[02:17:58]
<teslas_moustache (He/They)> Hello. I'm trying to install Debian/Yunohost on a Terramaster NAS and it's not going well.
What's strange is that I installed Debian 12, but any Debian 11 image I try boots to a grub prompt.
[02:18:01]
<teslas_moustache (He/They)> Same problem with the Yunohost image itself
[08:14:11]
<nicofrand> Hi there! I installed wordpress flawlessly (<3) and I am trying to setup a SMTP account through a wordpress plugin, but the authentication fails everytime. I am using an external smtp but the domain is the same as the one on which YNH is installed. Does anyone see a reason why the SMTP auth would fail or where to look? Roundcube works fine.
[11:21:53]
<Paprika> Hello, recently I tried using the Redirect app to expose a subdomain to a machine in the local network. I selected the nginx proxy option, set the target to http://192.168.1.155/ and installed the app. However, that more or less does nothing - accessing the subdomain via browser tells me that there's no connection.
Anyone faced anything similar?
[11:22:24]
<Paprika> Hello, recently I tried using the Redirect app to expose a subdomain to a machine in the local network. I selected the nginx proxy option, set the target to http://192.168.1.155/ and installed the app on that subdomain. However, that more or less does nothing - accessing the subdomain via browser tells me that there's no connection.
Anyone faced anything similar?
[12:28:11]
<tonton> Hi, I installed conduit but I can't seem to login. I'm assuming it doesn't use LDAP or PAM so I need to create users?
So I'm also wondering, does dendrite or synapse to get users from LDAP (or PAM)?
[12:45:03]
<orhtej2> > <@tonton:envs.net> Hi, I installed conduit but I can't seem to login. I'm assuming it doesn't use LDAP or PAM so I need to create users?
> So I'm also wondering, does dendrite or synapse to get users from LDAP (or PAM)?
Dendrite does not support ldap upstream, synapse seems to have ldap properly configured
[13:01:49]
<tonton> orhtej2: Awesome, I remember reading synapse had issues with performance - maybe 2 years ago - does it work well on modest hardware today, do you know?
[13:04:51]
<Mateusz Szymański> > <@tonton:envs.net> orhtej2: Awesome, I remember reading synapse had issues with performance - maybe 2 years ago - does it work well on modest hardware today, do you know?
it's still a resource hog :/
[13:05:32]
<Mateusz Szymański> `dendrite` provides CLI for creating users and it's a one-time operation so perhaps that's a good bet?
[13:05:47]
<Mateusz Szymański> OTOH Synapse is the only actively developed server as far as I can tell
[13:08:39]
<tonton> ah. Thank you. I'll have to think a bit then.... X(
[13:18:31]
<Paprika> Update: I think I realize this might be due to SSL not being present on the local machine that I'm redirecting to. Basically, I'm redirecting from https to http, which I assume might not work out of the box. If anyone has some insight to point me in the right direction to set this all up, that would be great. I've also thought of hosting the internal site that I have on the local machine on ynh instead, using the my\_webapp app (site uses php). However, that just gives me a 504 timeout when trying to connect to the local MySQL server that I have on that other machine.
[13:31:35]
<Mateusz Szymański> > <@botagiuks:tiesiog.lt> Update: I think I realize this might be due to SSL not being present on the local machine that I'm redirecting to. Basically, I'm redirecting from https to http, which I assume might not work out of the box. If anyone has some insight to point me in the right direction to set this all up, that would be great. I've also thought of hosting the internal site that I have on the local machine on ynh instead, using the my\_webapp app (site uses php). However, that just gives me a 504 timeout when trying to connect to the local MySQL server that I have on that other machine.
That's actually correct as NGINX will handle SSL termination in this case
[13:32:20]
<Mateusz Szymański> MySQL is not and should not listen on 0.0.0.0, hence it's unreachable from the outside
[13:34:05]
<Mateusz Szymański> reverse_proxy mode should work for the case you're describing, any NGINX logs or logs from network tab in developer tools that would indicate what's failing exactly?
