[09:56:39]
<yvanq> Hello, I try to restore my last app on new server but I have a problem and it refuse
`WARNING - ERROR 1045 (28000): Access denied for user 'hubzilla'@'localhost' (using password: YES)`
`ERROR - Impossible de restaurer hubzilla : Une erreur s'est produite dans le script de restauration de l'application`
Here is the log (as yunopaste don't work on my unit)
https://elemac.fr/nextcloud/s/5DmaT4HawAoQ7gy
[10:01:38]
<orhtej2> sooooo it created postgresql db then tries to restore mysql as this is what the backup contains, not sure how to migrate (I would perform the surgery on backup archive so that mysql-export is compatible with postgresql, but that's pain)
[10:15:28]
<orhtej2> alternatively you can edit the manifest inside the backup to say dbtype is mysql
[10:17:33]
<yvanq> > sooooo it created postgresql db then tries to restore mysql as this is what the backup contains, not sure how to migrate (I would perform the surgery on backup archive so that mysql-export is compatible with postgresql, but that's pain)
It's strange, I made the backup on the same system yunohost 11.2.30.2 (and almost same computers)
[10:19:05]
<orhtej2> > <@yvanq:matrix.org> It's strange, I made the backup on the same system yunohost 11.2.30.2 (and almost same computers)
all I'm saying is the switch from mysql to postgresql was not handled properly in hubzilla_ynh, it's not your fault
[10:33:21]
<yvanq> And the archive is just so huge ! 33gb ! (with php.log deleted), can I remove the huge store/store.... folder witch is so fat ?
[10:34:19]
<orhtej2> > <@yvanq:matrix.org> And the archive is just so huge ! 33gb ! (with php.log deleted), can I remove the huge store/store.... folder witch is so fat ?
doubt it, what's so big inside?
[10:56:20]
<yvanq> https://aria.im/_matrix/media/v1/download/matrix.org/gyoZWjrxsnkIcFppzstsEKQY
[10:57:02]
<Juan Carlos> somebody else with issues while trying to download resources from the yunohost repo?
[10:57:02]
<Juan Carlos> Hi
[10:59:31]
<orhtej2> > <@outofservice:matrix.org> somebody else with issues while trying to download resources from the yunohost repo?
#define 'issues while trying etc etc'
[11:07:03]
<Juan Carlos> https://aria.im/_matrix/media/v1/download/matrix.org/CLDknFhwzrUerzsJzvmZwpqT
[11:07:03]
<orhtej2> > <@yvanq:matrix.org> sent an image.
sounds like actual app data oO
[11:07:03]
<Juan Carlos> In the irc room, it seems everybody is sleeping
[11:07:03]
<orhtej2> > <@outofservice:matrix.org> In the irc room, it seems everybody is sleeping
isn't it bridged to matrix?
[11:07:04]
<Juan Carlos> I mean, in the server, not in my computer...
[11:07:04]
<orhtej2> forge is having some intermittent problem I guess, try again later?
[11:07:04]
<Juan Carlos> some hours ago I faced an isusue in the main web page, talking about no enough room in /var/....
[11:07:04]
<Juan Carlos> https://aria.im/_matrix/media/v1/download/matrix.org/DiyrmOsUWzpGBWvxnKdriqAO
[11:07:05]
<orhtej2> > I can't wrap my head around it but backing up Nextcloud with borg [fails](https://paste.yunohost.org/raw/fapixidequ) and it has something to do with backing up of `nginx/conf.d/domain.tld/nextcloud.conf` and `nginx/conf.d/domain.tld/nextcloud.d/`. There is already mount for`nginx/conf.d/domain.tld/`, hence linking subentries for it fail. Any ideas on how to resolve? Perhaps this should be a plain directory with symlinks, not mount?
