yunohost-apps / calckey_ynh

Calcket package for YunoHost

Calckey for YunoHost

Read this README in other languages.

This package allows you to install Calckey quickly and simply on a YunoHost server.
If you don't have YunoHost, please consult the guide to learn how to install it.

Overview

A greatly enhanced fork of Misskey with better UI/UX, security, features, and more! https://i.calckey.cloud/

Calckey is based off of Misskey, a powerful microblogging server on ActivityPub with features such as emoji reactions, a customizable web UI, rich chatting, and much more!
Calckey adds many quality of life changes and bug fixes for users and instance admins alike.

Shipped version: 13.1.4.1~ynh1

Demo: https://i.calckey.cloud/

Screenshots

🔴 Antifeatures

• Upstream not maintained: Calckey has been replaced by Firefish, install that new app instead. A migration procedure is being developed for existing instances.

Documentation and resources

• Official app website: https://i.calckey.cloud/
• Upstream app code repository: https://codeberg.org/calckey/calckey
• YunoHost Store: https://apps.yunohost.org/app/calckey
• Report a bug: https://github.com/YunoHost-Apps/calckey_ynh/issues

Developer info

Please send your pull request to the testing branch.

To try the testing branch, please proceed like that:

sudo yunohost app install https://github.com/YunoHost-Apps/calckey_ynh/tree/testing --debug
or
sudo yunohost app upgrade calckey -u https://github.com/YunoHost-Apps/calckey_ynh/tree/testing --debug

More info regarding app packaging: https://yunohost.org/packaging_apps

calckey_ynh's Issues

Instructions on how to migrate from Misskey YunoHost to Calckey YunoHost

Currently we have instructions on how to migrate standard deployments of Misskey to Calckey here: https://codeberg.org/calckey/calckey/src/branch/develop/docs/migrate.md

However, someone recently asked how to do that for YunoHost, and sadly I don't have many pointers to give. If @oufmilo or someone else could elaborate on that process I'd be happy to add it to the official docs.

EACESS issue when restoring Calckey

I'm trying to restore a Calckey backup (from v13.x).
It fails with this error:

DEBUG sudo -u calckey env PATH=/opt/node_n/n/versions/node/19/bin:/opt/node_n/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/root/.cargo/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/
sbin:/bin pnpm install                                                                                                                                                                                                      
DEBUG Internal Error: EACCES: permission denied, mkdir '/var/www/calckey/.cache/node/corepack'                                                                                                                       
DEBUG Error: EACCES: permission denied, mkdir '/var/www/calckey/.cache/node/corepack'

I tried to remove the .cache folder from the backup archive, but no luck.
I guess it's because that pnpm install command (https://github.com/YunoHost-Apps/calckey_ynh/blob/master/scripts/restore#L44) try to create the folder ?

ping @ThatOneCalculator, in case you have some clue 😅

Not in the YunoHost app catalog

Hello!

I'm the lead developer of Calckey, and I'm very happy to see that a Yunohost package has been made for Calckey. I don't use Yunohost personally, but a lot of people have been asking for it :)

I tried going to the link in the readme (https://install-app.yunohost.org/?app=calckey), but it says that "This application is not in YunoHost's app catalog." When can users expect to see this in the catalog?

Error restoring from backup when files have been added to /var/www/calckey/packages/backend/assets/

Describe the bug

• restoring Calckey backup on freshly setup Yunohost server fails when backup has files in /var/www/calckey/packages/backend/assets/

Context

• OVH baremetal dedicated; 32GiB ram, 1TiB ssd
• Debian 11
• YunoHost version: 11.1.18
• I have full root access to the server
• nothing has been manually tweaked
• attempting to restore Calckey Version 13.1.4.1~ynh1

Steps to reproduce

• upload $backup.tar via SCP
• in webadmin, navigate to Backup > Local archives /home/yunohost.backup > 10 days ago $backup (3 GB) /home/yunohost.backup/archives/$backup.tar
• deselect all but 'Calckey $Name calckey (3 GB) Version 13.1.4.1~ynh1'
• click 'Restore'
• installation of app/dependencies et al completes successfully, restoration of first encountered file in /var/www/calckey/packages/backend/assets/ fails with permissions issue, restore is rolled back
• {same happens if attempted again after reboot}

