Yeah, but you can also use file level technology like par files for rar compressed files or file system level or hardware level tech that allows for repairing some level of corruption. This often uses a little more space depending on how much corruption you want it to be tolerant of, but again, unless you have a ton of data, even using 1.25x the space needed for the data itself might still be cost effective for most home users.
If the tape drive attaches with SATA like a regular DVD drive or HDD then just get an adapter that connects USB to SATA. USB is unlikely to be a bottleneck with how slow tape is anyway, so having it internal isn’t that important.
But tape drives are only worth it if you need a heck of a lot of cartridges or absolutely need 30 year storage rather than maybe 10 years for an offline physical hard drive (assuming they’re both stored properly). Though I have drives decades old that still have readable data. But even if you copy out the data every 5-10 years to a new HDD, you’re likely to spend less than just the tape drive itself. And drives get larger over time so you’ll need fewer of them whereas the tape drive will have a set maximum storage space. And what if the tape drive dies and the company is out of business? You may end up not being able to read your tapes. SATA on HHDs has continued backward compatibility for decades.
Anwyay, for a home user with a few hundred TB to store, I’d be surprised if the cost would be better for tape. It’s really useful for places that store a few hundred TB per day, though. That’s where tape can be a boon. The primary advantages are it takes up a lot less physical space per TB and lasts longer. And once you do get past the cost of the drives, it does eventually start to be cheaper.
Also, remember it’s way slower than a physical HDD much less an SSD. Even an HDD connected via USB.
Should specify, I need it exposed to the internet as well so I can sync it to those multiple devices when I’m at my partner’s place or traveling and so my partner can use it from her place.
And I’d prefer something that allows some type of SSO or external authentication mechanism so it can handle multiple logins with different calendars and I can put some layers of security in front of it. Radicale requires addon applications to do that and most aren’t maintained.
Supposedly Radicale can do it with a WSGI server, but I never found one I got working.
Only one that came close was Baikal, but it had some issues and development has been really slow on it. It’s been quite a few years though, maybe I will try that one again.
Definitely easy, but not really a relevant response given the community we’re in. Plus there’s a lot of use in having your calendar synced to multiple places, not just your phone.
I never found an easy solution to set up for caldav and carddav yet personally. Most are either very complex to set up or they are no longer maintained which is bad for security vulnerabilities. So, I’m still stuck with Google because I need my reminders everywhere I go…ADHD.
Yeah getting rid of windows solves many issues, LOL. I have always refused to run any server with windows, outside of being forced to at work. 😁
True, I was oversimplifying a bit. It’s not a full VM, but often it is a full OS depending on how the container was created. I’ve seen crappy containers that start with a full desktop version of Ubuntu for instance. They are often optimized installations for running virtually, but more for things like the base for VDIs, not individual application containers.
Based on a quick read it sounds like the issue will be that the bad actors won’t get dropped immediately. Meaning whatever servers and services are in front of the firewall that fail2ban configures, will still receive the performance hits. But your backend services will still be protected.
I have a server specifically set up as the ingress server. That runs fail2ban and Traefik and I have crowdsec set up with a Traefik bouncer plugin. So fail2ban runs in front of everything to catch the worst, most obviously bad stuff and drop the traffic very quickly and efficiently so it doesn’t get any further including things trying to access services other than the web services. Then crowdsec is focused on more complex threats as they flow through the reverse proxy. Seems to work the most efficiently for me, but I haven’t had a chance tondo any real analysis on how much improvement it might be over any other configuration.
I didn’t look too closely at details, but generally that’s relatively normal the more individual applications you have running, especially with the overhead of them all being virtualized. Often the overhead of running that many virtualized operating systems can be problematic depending on how the container author chose and configured the container’s OS. A lot of people seem to just randomly pick an OS for their containers just so they can distribute their application with docker as easily as possible, without considering the consequences.
Here are the options off the top of my head if disk wear is your primary concern:
Turn swap off at the host level. Note, though, that you may end up in bottlenecks if multiple of them do very memory intensive tasks at the same time, like image and video processing, and docker can’t swap out the VMs that are idle to the disk and be able to use the RAM fully for those active tasks. For some of those apps you can configure when they do maintenance tasks and reduce the likelihood.
