What do you all think about having a regular weekly event to package an important App as a group in a live session, maybe even live streamed, where anyone on the call can also contribute their small part, so one person doesn't have to slog through it all by themselves?

Others can watch and learn and ask questions & chit chat.

Any volunteers?

Rob here (robi is my Croatian nickname and often computer username) for those in other cultures, short for Robert.

As a Strategic Advisor and (Business & Innovation) Catalyst, I'm mostly a full stack janitor

As a leader in several startups, I reserve a keen eye for emerging technologies (hi Cloudron and it's Apps); I have been a speaker, mentor, advisor for Google, NASA, IBM, and many more, finding safe, creative solutions all the way from the mundane to the complex with equal ease, which is one thing that keeps my passion ignited.

Remember WinAmp? Yeah, I was there during the mp3 revolution, in the credits until the recent reshuffle.

I've invented a bunch of stuff, including the first PCIe flash storage device in small form factor (SFF), now you'll recognize it as the M.2 PCIe SSD device for which there is now a slot on every motherboard being made. (didn't make a dime.)

Made core discoveries (read blunders) in physics (magnetism, light, water), mathematics, statistics, AI, ML, medicine, law, etc, which you can ask me about outside this thread.

3 decades ago I got bitten by the sysadmin bug and never left, despite all the C level gigs.

After 15 years, I am still on the board of directors for https://BayLISA.org which is one of the oldest meetups in Silicon Valley. We've had the Cloudron team speak at one of our meetups which should be online on YT.

I play a long running pickup soccer game every Saturday (same folks come out for 8+ years) and tennis 1-2 times a week if possible. Love ping pong and ultimate frisbee too.

If you want to see a small part of my other crazy projects, see https://venture.toldyouso.com

Otherwise a few here have found me on Twitter.

There's more, but that's plenty for now

Porter is a Kubernetes powered PaaS that runs in your own cloud provider.

Repository: https://github.com/porter-dev/porter

Porter is an alternative to Heroku that spins up the underlying infrastructure in your own cloud provider. The complexity of the underlying infrastructure is abstracted away from you just like Heroku, but you can fully control the underlying infra when you need to (as your applications scale and "grow out" of the platform).

I was always hosting my own things since the first linux install back in the early 90s. When I came across Sandstorm I thought it was the bees knees, but it was far from pleasant to use with all the rough security and then the project shutdown! (for a while).

I actually called up the founder since I knew of him, and had a chat about it.. he was rearranging his life as he went to work for Cloudflare.

Shortly after I found Cloudron and haven't had to look back at anything else.

I even had @girish give a presentation on the mighty Cloudron at our meetup BayLISA. It was very impressive.

SO in some small way I am responsible for infecting a bunch of BayLISA members with the Cloudron bug and run 3 different Cloudrons myself spanning hundreds of domains now. Go Cloudron.

As I run 50+ apps in my Cloudron, I often have to check on services during troubleshooting, and when loading the System Info (/#/system) page it displays the Services, BUT as soon as I want to click on one, the Disk Usage loads and pushes all the Services down a page. (Bait and scroll?! )

Always makes me wonder why Services doesn't have it's own tab to avoid this problem.

Please fix.

As a very happy birthday present, this came about synergistically, and now we have the worlds first Cloudron App running in a Sysbox container runtime from Nestybox.com

Gratitude to this community and the Nestybox team.
This couldn't happen w/o @marcusquinn & @Rodny-Molina

Marcus provided the Cloudron based development environment which made this super easy and convenient, and Rodny provided the integration muscle and Sysbox expertise.

Here is the first issue filed related to our work:
https://github.com/nestybox/sysbox/issues/151

Feel free to follow along or jump in.

That's all for now

Open Source Social Network (OSSN)
https://www.opensource-socialnetwork.org/

Open Source Social Network (OSSN) is a social networking software written in PHP. It allows you to make a social networking website and helps your members build social relationships, with people who share similar professional or personal interests. It is available in 16 international languages.

https://github.com/opensource-socialnetwork/opensource-socialnetwork/

It's very easy to click on the tile opening the app when the intent was to open the configuration page.

Please increase the config clickable area as there is plenty to click the app. If that means making the icons larger, so be it, but their background area should be much larger.

I dropped a line on their Github issue to see if anyone is interested.

Currently the way to roughly clean up some backups we don't want taking up space is to change the retention policy for a while, then come back later and change it back.

It would be nice to have a delete function for individual backups w/o having to shorten the policy.

perhaps @staff can include these changes by default in the next package update.

@onlybro try it

@hardwarehulp how and where did you install it?

@timconsidine ah, let's see what @marcusquinn thinks about that, as he's been using EspoCRM for similar things.

@timconsidine can you describe what you're trying to do, in terms of the app?

I decided to reboot the box for security upgrades (from notifications) and it came up without errors this time.

After a server restart, graphite won't start. Reconfig doesn't help.

@girish said in Cloudron+ZFS?:

I think this might also help us in setting per-app data directory size limits.

Yes

There was another long there elsewhere about how running out of disk space should not bring down the server. I think with having appdata in a volume with specific size (assuming zfs allows us to resize volumes easily etc), then that's a big win!

Yes it does! From 10 years ago I recall it's only resizing up. Downsize was a lateral move across volumes.

Any idea on how these things perform? As long as it's reasonable, that should be ok. It's not like most apps are very disk i/o heavy.

Yes, it runs better than most any other filesystem because of the efficiency and caching that ZFS does. It can also be further enhanced with their ZIL (intent log) that's kept in RAM. For busy systems that is a godsend.

I was one of the key members of the team actually created the first ZFS system on the planet, stretched across two nodes.

https://www.researchgate.net/publication/333250276_OCP_Hackathon_2015_N-Way_OpenZFS_Redundancy_and_Horizontal_Scaling_on_Flash_Memory_Switched_Fabrics

@girish no, they're pretty useless. Their web UI S3 console is such crap it can't handle the chatty API requests and keeps timing out. Also I may be wrong that multiple directories are because of failures and restarts. It just looks like multiple changed apps per day get a new dir.

So I am attempting other workarounds. Like creating a new bucket and just nuking the old one.

rsync isn't great for object store backups as it makes a ton of small files.
tgz isn't great as it's a lot of repeated information.

We need something hybrid that is the best of both.
Something like backing up to a local Minio much more quickly then doing an object to object store transfer offsite, which is much more efficient. This may also offer an opportunity to dedupe and further optimize.

@girish the stability of ZFS on linux has improved drastically since I first played with it. It's definitely worth trying now.

Also, when you tried BTRFS, it was plagued by all of the immaturity issues, ZFS didn't have at the time.

Now that distros lets you add ZFS volumes by default, it's pretty awesome for all the benefits you get.