
[19:49:56]
<tituspijean[m]> A workaround would be to use internal domains with your server as DNS.
[19:50:36]
<tituspijean[m]> I had drafted a tutorial for an intranet with Tailscale... I should finish it...
[19:52:24]
<tituspijean[m]> @magostinelli:fedimedia.it:
```
yunohost app shell libretranslate
libretranslate@test:~$ ./scripts/install_models.py
Updating language models
Found 96 models
Downloading Albanian → English (1.9) ...
Downloading Arabic → English (1.0) ...
Downloading Azerbaijani → English (1.5) ...
Downloading Basque → English (1.9) ...
Downloading Bengali → English (1.9) ...
Downloading Bulgarian → English (1.9) ...
Downloading Catalan → English (1.7) ...
(and going...)
```
[19:52:29]
<tituspijean[m]> https://aria.im/_bifrost/v1/media/download/AQB2ki0X4_ygNqoeMC5jefz0DExRMbgs6bfPiLB4R7WxO_YEyyBq-zMLD_i0KCCgeVfFuwzZSBFvADz1eDRUVv9CeZVBX0iQAG1hdHJpeC5vcmcvQmZjUE5ySFRqUWJxY1FTV2xhUUd4eU1u
[19:52:59]
<tituspijean[m]> though the service seems to ignore the list of models to actually display (in the .env file)
[19:54:22]
<tituspijean[m]> @magostinelli:fedimedia.it you got absolutely no output from the command?
[19:54:43]
<florent> I wonder if there exist spaces after or before "on\_calendar".
Could you also share `echo "on_calendar='$(yunohost app setting borg on_calendar)'"`?
Also I see the upgrade has failed. Does it mean that the backup prior to the upgrade has been restored?
[20:31:29]
<ljf> SO your restore is in 2 steps
[20:32:58]
<ljf> You can create a tar with the app (without data) BUT with 3TB you should just export and rsync (or borg extract directly on the new server after reconnect the borg repo) the 3TB data
[20:44:39]
<n4ta> Ooh, that sounds like just what I need! If it isn't too much of a bother, could I politely ask for a slight nudge in the direction of where to find more info on how to do that? I don't want to give *too much trouble* though, I'm just kind of in the dark for the most part 😅
[20:48:21]
<tituspijean[m]> @n4ta:matrix.org try setting `10.10.10.0` as your DNS in the wireguard configuration (yes that's a real IP, in that context it's the server's)
[20:48:42]
<tituspijean[m]> you might need to download the configuration files again after that tweak
[20:49:01]
<tituspijean[m]> (or basically check that there is a DNS directive in your clients' configuration files for wireguard)
[20:49:42]
<tituspijean[m]> but you really should not use .local domains for that.
[21:05:46]
<n4ta> I added it in both the Wireguard_Client and my phone, but no avail (maybe I'm misinterpreting?)
Either way, I've been avoiding regular domains for far too long when every single page tells me to use those instead for convenience, so I might switch to those instead. Either way, thank you so much for your time and I hope you have a great day (or night)! :]