[13:36:34]
<Mateusz Szymański> is the app running on IP you're using and is said IP reachable from your YNH box?
[13:41:03]
<Paprika> The user that I have setup on the local machine in MySQL has been granted access from 192.168.1.0/24. In that case, at the very least, I should be able to utilize the redirect in the local network. However, that just gives me a connection failed in the browser (if I try to access the subdomain from the local network). If do it from outside the network, I get the SSL error. And even then, from my understanding, I don't need to grant access on 0.0.0.0 for the MySQL user, at least I would think so.
When I was trying to access the subdomain from outside, this is the error I noticed in the nginx logs:
`[...] upstream timed out (110: Connection timed out) while reading response header from upstream, client: x, server: domain.tld, request: "GET /path/to/index.php HTTP/2.0", upstream: "fastcgi://unix:/var/run/php/php7.4-fpm-my_webapp.sock", host: "domain.tld"`
When testing whether the local machine is reachable from YNH box, I tested with `ping [localmacineIP]`, which successfully pinged the local machine from the YNH box.
[13:41:43]
<Paprika> The user that I have setup on the local machine in MySQL has been granted access from 192.168.1.0/24. In that case, at the very least, I should be able to utilize the redirect in the local network. However, that just gives me a connection failed in the browser (if I try to access the subdomain from the local network). If do it from outside the network, I get the SSL error. And even then, from my understanding, I don't need to grant access on 0.0.0.0 for the MySQL user, just on the IP that the YNH box is on (I believe).
When I was trying to access the subdomain from outside, this is the error I noticed in the nginx logs:
`[...] upstream timed out (110: Connection timed out) while reading response header from upstream, client: x, server: domain.tld, request: "GET /path/to/index.php HTTP/2.0", upstream: "fastcgi://unix:/var/run/php/php7.4-fpm-my_webapp.sock", host: "domain.tld"`
When testing whether the local machine is reachable from YNH box, I tested with `ping [localmacineIP]`, which successfully pinged the local machine from the YNH box.
[13:43:15]
<Paprika> I also tried to check MySQL error logs, but they are empty.
Also, I have no idea whether the my_webapp app uses apache2 or what, so I don't really know where to look for logs for that app itself. 🤔
[13:44:12]
<Paprika> The user that I have setup on the local machine in MySQL has been granted access from 192.168.1.0/24. In that case, at the very least, I should be able to utilize database from within the local network. However, when I try to access the subdomain that redirects it to the local machine, that just gives me a connection failed in the browser (if I try to access the subdomain from the local network). If do it from outside the network, I get the SSL error. And even then, from my understanding, I don't need to grant access on 0.0.0.0 for the MySQL user, just on the IP that the YNH box is on (I believe).
When I was trying to access the subdomain from outside, this is the error I noticed in the nginx logs:
`[...] upstream timed out (110: Connection timed out) while reading response header from upstream, client: x, server: domain.tld, request: "GET /path/to/index.php HTTP/2.0", upstream: "fastcgi://unix:/var/run/php/php7.4-fpm-my_webapp.sock", host: "domain.tld"`
When testing whether the local machine is reachable from YNH box, I tested with `ping [localmacineIP]`, which successfully pinged the local machine from the YNH box.
[13:46:48]
<Paprika> Just to clarify, in the firewall, I do allow traffic from 192.168.1.0/24 to any port 80 and 443 on the local machine. I am not sure if I need to have them open to 0.0.0.0.
[13:49:15]
<orhtej2> > <@botagiuks:tiesiog.lt> The user that I have setup on the local machine in MySQL has been granted access from 192.168.1.0/24. In that case, at the very least, I should be able to utilize database from within the local network. However, when I try to access the subdomain that redirects it to the local machine, that just gives me a connection failed in the browser (if I try to access the subdomain from the local network). If do it from outside the network, I get the SSL error. And even then, from my understanding, I don't need to grant access on 0.0.0.0 for the MySQL user, just on the IP that the YNH box is on (I believe).
