Probably not what you’re asking for, but I have an impression, that your primary motivation is curiosity and just good feeling of using the open platform, so I figured I’ll mention it.

I’m using ESP32-C3 boards with some sensors and ESPHome to monitor air quality in my house. The board is RISC-V based and can be bought for real cheap. (single digit $ price generally) ESPHome is quite easy to work with and (If you’re realistic with your expectations around very low power device) also quite powerful.

Honestly the ESPHome itself is almost too good if you’re really curious as it abstracts the differences between various boards quite well. You’re just editing a yaml file to define your desired functionality.

Even if you’re hesitant to do some soldering, you can get pretty far if you buy board and sensors with pre-soldered pins and some jumper wires.

There’s also Rustic. It uses the same repository format as restic. It already has some pretty neat features and since latest release a ton of built-in backends.

I’m curious. How would you identify who’s guest and who’s not in this case?

With multiple networks it’s pretty easy as they are on a different network.

Obtainium will check regularly for new versions and update automatically. So that’s definitely a benefit if you’d like to keep the apps updated.

As for Mull, you could add its f-droid link into Obtainium if you’d like to have all updates via a single app.

I feel so sorry for recommending a closed source app in this community, but Genius Scan from Grizzly Labs is the only non-oss app I still use. I think I paid around €30 for the enterprise version so it doesn’t bother me with cloud nonsense.

It’s all local only (if you want) and the scanning quality is the best I’ve found. (I used OpenNoteScanner for a few months, sadly it’s not even close both in terms of quality and convenience)

I figured I’ll mention it as an alternative to MS Lens app that likely sucks in every bit of information it can get its hands on.

That specific repository has no releases so it won’t work AFAIK. You need a repository with releases, that have apk attached. (Typically the developer would set up a CI workflow to build and attach apk for every release)

Edit: For example AuroraStore has releases with apks. So you can just enter gitlab repo for AuroraStore into Obtainium and it will install it and keep it updated.

Well in that sense Rust is even more predictable than Java. In Java you can always get back exception that bubbled up the stack. Rust function would in that case return Result that you need to handle somehow before getting the value.

Predictable in what sense?

Same with Adguard Home here sitting at 38MB.

In Tailscale you can set up an exit node which lets you access the entire internet via its internet connection.

You could set up an exit node that would let you access the internet via some (anonymizing) VPN providers like Mullvad or any other.

This sounds like Tailscale is simply setting up this exit node for Mullvad on their side and providing it as a service. So it’s not like using another VPN anonymizers is impossible, it’s just convenient to use Mullvad.

RAID is not backup. RAID is used for increased capacity, throughput or uptime. (Depending on configuration)

Multiple volumes would likely get corrupted just as much with faulty RAM as RAID would. Besides RAM there’s controller, CPU, power supply and possibly more single points of failure in that NAS, that would destroy both RAID and multiple volumes.

So assuming you have external backup, I’d go with RAID for better uptime as opposed to some custom multi volume pseudo-RAID for the same.

I really like their embrace of open source. Seeing their email app on f-droid first is quite refreshing. And when they started developing it, I just subscribed to github issues with features I considered crucial for me so that I’d get notification once they were implemented.

How often do you get at least changelog with closed source apps? I’d have to check every couple months whether they implemented features I need had this not been developed in the open.

In context of self hosting it’s probably worth pointing out, that SQLite specifically mentions NFS on their How To Corrupt An SQLite Database File page.

SQLite is used in many popular services people run at home. Often as only or default option, because it does not require external service to work.

I haven’t seen anyone recommend Infomaniak Mail. I think it’s great option. It’s €1.50/month for 5 mailboxes with unlimited storage. You can add multiple domains and mailbox aliases for free. (no limit on either as far as I can tell) You get calendar and contacts as well. They also offer entire office suite, but that’s going to cost more.

They offer pretty good webmail interface, that’s not just Roundcube or other OSS webmail solutions. (which are okay, but usually limited by the fact that it’s IMAP on the backend) They offer apps for mobile calendar/contact sync and they also have (quite new, but already very good IMO) email app. These are all open source. You obviously have IMAP, CalDAV and such if you want to use your own client.

It’s not some one man show provider, they are pretty big cloud provider in Switzerland. So you also get custommer support that from my experience is pretty fast to respond.

To be fair, the onboarding experience in the app is impressively good. It’s extremely straightforward with no unnecessary buttons or steps. So I kind of see why they might be hesitant to add complexity by supporting self-hosted backend as an option.

I use Telegraf for most of the metrics.

It looks interesting as an app, but in context of self-hosting there are couple of speed bumps:

• The server side is quite complicated. (compared to Joplin for example) It’s multiple services and it also needs Mongo, Redis and S3. Makes sense for them to do it this way to be able to scale up, but for few users hosted locally it’s quite a lot of moving parts.
• You need to compile client apps to self-host. This effectively kills this as an option on iOS.

I was kind of the same, but I still collected metrics, because I just love graphs.

Over time I ended up setting alerts for failures I wish I was aware of earlier. Some examples:

• HDD monitoring - usually drive is showing signs of failure couple days before it fails, so I have time to shop around for replacement. If I had no alert set, I’d probably only notice when both sides of a mirror failed which would mean couple days of downtime, lot of work with backup restoration and very limited time to find drive for reasonable price
• networking issues - especially VPN, it’s much better to know that it is broken before you leave house
• some core services like DNS. With two Adguard instances it’s much better to be alerted when one is down, than to realize that you suddenly have no DNS when both fail and you can’t even google stuff without messing with your connection settings.
• SSD writes - same as HDDs, but in this case the alert is around 90% declared TBW lifetime claimed by manufacturer and I tend to replace them proactively as they are usually used as system disk without mirror, which holds no valuable data, but would again lead to extended unplanned downtime
• CPU usage being maxed out for long time - I had one service fail in a way where it consumed 100% of all cores. This had no impact on other services because process scheduler did its job, but I ended up burning kilowats of electricity as this continued unnoticed for weeks. This was before energy prices went up, but it was still noticeable power consumption. (Had double CPU server back then, that consumed a lot of juice when maxed out)

That’s fair. Wonder why that is, because my experience is quite the opposite.

The metrics I shared above actually had the Pihole running on much more powerful HW. (proper server with quite beefy CPU) The Adguard stats are from old Intel NUC which is perfomance-wise about on par with Rpi3B+. As you can see it barely uses any resources at all. So I’m surprised to see you reporting the performance as really bad.

I was testing Adguard on small openwrt based device and it still ran fine. Rpi3B+ has order of magnitude faster HW than that. I just don’t see how would Adguard be slower or even noticeably slow. Or even Pihole. Both could run about 40 copies of the service on single Pi3.

Whis is not to say I don’t trust you, it’s just strange.

You can’t really go wrong with any of those. They are both very solid options. Having said that, if I had to recommend one, I’d go with Adguard, because:

• The interface is better. Most notably the query log interface. Searching the logs with some long time span makes Pihole spike in memory usage and is super slow. (there’s no server-side pagination)
• Custom filters are more powerful thanks to modifiers, which AFAIK Pihole does not support. Some of it can be configured via dnsmasq (without user friendly interface), some I had not found any solution for. Good example is dnstype modifier, which I sometimes use to block AAAA responses for sites, that have set AAAA records, but the service actually does not work over IPv6. So I can disable IPv6 for certain domains if I need to. (or other way around, force IPv6 only)

Some of the above might have changed, I haven’t used Pihole for about a year.