Posts made by robi

@humptydumpty said in VPS down, restart button won't work, host denies any downtime - Cloudron update related?:

Can I run a mail server on Cloudron's free tier?

yes

@johannesjom forks?

@johannesjom go for it.. it would be announced otherwise.

is your cloudron CLI up to date?

@girish okay, I thought this was selective for web based apps that it makes sense to limit in that way, not for the entire cloudron all at once.

For any affected mobile apps, one can use a pw enabled proxy-redirect to a secondary domain of the App.

@privsec

1. visit a friend/place who has a blazingly fast connection. (corps, ISPs, datacenters, universities, .FTTH, ..)
2. avoid uploading twice. (upload once, then xfer from there to the other)

There are many clients that can access S3, so you can try those to see if they work better or use a different configuration.

Avoid large transfers over SSH as it's inherently slow.

@girish why is a VPN setup a solution for basic http(s)/proxy auth?

That's like saying we're looking into SFTP access so we can run an SSH proxy to port forward HTTP(s) connections to the app gated by the SSH login.

@nebulon it can be gleaned from backup DB dumps

PSA:
Please note that AVX support is only a feature of the CPU generation, not any company or package bundle.

Virtualization platforms may choose to enable/disable CPU features, so there's that too.

For those that cannot afford it for whatever reason, consider offering services in return, say packaging or marketing or referrals for which the team can respond in kind.

What else does @staff appreciate?

@girish no there are other actual errors in there, nginx, tls, certs, and port. They're in bright red on my screen, the last few anyhow.

It may help to restart the app and see them fresh.

https://my.demo.cloudron.io/logs.html?appId=f7b6c07f-1af0-482b-bef0-98870fdbf15a

If you 'upload' your vids via streaming, there are services that connect to multiple destinations at once.

@zylstra from which app?

did you check the files there?

@martinkbs said in Best way to house/use content within Jellyfin?:

mounted with rsync

do you mean rclone?

@privsec you can play with rclone mounts for this, however you want to minimize the network distance to avoid hiccups.

UPDATE:

since the filename is so long.. I missed it as it gets hidden by the width of the window and doesn't wrap.. the file not found message is because the import UI attaches .tar.gz to any file name it doesn't recognize, even .tgz

Once the file was renamed or the extension removed, it worked.

@girish can you add that to the list of improvements in file handling during import?

@nebulon right, this also opens up mountable encrypted filesystems within a file.

traditionally that is managed with the quota system, or one can do so with mounts and partitions.

Just ran into this again.. and can't find a place to put it on the local filesystem for it to actually restore.

Even if all permissions are fixed, file is not found, heh.

Okay, it's a bit odd since that's the first time I've seen Surfer app crash due to this memory condition, presumably because the GC failed or was started too late with too little overhead available to complete.

Have tens of other Surfer apps that do not exhibit this behavior.

@girish It might be more efficient if you reproduce this one yourself in your demo/test cloudron with a quick install then having me post server non-specific details which you'll want to verify anyway.

check your DNS provider settings..

Let's just make sure where you're attempting to log into.. your VPS where you installed cloudron at your domain.com and manage apps, or your cloudron.io account to manage your subscription only?

As the app upgrades it seems to have many more features and subsystems that aren't being used.

The logs show a few errors on localhost as well as several warnings on unconfigured parts of the subsystems.

Anything we can improve?

The graphs seem to indicate something eating more resources over time.

It's not a busy site, and simply static pages.

@marcusquinn that's a different category, no? feedback vs forum.

Like comparing uservoice and nodebb.

https://standardnotes.com/blog/introducing-self-hosting-v2

@jdaviescoates with 20 upvotes and counting...

How about some more upvotes on this?

https://www.rukovoditel.net/testimonials.php

@scooke just search for "how to minio gateway mode" and check on sub/path support in their docs.

@jdaviescoates You could set up an S3 gateway in front that does support path style with Minio.

@LoudLemur How big is it?

couldn't one simply use any existing redis service from another container?

@marcusquinn what's the plugin for that?

The simpler option might be using a 'copy as markdowon' browser plugin which will grab those links as such and display them inline in the forum post.