holy cow nextclout nginx conf is declared twice x_x
[11:07:05]
<orhtej2> I understand ;) I think this will get resolved eventually
[11:07:05]
<Juan Carlos> ok. Just wanna you to know
[11:10:00]
<Salamandar> might be why the CI has issues too
[11:10:01]
<orhtej2> > <@rodinux:matrix.org> I have something also with `yunohost tools update` `Error: Unable to download the default app catalog: Corrupted JSON read from https://app.yunohost.org/default/v3/apps.json (reason: Unterminated string starting at: line 1 column 65137 (char 65136))`
infra is having some downtime 🤷
[11:10:02]
<orhtej2> `curl: (28) Failed to connect to app.yunohost.org port 443 after 21143 ms: Couldn't connect to server`
[11:10:02]
<rodinux> I have something also with `yunohost tools update` `Error: Unable to download the default app catalog: Corrupted JSON read from https://app.yunohost.org/default/v3/apps.json (reason: Unterminated string starting at: line 1 column 65137 (char 65136))`
[11:10:02]
<Salamandar> oh !
[11:27:33]
<Salamandar> :D
[11:28:14]
<rodinux> I confirm
[11:29:17]
<Salamandar> yes, i wiped out a bit of log files
[11:29:39]
<Salamandar> in a few days we'll stop having such issues, we are in the rpocess of migrating the infra
[11:29:53]
<Salamandar> @orhtej2:circledsquareroot.ovh that's most certainly why the readmy generator was out too…
[11:30:06]
<rodinux> 🤞
[11:30:35]
<yvanq> I try to restore with "mysql" mention at the end of manifext.toml
\[resources.database\]
type = "mysql"
If it fails, is there a way to convert db mysql to progresql on the old pc then make again a backup to restore it on the new one ?
[11:31:08]
<orhtej2> > <@yvanq:matrix.org> I try to restore with "mysql" mention at the end of manifext.toml
> \[resources.database\]
> type = "mysql"
> If it fails, is there a way to convert db mysql to progresql on the old pc then make again a backup to restore it on the new one ?
I mean SQL is **almost** standard, no clue how much tinkering it requires in this specific case 🤷
[11:43:38]
<OutOfService> Is there any working subproyect to deploy yunohost via ansible?
[11:44:13]
<tituspijean> Hi, there's an unofficial work on it: https://forum.yunohost.org/t/ansible-collection-ansible-yunohost-v1-1-4-is-out-lets-contribute/17322/21
[12:14:26]
<OutOfService> great. Thanks a lot!!
[13:09:34]
<yvanq> > I mean SQL is **almost** standard, no clue how much tinkering it requires in this specific case 🤷
tryed to restore with mysql option, error...😪 even with mysql, php's services restart
Hubzilla page says :
`Apologies but this site is unavailable at the moment. Please try again later.`
[13:30:37]
<orhtej2> > <@yvanq:matrix.org> tryed to restore with mysql option, error...😪 even with mysql, php's services restart
> Hubzilla page says :
> `Apologies but this site is unavailable at the moment. Please try again later.`
Restore log?
[13:31:04]
<yvanq> where ?
[13:31:04]
<orhtej2> Its available in webadmin tools logs
[13:31:14]
<yvanq> i have an error with yunopaste also🙄, I paste somwhere else
`Something wrong happened while trying to paste data on paste.yunohost.org : 502,`
[13:31:14]
<orhtej2> > <@yvanq:matrix.org> https://elemac.fr/nextcloud/s/8a33cFebrRdHSYF
it's the same error as it provisioned postgresql database again
[13:31:14]
<yvanq> > Its available in webadmin tools logs
https://elemac.fr/nextcloud/s/8a33cFebrRdHSYF
[13:33:02]
<yvanq> changed this mention at the end of manifext.toml and a php hubzilla appears in phpmyadmin
```
[resources.database]
type = "progresql"
```
to
```
[resources.database]
type = "mysql"
```
[13:33:02]
<orhtej2> did you edit manifest.toml inside the backup archive to read 'mysql' as database type?
[13:33:02]
<yvanq> yes, and it's a night mare to work on a 32gb
[13:33:04]
<orhtej2> it clearly provisioned a PSQL DB, then failed to restore MySQL backup 🤷
[13:33:04]
<yvanq> it seems to be somewhere else than in the manifext.toml this migration operation
[13:33:04]
<orhtej2> so IDK, you shared the wrong log or something?