>
> When I was trying to access the subdomain from outside, this is the error I noticed in the nginx logs:
> `[...] upstream timed out (110: Connection timed out) while reading response header from upstream, client: x, server: domain.tld, request: "GET /path/to/index.php HTTP/2.0", upstream: "fastcgi://unix:/var/run/php/php7.4-fpm-my_webapp.sock", host: "domain.tld"`
>
> When testing whether the local machine is reachable from YNH box, I tested with `ping [localmacineIP]`, which successfully pinged the local machine from the YNH box.
Mysql is not listening on external ips, only on 127.0.0.1
[13:50:13]
<orhtej2> What if you curl/wget the ip from redirect box?
[14:04:07]
<Paprika> If I `curl -i sub.domain.tld` from YNH, I get `curl: (6) Could not resolve host: sub.domain.tld`
[14:11:48]
<Paprika> To clarify on the situation itself, that's what I have currently:
1. The local machine I speak of has some servers running. Those servers push data to the MySQL server that runs on that local machine as well. For testing purposes, I also have apache2 running, which serves those .php pages. They do work and display everything correctly, but that's because it's all running on the same machine.
2. I also have a YNH container that's on the same local network. They can see each other via ping.
What I wanted to do were two options:
1. Have my\_webapp serve the .php pages that would access the local MySQL server on the other machine. That would be the easiest option as I wouldn't have to worry about HTTPS, SSL, domains and so on.
2. Have a subdomain.domain.tld redirect to the internal site on the local machine.
With the first option, I get 504 Gateway Time-out by Nginx.
With the second option, I either cannot resolve the host / can't establish a connection via browser, or get an SSL error if I access the subdomain from the outside.
Overall, I am fine with either option, it's just that I cannot get it to work.
[14:17:20]
<Paprika> To clarify on the situation itself, that's what I have currently:
1. The local machine I speak of has some servers running. Those servers push data to the MySQL server that runs on that local machine as well. For testing purposes, I also have apache2 running, which serves those .php pages. They do work and display everything correctly, but that's because it's all running on the same machine.
2. I also have a YNH container that's on the same local network. They can see each other via ping.
What I wanted to do were two options:
1. Have my\_webapp serve the .php pages that would access the local MySQL server on the other machine. That would be the easiest option as I wouldn't have to worry about HTTPS, SSL, domains and so on.
2. Have a subdomain.domain.tld redirect to the internal site on the local machine.
With the first option, I get 504 Gateway Time-out by Nginx (php timeout, but why is it timing out?)
With the second option, I either cannot resolve the host / can't establish a connection via browser, or get an SSL error if I access the subdomain from the outside.
Overall, I am fine with either option, it's just that I cannot get it to work.
[14:18:25]
<Paprika> To clarify on the situation itself, that's what I have currently:
1. The local machine I speak of has some servers running. Those servers push data to the MySQL server that runs on that local machine as well. For testing purposes, I also have apache2 running, which serves those .php pages. They do work and display everything correctly, but that's because it's all running on the same machine.
2. I also have a YNH container that's on the same local network. They can see each other via ping.
What I wanted to do were two options:
1. Have my\_webapp serve the .php pages that would access the local MySQL server on the other machine. That would be the easiest option as I wouldn't have to worry about HTTPS, SSL, domains and so on.
2. Have a subdomain.domain.tld redirect to the internal site on the local machine.
With the first option, I get 504 Gateway Time-out by Nginx (php timeout, but why is it timing out?)
With the second option, I either cannot resolve the host / can't establish a connection via browser, or get an SSL error if I access the subdomain from the outside.
Overall, I am fine with either option, it's just that I cannot get it to work. Maybe I'd prefer the first one as this isn't going to be used for long, more or less a one time thing, but I cannot figure out why is it timing out, and I have no idea where to look for the reason of the time out.