@girish ideally an internal DNS mapping if dynamic, and an IP pattern to use if static, both of which should be viewable from the UI as needed.

https://www.seotoaster.com/

https://www.seotoaster.com/download-seotoaster-cms.html
https://www.seotoaster.com/shopping-cart/download-seotoaster-ecommerce-free-open-source-online-store-website-builder.html
https://www.seotoaster.com/crm/download.html
https://www.seotoaster.com/website-plugins-marketplace.html

Even has SeoSamba marketing platform and mobile apps for management.

The orchestration of this depends on if you have the users have access to Cloudron UI and LDAP or not. From there you can separate things in the UI or have start-pages for each org so they can jump to their apps easily.

@marcusquinn their support should be able to tell you knowing how the back end works. Some systems are highly optimized for injest, but less for garbage collection, which can be deferred since it causes injest disruption. If they do this at 90day intervals, that's a pretty poor architectural set of choices.

If it's not usable, it should not be charged or passed on to the customer. They'd be smarter to hide it with larger customers where it matters less than smaller ones where it's very obvious.

@LoudLemur Just add all domains and install apps on each. Easy.

@privsec that depends on the settings in handbrake, look into it.

@marcusquinn and they charge for it?

@privsec MKV is a container format that wraps a bunch of useful stuff in one file.

If something doesn't play, it'd be nice to get the details of that file and the inner structure so you get more info on what's not working.

So if one mkv has an mp4 inside and plays fine, yet another has an avi and doesn't, that's a strong clue.

@nebulon A setting to start apps before a backup, then have them update and shutdown after might be a workflow that applies here.

@girish is this documented? app to app mounts?

@nichu42 right, here are the options:
https://boto3.amazonaws.com/v1/documentation/api/latest/guide/credentials.html

@privsec either use automated DNS setup and manage the exceptions or use a manual setup and manage DNS elsewhere.

best to front it somehow so it's not direct access. API or other DB multiplexing sw.

@Mastadamus why not do it behind?

@privsec yes

@jdaviescoates if you get a public link, does it play in the browser?

compare that to the link the app tries to use?

@nichu42 hmm, ok.

what's the error if you rename the .yaml file? (just to make sure that's where it's looking for the info..)

@nichu42 yeah those are s3 credentials for the bucket that need to be provided..

see if you can find example commands for the upload step.

For example, In App settings, the left pane menu items would have an underlined character that would allow for keyboard navigation.

Subsequently all buttons should be keyboard activation enabled.

This would significantly speed up administration of repetitive tasks and improve UX.

https://www.reddit.com/r/Mastodon/comments/yzp1qz/mastodon_server_hardware_requirements/

This is a discussion on Reddit about what it takes to host a Mastodon instance (Mastodon uses the ActivityPub protocol). The requirements are quite high. One large instance with 29,000 users is hosted on this hardware:

2 AMD 5950 16c/32thread 128GB ram with nvme ssds running sidekiq jobs
1 AMD 5950 16c/32thread 128GB ram with nvme ssds running elasticsearch and redis
1 AMD 5950 16c/32thread 128GB ram with nvme ssds running Postgres DB
2 AMD 5950 16c/32thread 128GB ram with nvme ssds running PUMA (mastodon web app) and nginx (only one is currently powered on - other is being reworked
1 AMD Epyc 32c/64th 400GB ram with 10TB of nvme ssd acting as a s3 storage provider and another nginx/PUMA web front end

This is a total of 96 CPU cores! Another person with 500 users says,

"For those interested in a data point, I am running an instance with nearly 500 users. Currently running on an AWS Lightsail instance with 4GB Ram, 2 vCPU, for $20/month, and an S3 instance for media storage (1TB - nearly full) for $3/month. I am constantly monitoring resources - as the system is running well but nearly topped out.

I will likely be moving to 8GB RAM, 4 vCPU instance, and 2TB S3 in the coming weeks."

These requirements are quite high. I also recently learned that bandwidth is a continuous 1-2 Gbits/s. No wonder Mastodon is having trouble keeping up with the influx of Twitter users!

@jdaviescoates playback may require the bucket to be https accessible, which is a setting.

@jdaviescoates that's a good pivoting data point @jdaviescoates , mine is a wildcard setup too. Likely something specific to that branch of code..

@marcusquinn this would not make it so occasional, and provide for other automations to make it more frequent.

@marcusquinn I think we need an n8n workflow to take an app from the appstore and tweet its link with a webhook/rss feed for us and our n8n installs pick it up and RT it!

@girish done https://forum.cloudron.io/topic/8487/enhance-app-proxy-app-with-path-and-port-support