Look at each individual VM in docker and see which are using the most swap and force docker to limit their allowed resources. You can even tell docker to make an application not use swap by setting the memory and memory-swap values to the same value or reduce their swap usage by setting memory-swapiness to a low number, or you can install those applications to the host system and configure the host to reduce swap or disallow swap usage through systemd or whatever rather than running them in docker. You may also find one of the applications you’re running has a memory leak issue. Good OSs often handle that, but who knows what IS the container is running and the docker layer is concealing from the host kernel and not handling quite as well as it being installed on “bare metal” might. Virtualization adds a lot of variables for these kinds of low level resource control scenarios. So just trying it on the host directly might reveal something if you’re interested in digging like I usually am, lol.
Get a second, small, inexpensive drive and set that up as the swap partition for the host.
Look into what database systems these are using. Often if they each are running their own separate DBMS containers, the DBMS maintenance overhead can cause some additional churn. I have a single instance of PostgreSQL and one of MariaDB running on a separate server that all of my applications connect to. Some of these applications may be not only spinning up a whole VM for the application, but an entirely other VM for their database server. Though I think some of these probably use sqlite which is just a file, I think manynof these use PostgreSQL and/or MySQL/MariaDB by default. This also helps because I only need a single backup for all of my databases. The negative is that if that DB server goes down, then all the apps go down too, but if you already have everything on one server that might not be as big of a deal. I have a lot of smaller servers running smaller subsets of applications based on their resource needs. Helps me conserve electricity rather than running one or two big servers that are almost always highly active. But I enjoy optimizing that kind of stuff, it’s tedious.
Yeah FUTO uses a “source first” model of licensing that isn’t fully FOSS. It does mean all the source code is available to everyone. But it is only for non-commercial use even of derivative works. And a few other limitations that violate some of the tenants of FOSS. Personally, I don’t see a major issue as long as the company remains as it is. The issue with this kind of thing is a lot of time these nonprofits get acquired, consensually or hostilely, and the new company revokes some things and makes it all proprietary and then sues everyone for profit. Or other really shady things can happen if licenses aren’t open enough. Of course the flip side is that big corporations rarely obey copyleft licenses or other things, especially now that LLMs do the stealing of code for them.
So, for now GrayJay and some other FUTO apps are sufficiently open and free for me, but that could change.
Yeah, thanks for the clarification. I know I infodumped a bit, but when it comes to safety I tend to like to make sure to mention the worst case scenarios.
If you will have additional support and a shelf below it, it might be ok to use, but yeah without a drill to make some new holes in some ears meant for another device it’s hard to say if they’d be compatible. Hole patters like that which attach on the device as opposed to the rack or standard components can really be any pattern. They have to account for both the structural integrity of the case of the device as well as what is underneath and thus would be exposed by the holes.
I just did a quick search on the model and looked at the first spec sheet that popped up, so not sure my info is correct, but its says it’s 1U and about 50lbs. That seems like it’s pushing the ability to use front ears alone. I saw something that mentioned a lot that came with adjustable depth rails, not ears. But again could be a slightly different model or different battery configuration or something.
Anyway, I’ve seen 2U ears rated for like 60lbs or so. And those are sometimes pushing it IMHO. But 1U at 50lbs would have to be extra thick and hardened to stand the torque with how narrow they’d be just based on some head physics/math, but I could be wrong.
But anyway of use use something not rated for it you might get away with it for a bit, but it would likely be really fatigued over time, especially as the rack is, bumped, shake, or moved, not to mention in an earthquake. Of it would snap and fall onto whatever is directly under it. And if there’s more than an inch or so of space there, that could be disastrous. Like what if normally you have a shelf directly under it which might be providing structural integrity to the rails to not move much. But you remove the shelf while reconfiguring, the rails flex slightly as they do, and put just enough extra torque on the ears to snap them. Now whatever was under that shelf might be a few inches lower and just that distance with 50lbs could easily destroy an unshielded component or 10. Or badly damage sheilded ones. I always think worst case whenever I’m going to exceed rated standadards even if the at may never happen. But it’s definitely not unlikely.
Personally, I’d either use side rails like the original kit said it came with, a shelf rated for the weight, or possibly both front and back ears if you have a shallow depth rack that fits it.