[14:23:45]
<Mateusz Szymański> > <@botagiuks:tiesiog.lt> To clarify on the situation itself, that's what I have currently:
>
> 1. The local machine I speak of has some servers running. Those servers push data to the MySQL server that runs on that local machine as well. For testing purposes, I also have apache2 running, which serves those .php pages. They do work and display everything correctly, but that's because it's all running on the same machine.
> 2. I also have a YNH container that's on the same local network. They can see each other via ping.
>
> What I wanted to do were two options:
>
> 1. Have my\_webapp serve the .php pages that would access the local MySQL server on the other machine. That would be the easiest option as I wouldn't have to worry about HTTPS, SSL, domains and so on.
> 2. Have a subdomain.domain.tld redirect to the internal site on the local machine.
>
> With the first option, I get 504 Gateway Time-out by Nginx (php timeout, but why is it timing out?)
> With the second option, I either cannot resolve the host / can't establish a connection via browser, or get an SSL error if I access the subdomain from the outside.
>
> Overall, I am fine with either option, it's just that I cannot get it to work. Maybe I'd prefer the first one as this isn't going to be used for long, more or less a one time thing, but I cannot figure out why is it timing out, and I have no idea where to look for the reason of the time out.
To clarify:
- you have some website running on http://192.168.1.155/ which works correctly when you're inside LAN and type said URL to browser
- you have YNH box available from the outside pointed to by sub.domain.tld
- when on YNH box you can `curl -L http://192.168.1.155/ ` and see web page contents
- when you type `sub.domain.tld` into browser you get HTTP 504? This is where redirect_ynh is installed that should proxy_pass to http://192.168.1.155/ , right?
[14:24:41]
<Mateusz Szymański> points 1-3 are prerequisites for point 4 to work as expected
[14:26:40]
<Paprika> As always, I was just dumb.
For the option 1 to work, I was supposed to open port 3306 on the local machine for YNH IP. I just forgot to do that.
A simple `ufw allow from YNH_IP to any port 3306` fixed the issue.
[14:29:33]
<pepecyb> A question about the custom webapp "my_webapp":
Is it possible to have multiple installations of my_webapp? The problem I see is that the installation of my_webapp is done under /var/www/my_webapp. However, this would lead to conflicts with a further installation. When installing with the web admin centre, however, you cannot select the destination of the web folder.
Is this possible with the cli installation? And if so, what would the corresponding argument (different path) be?
[14:36:34]
<Mateusz Szymański> > <@pepecyb:matrix.org> A question about the custom webapp "my_webapp":
>
> Is it possible to have multiple installations of my_webapp? The problem I see is that the installation of my_webapp is done under /var/www/my_webapp. However, this would lead to conflicts with a further installation. When installing with the web admin centre, however, you cannot select the destination of the web folder.
>
> Is this possible with the cli installation? And if so, what would the corresponding argument (different path) be?
yes, the app [does support multiple installations](https://github.com/YunoHost-Apps/my_webapp_ynh/blob/9defce59be9422a43b8684556a5dfa71589204a8/manifest.toml#L19), next instances will be called something like `my_webapp__2`. The destination folder is automatically assigned, should not affect app in any observable way assuming you enable SFTP access and use that for managing files served. Why are you concerned about the folder that serves the app?
[15:08:18]
<Paprika> > <@orhtej2:matrix.org> To clarify:
> - you have some website running on http://192.168.1.155/ which works correctly when you're inside LAN and type said URL to browser
> - you have YNH box available from the outside pointed to by sub.domain.tld
> - when on YNH box you can `curl -L http://192.168.1.155/ ` and see web page contents
> - when you type `sub.domain.tld` into browser you get HTTP 504? This is where redirect_ynh is installed that should proxy_pass to http://192.168.1.155/ , right?