Expected behavior

• app to be restored to working state

Logs

https://paste.yunohost.org/raw/udepibivab

2023-05-07 16:47:51,803: DEBUG - [16:47:51] 'copy:backend:custom' errored after 324 ms
2023-05-07 16:47:51,804: DEBUG - [16:47:51] Error: EACCES: permission denied, open '/var/www/calckey/packages/backend/assets/clouds-high.jpg'
2023-05-07 16:47:51,804: DEBUG - [16:47:51] 'build' errored after 327 ms
2023-05-07 16:47:51,812: DEBUG -  ELIFECYCLE  Command failed with exit code 1.
2023-05-07 16:47:51,825: DEBUG -  ELIFECYCLE  Command failed with exit code 1.
2023-05-07 16:47:51,838: DEBUG - + ynh_exit_properly
2023-05-07 16:47:51,839: DEBUG - + [[ restore =~ ^install$|^upgrade$|^restore$ ]]
2023-05-07 16:47:53,344: ERROR - Could not restore calckey: An error occured inside the app restore script

Why is it marked as "low quality"?

Calckey:

Misskey:

If the Calckey package is based off the Misskey package, why does the Calckey package have this label but the Misskey one doesn't?

Can't restore from backup after server change

Describe the bug

The restoration of my Calckey backup failed.

Context

• Hardware: VPS on Debian 11
• YunoHost version: 11.2.3 (stable)
• I have access to my server: Webadmin and SSH
• Are you in a special context or did you perform some particular tweaking on your YunoHost instance?: Yes, fresh Yunohost installation and restored the backup I made of the Calckey install
• Using, or trying to install package version/branch: See above
• If upgrading, current package version: n/a

Steps to reproduce

• Backup existing Calckey install
• Install Yunohost on a new server
• Add the Calckey domain name and run LetsEncrypt
• Restore the backed up install

Expected behavior

It should've restored

Logs

https://paste.yunohost.org/raw/ikosupaqev

Workers die immediately after spawning with "trap invalid opcode" in dmesg

Describe the bug

Workers die immediately after spawning with the following error in dmesg:

Context

• Hardware: VPS
• Yuno versions:

yunohost:
repo: stable
version: 11.1.21.4
yunohost-admin:
repo: stable
version: 11.1.10
moulinette:
repo: stable
version: 11.1.4
ssowat:
repo: stable
version: 11.1.4

• Server access: SSH
• Special context: I don't think so
• Using: 13.1.4.1~ynh1

Steps to reproduce

Install Calckey to secondary domain

Expected behavior

Workers should not die upon spawning.

Logs

Install: https://paste.yunohost.org/raw/ehuvekeseg

nginx domain error:

2023/06/22 02:27:46 [error] 9469#9469: *13859513 connect() failed (111: Connection refused) while connecting to upstream, client: xx.xx.xx.xx, server: myhost.org, request: "GET / HTTP/2.0", upstream: "http://127.0.0.1:3020/", host: "myhost.org"

Journal:

Jun 22 02:19:41 myhost.town npm[31938]: (node:31938) ExperimentalWarning: Import assertions are not a stable feature of the JavaScript language. Avoid relying on their current behavior and syn>
Jun 22 02:19:41 myhost.town npm[31938]: (Use `node --trace-warnings ...` to show where the warning was created)                                                                                 
Jun 22 02:19:43 myhost.town npm[31938]: (node:31938) ExperimentalWarning: Importing JSON modules is an experimental feature and might change at any time                                        
Jun 22 02:19:50 myhost.town npm[31454]: ERR  *        [core cluster]        [11] died :(

(Repeat for each spawned worker)

/var/log/messages:

Jun 23 20:04:59 myhost kernel: [81392.870690] traps: Calckey (worker[225232] trap invalid opcode ip:7f26d2cc68d3 sp:7ffdfe003ed0 error:0 in libvips-cpp.so.42[7f26d2ab3000+a42000]

(Repeat for each spawned worker with slightly different addresses)

This is almost certainly upstream but I'm posting here per forum suggestions.

Warning for next release

We're switching from yarn to pnpm.

The process for installing packages is now:

corepack enable
corepack prepare pnpm@latest --activate
pnpm i

Futhermore, all yarn run should be replaced with pnpm run and all yarn workspace with pnpm --filter

Testing is not the main branch + is out of sync

You should make testing the default branch to align with common Yunohost guidelines :)

14.0 RC3 - Still not able to import following.csv

I was able to upgrade my setup to the latest RC, but still the same problem as before:

Uploading a following.csv works, the server tells me that it's starting to process the import, nothing happens.