I hate to mention really bad options, but If you’re set on ears, it’s possible you could find some and maybe line one or two holes and drill out any that don’t line up. But make sure to take into account your compromising the weight rating if you drill. I have also seen setups with a 1U device attached to 2U ears. They attached the device at the top and then had two custom rails/beams angled up from the bottom of the ears attached where the bottom 2U device hole would have been and angled up to the rear bottom of the 1U device and somehow attached there, drilled holes and bent the ends flat or around the back I think. That gave her the additional stability to reduce the torque put on the ears when the rack moved or was bumped. Seemed like more work than it was worth, but she also had another narrow component under the UPS and these rails were positioned outside the width of that, so she didn’t lose the extra 1U from the too tall ears. But I think she might have gotten a lot of stuff free or super low cost from “dumpster diving” and she did have mad skills like that.
But I don’t do custom stuff like that on my own, expensive devices. Might add some extra support if something looks too wobbly but is still in spec, but I’d never compromise a support part, much less add my own part not intended for that use without some serious math I’m probably not interested in doing, LOL.
I know that’s a lot and I wrote quickly and didn’t have time to clean up my thoughts or double check specs, but I like to show options and consequences of choices. Usually the “right way” is more expensive initially, but saves hundreds or thousands of times as much money over the “wrong way” failing. 🙂
So find verified specs for the device, or just measure and weigh it on a bathroom scale. Then find similar parts and see what they’re rated for and don’t use parts for 2U on 1U thst doesn’t give you the support or rating listed since your attachment footprint is only half on the device, unless you’re confident enough in your metalworking skills, geometry and materials science to do what “that girl” did with the support beams, lol.
No need to have an external domain. And you can set up an internal DNS to route any domain name to any IP address. Just be sure not to point anything to your external IP address from your ISP or any globally addressable IPv6 addresses.
Issue is that it might confuse your devices that jump between internal access and external access like your phone if you use a domain name that’s owned by someone else, like example.com. I have two domains, one for my VPSs and one for internal stuff. So when my devices are outside the network they can’t resolve any of the DNS addresses that are set up inside and can’t access any of the IP addresses. That is unless I connect with vpn. I also have pangolin for a few things that are accessible externally, but that’s rare since apps can’t use those kinds of authentication systems, just web front ends.
What are the differences? I always wish forks would state how they are different, but this one pretty much just says it “diverged substantially”. Is it significant features were missing? Is it a different ethical or licensing model or different views on “vibe coding”? I’m always turned off by a project with a long AGENTS.md. It usually means “AI” is not well controlled if that much guidance is needed just for preferred coding style. And that typically means usually poor efficiency, stability, and security in exchange for fast feature implementation.
It’s what always happens. People never learn. Discogs, crowdsec, google maps, all of the things that have ever benefitted from crowd sourcing inevitably end up locking up the information and it becomes polluted with junk when they have to start using poorly sourced and moderated commercial data.
I usually use my phone with WiFi off or set up my phone as a hotspot while not connected to WiFi so my laptop can connect through my cell carrier’s network. Always helps for testing external connections quickly and easily.
GrayJay is what I usually use. It’s working currently.
With a reverse proxy there’s no need for hoops unless you add some authentication mechanism on top of that. I use reverse proxies for several things so that they flow through filtering like crowdsec or fail2ban. It just routes the single port to the device on question as is. The idea is to make it indistinguishable from a fully exposed port to the user. Just have to make sure whatever system runs the reverse proxy is secured. I always make sure that nothing significant is on the system that is exposed to the internet other than the reverse proxy server and all other ports are closed internally and externally. Ideally a device that has multiple Ethernet ports works well so you can keep an ssh port open on the internally facing ethernet port for troubleshooting. Nice if they are 2.5Gbit or 10Gbit ports, too, since they route a lot of traffic. But not needed.
Yubikeys work great for me for 2FA or passwordless auth. NFC and USB interfaces. Very secure. Pretty durable, though USB-C is inherently less durable due to the size and complexity than the USB-A footprint. But I’ve never had an issue with mine in my pocket or purse right with my keys. But if you bend it while it’s inserted in something it can be an issue.
This would be the 3rd time since I’ve been using it that it has broken compatibility with Vaultwarden, but Vaultwarden always fixes it pretty quickly. But usually browser plugins end up updating before I get around to updating Vaultwarden.
That sucks. It was one of the few GUIs that did a good job with most aspects of Docker Swarm. I knew it was heading in a shitty direction, but I was hoping something else would come along to fill the gap in the mean time.