Just to answer to your questions as my redirect_ynh attempt was unsuccessful, but I went the option 1 way, you are correct for points 1-4. However, for point 5, it was that if I tried to access `sub.domain.tld`, it would end up simply saying that the host cannot be resolved if I used `curl -i sub.domain.tld`, while in the browser I wouldn't be able to establish a connection when accessing that subdomain. I would like to say that I have any idea why it didn't work, but I can't. If it wasn't for my_webapp, I would be still bugging my head around how to do this redirection differently to make it work.
[15:10:50]
<Paprika> However, this situation did raise some questions. For one, I wasn't able to find out how my\_webapp works. I assumed it would use apache2 or lighttpd, but I couldn't find any logs for these services, which is why it was so cumbersome to try and pin point the issue when accessing the .php files. On a local linux container that I have, I simply have lighttpd or apache2 running and I can just find the logs in /var/log/apache2, for instance. Naturally, that's where I searched for logs in YNH.
[15:11:07]
<pepecyb> Mateusz Szymański: Aaaah, thank you very much! I hadn't found that multiple installations are possible. Great, then I can start a new "project" when I get the chance. 🙂 👍️
[15:17:55]
<Mateusz Szymański> YNH is all about NGINX so that's what's serving your `my_webapp`, the logs are in `/var/log/nginx/domain.tld-xxxx.log`
[15:48:48]
<teslas_moustache (He/They)> > Hello. I'm trying to install Debian/Yunohost on a Terramaster NAS and it's not going well.
>
> What's strange is that I installed Debian 12, but any Debian 11 image I try boots to a grub prompt.
Sorry. I guess I have two questions.
1) since I can only seem to get Debian 12 to install, does Yunohost work *at all* with 12? If not, when might we expect 12 support?
2) does anyone have any thoughts on what I could try to get Debian 11 installed?
[18:15:45]
<orhtej2> > Sorry. I guess I have two questions.
> 1) since I can only seem to get Debian 12 to install, does Yunohost work *at all* with 12? If not, when might we expect 12 support?
> 2) does anyone have any thoughts on what I could try to get Debian 11 installed?
Cf https://forum.yunohost.org/t/alpha-stage-testing-for-yunohost-12-0-on-debian-bookworm-but-not-yet-for-the-bullseye-bookworm-migration/28188/1
[21:24:36]
<teslas_moustache (He/They)> > Cf https://forum.yunohost.org/t/alpha-stage-testing-for-yunohost-12-0-on-debian-bookworm-but-not-yet-for-the-bullseye-bookworm-migration/28188/1
I see. So installing YNH on a fresh Debian 12 install pretty much works, it's the *migration* from 11 to 12 that has major problems to work out?
[21:25:21]
<Aleks (he/him/il/lui)> not just this, DKIM-signing needs to be fixed, some apps do not work yet on 12.x, and there may be other undiscovered issues because this is alpha-stage
[21:26:44]
<orhtej2> > I see. So installing YNH on a fresh Debian 12 install pretty much works, it's the *migration* from 11 to 12 that has major problems to work out?
I have no experience running YNH 12 in 'prod' env so IDK, tread with caution I guess
[21:26:44]
<Aleks (he/him/il/lui)> https://www.youtube.com/watch?v=7Q8hAb230OE
[21:26:44]
<teslas_moustache (He/They)> This is a home setup. Kinda hard to actually call it "production" right now. I'll give it a shot just so I can move on from dicking around with BIOS bullshit.
[21:27:28]
<teslas_moustache (He/They)> > <@Alekswag:matrix.org> https://www.youtube.com/watch?v=7Q8hAb230OE
Heard
[21:37:44]
<Aleks (he/him/il/lui)> (ah also the user portal / sso was reworked and the paint is still fresh and pretty sure there are various bugs or unexpected behaviors here and there)
[23:50:36]
<miro5001> Bonne lecture https://sgbd.developpez.com/actu/355655/RIP-Redis-Comment-Garantia-Data-a-orchestre-ce-qui-pourrait-etre-considere-comme-le-plus-grand-vol-de-l-histoire-de-l-open-source-selon-Khawaja-Shams-et-Tony-Valderrama/