[13:33:45]
<yvanq> > <@Salamandar:matrix.org> paste.yunohost is back up
cool !
https://paste.yunohost.org/raw/yojicekehu
[13:33:46]
<orhtej2> > <@yvanq:matrix.org> cool !
> https://paste.yunohost.org/raw/yojicekehu
ok, this one actually restored the mysql db correctly
[13:33:46]
<Salamandar> paste.yunohost is back up
[14:36:19]
<yvanq> but Hubzilla page says :
`Apologies but this site is unavailable at the moment. Please try again later.`
I change also this line in manifext.toml and give a try
```
[resources.apt]
packages = "postgresql, php8.2-curl, php8.2-gd, php8.2-pgsql, php8.2-mbstring, php8.2-xml, php8.2-zip, php8.2-cli, php8.2-imagick, php8.2-gmp, php8.2-intl"
```
to
```
[resources.apt]
packages = "mysql, php8.2-curl, php8.2-gd, php8.2-pgsql, php8.2-mbstring, php8.2-xml, php8.2-zip, php8.2-cli, php8.2-imagick, php8.2-gmp, php8.2-intl"
```
[14:36:35]
<yvanq> Doesn't work restore failure...
[14:37:52]
<yvanq> https://paste.yunohost.org/raw/unefoqudav
[14:38:51]
<Salamandar> 2024-09-25 15:53:30,696: WARNING - E: Package 'mysql' has no installation candidate
[14:39:31]
<Salamandar> humm
[14:40:08]
<Salamandar> should be mariadb-server
[14:41:06]
<Salamandar> iirc
[15:41:44]
<orhtej2> > <@yvanq:matrix.org> but Hubzilla page says :
> `Apologies but this site is unavailable at the moment. Please try again later.`
> I change also this line in manifext.toml and give a try
>
> ```
> [resources.apt]
> packages = "postgresql, php8.2-curl, php8.2-gd, php8.2-pgsql, php8.2-mbstring, php8.2-xml, php8.2-zip, php8.2-cli, php8.2-imagick, php8.2-gmp, php8.2-intl"
> ```
>
> to
>
> ```
> [resources.apt]
> packages = "mysql, php8.2-curl, php8.2-gd, php8.2-pgsql, php8.2-mbstring, php8.2-xml, php8.2-zip, php8.2-cli, php8.2-imagick, php8.2-gmp, php8.2-intl"
> ```
I think you need to adjust the config to include the newly generated db credentials
[18:32:09]
<Igor Leturia> Hi! I've installed Yunohost in a new VPS for the first time. It is a Debian 12 and Yunohost 12. I know it is a beta, but still I understood it worked well for the most basic things, and I am having a lot of serious problems in just the first steps...
I only did the initial postinstall (main domain and first user), and this are the problems I have:
- I cannot SSH any more, not as root like before and not as the new first user either. It says permission denied. And I know the user and password are correct, they are the same ones I can enter in Yunohost admin and they work.
- In Yunohost admin, I cannot see the domains, the web page is blank.
Any ideas?
[18:45:58]
<Aleks (he/him/il/lui)> can you double check your first user is in the admin group (it should be but who knows) in the webadmin > Users > Groups and permissions(?)
[18:47:07]
<Aleks (he/him/il/lui)> the "domain page is blank" issue is a known bug ... the fix exist but was not release ... but let me try creating a release because that's clearly a major issue right now
[18:48:42]
<Igor Leturia> Yes, the user is in the admin group.
[18:49:30]
<Igor Leturia> > <@Alekswag:matrix.org> the "domain page is blank" issue is a known bug ... the fix exist but was not release ... but let me try creating a release because that's clearly a major issue right now
However, can I install the new release without SSH access?
[18:49:59]
<Aleks (he/him/il/lui)> yes you can upgrade yunohost through the "system upgrade" thing in the home of the webadmin
[18:51:43]
<Aleks (he/him/il/lui)> i'm trying to think about hypothesis regarding the ssh access ... could be that you chose to *not* let yunohost take over the ssh conf during the initial install, but if that was the case you should still have access using root
[18:53:16]
<Igor Leturia> > <@Alekswag:matrix.org> i'm trying to think about hypothesis regarding the ssh access ... could be that you chose to *not* let yunohost take over the ssh conf during the initial install, but if that was the case you should still have access using root
I am sure I told it to change SSH conf, but if I did not, it would let me SSH as root, as you say...