I experienced the same on moppels.bar, which AFAIK is also hosted on Yunohost.

On calckey.world, the very same .csv worked just fine.

Where should I look (logfiles) for an explanation?

Thanks in advance.

502 Bad Gateway

Describe the bug

I restarted the Calckey service because I was having issues with federation and got the errors below in the log. I have no idea how to resolve this and my instance is completely inaccessible.

Context

• VPS running Debian 11
• YunoHost 11.1.15 (stable)
• Calckey 13.1.2~ynh1

Steps to reproduce

Log into Yunohost admin panel
Click on tools > services
Click on calckey
Click restart button

Expected behavior

The service would restart and reload the UI

Logs

https://paste.yunohost.org/tesahetaya

Edit: there are more logs that occurred after that one: https://paste.yunohost.org/unuvaxabem

v13.1.3

https://calckey.social/notes/9ccqvryj3t

Unable to restore from backup or get admin for a fresh install.

I just moved my Yunohost setup to a new machine. Everything worked quite nicely, except the restore of the Calckey backup.

There's been a similar report a few weeks ago:

#40 (comment)

I get the very same error message when trying to restore the backup and I also cannot make the first user on a fresh install admin.

Testing branch update.

Hey, can we get release candidate 2 on testing branch?

Warning for the next upcoming release (v13.2.0)

Word of warning for the next update: Rust will be needed to build Calckey.

Not in Catalog

Exclude cache on backup

Calckey writes a lot of cache files for yarn pnpm in the future ? #10 and node_modules, up to 2GB after a fresh install.

I suppose this could not be backed up, as it would rebuilt it anyway ?

Then we should exclude it from the backup.

I'd be happy to test this is someone does it :)

Calckey has been rebranded into Firefish

cf https://calckey.social/notes/9fatefrf0h

We should be doing an RC2 for v14 soon, and if everything is fine, we'll release v14 a few days later. However, we're in the process of rebranding, so this might affect new releases a little.

Restoring instance from backup.

I'm terribly sorry for this whole mess I caused, it was DNS...

Of course, I panicked and deleted my instance (but was sane enough that I did a backup before), and now I have a new problem...

When I try to restore from backup (I tried the 14rc backup and the backup it did before updating to 14) I get an error...

2023-06-16 02:57:15,047: DEBUG - error: could not create bin directory: '/var/www/calckey/.cargo/bin': Permission denied (os error 13)

Here are the logs: https://paste.yunohost.org/raw/amesudozig

I'm terribly sorry for the whole trouble. I hope you can forgive me and help me out last time.

BTW, installing calckey from scratch worked, but I would like to not lose my posts.

Originally posted by @Ciechom in #38 (comment)

In the previous issue, I included more logs.

Edit: After failed restore from backup, when I try to deploy a fresh calckey instance when I register as a new user, the admin panel is unavailable.

Logs filling with "bad system call"

Immediately after starting Calckey from the installation the service logs are filled with

Jun 19 13:41:00 vps115393 npm[52192]: Bad system call
Jun 19 13:41:02 vps115393 npm[52192]: Bad system call

This line recurs approximately every 2 seconds, blowing away any and all other diagnostic information.

By monitoring /var/log/syslog I see that these lines from the audit log may be related. they also reoccur every 2 seconds:

Jun 19 13:41:02 vps115393 kernel: [74017.068000] audit: type=1326 audit(1687182062.366:36192): auid=4294967295 uid=996 gid=996 ses=4294967295 subj=unconfined pid=1035102 comm="ip" exe="/usr/bin/ip" sig=31 arch=c000003e syscall=126 compat=0 ip=0x7f46195c9387 code=0x80000000
Jun 19 13:41:04 vps115393 kernel: [74019.069795] audit: type=1326 audit(1687182064.370:36193): auid=4294967295 uid=996 gid=996 ses=4294967295 subj=unconfined pid=1035124 comm="ip" exe="/usr/bin/ip" sig=31 arch=c000003e syscall=126 compat=0 ip=0x7f82a57ab387 code=0x80000000

Server is running a fresh out of the box installation of yunohost on a Debian 11 VPS.

So far I haven't been able to see what this is impacting. The Calckey instance does seem to be running correctly, I'm able to post and view posts etc, but would like to get the log spam to calm down if possible.