[18:53:52]
<Aleks (he/him/il/lui)> could it be that the first user id was already an existing unix user on the system beforehand ? (but yunohost should have complain about it if that was the case)
[18:54:22]
<Igor Leturia> Might it be because the change of the domain name to the new server's IP is not propagated yet, and thus the domain did not get a LEt's Encrypt certificate, and that is somehow affecting everything?
[18:55:15]
<Igor Leturia> > <@Alekswag:matrix.org> could it be that the first user id was already an existing unix user on the system beforehand ? (but yunohost should have complain about it if that was the case)
No, there was only the root user.
[18:55:19]
<Aleks (he/him/il/lui)> uuuuh i doubt that SSL certificate have anything to do with ssh ... but yeah maybe the DNS propagation could be related idk ? Could it be that "ping <domain>" pings some old IP ?
[18:55:24]
<Aleks (he/him/il/lui)> what if you try to ssh using the explicit global IP ?
[19:03:29]
<Igor Leturia> > <@Alekswag:matrix.org> what if you try to ssh using the explicit global IP ?
I am doing it that way... I meant if the certificate thing could be related to other problem of not showing the domains.
[19:06:13]
<Igor Leturia> If I ping the domain from my machine it calls the old IP, but if I use a web service it seems to be using the new one. However, maybe it was not yet propagated when I added the domain, but now I cannot manage the domains because the page is blnak.
[19:07:16]
<Aleks (he/him/il/lui)> yes, the fix for the blank domain page is on the way (i'm working on the release as we speak), i'm more concerned with the ssh issue
[20:04:25]
<Igor Leturia> Hi! The login problem might not be related to Yunohost, I have ordered a fresh reinstall of the OS to start the whole process again, and it is impossible for me to login anyway.
[20:05:42]
<Igor Leturia> So I will first have to solve that with the hosting providers, then I will install Yunohost 12 again and I will come back here if the domains screen blank problem persists. Or maybe I should wait for the release?
[20:06:52]
<Aleks (he/him/il/lui)> the domain blank screen thing should be fixed with the new version i just released moments ago
[20:15:25]
<Igor Leturia> Wow, thanks! How do I install that release?
[20:15:45]
<Igor Leturia> I installed the previous one in a fresh Debian 12 with the command "curl https://install.yunohost.org/bookworm | bash -s -- -d testing". Will that now install the new release?
[20:17:05]
<Aleks (he/him/il/lui)> yes, but you can also just upgrade an existing with yunohost in the webadmin > System upgrade
[20:17:32]
<Aleks (he/him/il/lui)> (or `yunohost tools update system` and `yunohost tools upgrade system`)
[20:25:44]
<Igor Leturia> Ok, thanks. I'll keep you posted when I regain access to the server...
[20:47:04]
<yvanq> Hubzilla restore work in progress, I finally made a backup from old server with https://myhubzilla-server/uexport and decide to make a fresh install (9.2.1~ynh1), my new problem, impossible to log in through yunohost ldap...
[21:02:05]
<orhtej2> > <@yvanq:matrix.org> Hubzilla restore work in progress, I finally made a backup from old server with https://myhubzilla-server/uexport and decide to make a fresh install (9.2.1~ynh1), my new problem, impossible to log in through yunohost ldap...
the login is user@domain.tld, not the user alone
[21:10:50]
<yvanq> Don't work.
[21:17:36]
<orhtej2> > <@yvanq:matrix.org> Don't work.
hm indeed
[21:20:16]
<yvanq> does anybody use new install hubzilla or streams on yunohost 11.2.30.2 ? since it creates progresql db ? I will open issues on theses app
[21:48:13]
<am> > does anybody use new install hubzilla or streams on yunohost 11.2.30.2 ? since it creates progresql db ? I will open issues on theses app
As long as I remember, I never had the possibility to use SSO with Hubzilla on YunoHost.
You need to sign-up as a new user 😉
As first user, you should get the admin rights on your Hubzilla instance.
[21:54:12]
<am> yvanq: