Podman support and optional MFA for WebUI login. 

On 7/27/2024 at 1:04 PM, Doppler said:

Not using it at the moment, it doesn't really fit with what I need. I don't want it moving the files by age, just on a daily/weekly/monthly routine. If I moved them by age I would never achieve my 80 percent standby rate for my drives. I prefer moving it in bulk after the files "lifespan" has passed. (How long I typically work/need it immediately) I add files every hour or so, and as far as I know it would just keep moving them over and never let my drives spin down. Maybe it can be set up differently but I do not know how. That's why a share by share use-case is far more important. Integrated functionality is always good. Maybe a "Pool" Generalization/Default setting that can be overridden on random shares. That would make it nice for different cache pools that take in different types of data. Which would affect all the shares using said pool. Just overkill but a more convenient initial setup. That was a whole word salad of a post, sorry, am a bit exhausted.  

Mover tuning plug-in can also do moves based on disk fill % so you can use it to have the mover only fire when it absolutely needs to if you want to keep your array spun down more often. and you can do per-share tuning for it too, still all run from the one scheduler though so you cant have one share move nightly, one move weekly and another only move when SSD is full for example.

 

Things like that are why I would like more mover controls built into the OS and not rely on layers of plug-ins and custom scripts to do these more complex setups. now that we have the ability to do multiple pools and eventually multiple arrays it will have to get more complex and configurable at some point.

I want an XFS pool. RAID is not needed.

I simply want to group NVMe drives into a single pool for data storage purposes.

 

ZFS uses RAID, while Btrfs can create a pool with a single disk, but I think XFS is more reliable.

Edited August 2, 20241 yr by unraidk
explained in detail

On 8/1/2024 at 10:35 PM, unraidk said:

I want an XFS pool. RAID is not needed.

 

So you want an "Unraid" pool. That's confirmed as coming, but no date/version has been announced for it yet. Hopefully in a future 7.0 beta.

On 8/1/2024 at 9:15 PM, Faceman said:

Mover tuning plug-in can also do moves based on disk fill % so you can use it to have the mover only fire when it absolutely needs to if you want to keep your array spun down more often. and you can do per-share tuning for it too, still all run from the one scheduler though so you cant have one share move nightly, one move weekly and another only move when SSD is full for example.

 

Things like that are why I would like more mover controls built into the OS and not rely on layers of plug-ins and custom scripts to do these more complex setups. now that we have the ability to do multiple pools and eventually multiple arrays it will have to get more complex and configurable at some point.

I suppose I could do something like that. I also wish plugins weren't "required" to have some of those nicer features. Thank you for the idea.

17 minutes ago, Doppler said:

I suppose I could do something like that. I also wish plugins weren't "required" to have some of those nicer features. Thank you for the idea.

Having solutions as plugins allows for faster development changes outside of core development timelines.

and a more streamlined and stable base OS for the average user that doesn't need every possible bell and whistle.

 

16 hours ago, Faceman said:

and a more streamlined and stable base OS for the average user that doesn't need every possible bell and whistle.

 

Can you be more specific?
are you wanting unraid on example(debain/ubuntu) host os and not Slackware or are you wanting a secure predefined out of box in slackware?

37 minutes ago, bmartino1 said:

Can you be more specific?

I believe the intent of this post was to add to Simon's explanation of why plugins are good instead of embedding everything in the OS by default.

 

On 8/5/2024 at 3:18 PM, SimonF said:

Having solutions as plugins allows for faster development changes outside of core development timelines.

 

17 hours ago, Faceman said:

and a more streamlined and stable base OS for the average user that doesn't need every possible bell and whistle.

 

 

On 8/5/2024 at 3:18 PM, SimonF said:

Having solutions as plugins allows for faster development changes outside of core development timelines.

 

The problem is discoverability - It's not obvious what's missing nor what's available as a plugin to a new or even seasoned user.

This can be mitigated by having some plugins be part of a base install. This allows updating them outside of the core timeline, while keeping them obvious/known.

 

 

Edited August 8, 20241 yr by Espressomatic

2 hours ago, Espressomatic said:

 

The problem is discoverability - It's not obvious what's missing nor what's available as a plugin to a new or even seasoned user.

This can be mitigated by having some plugins be part of a base install. This allows updating them outside of the core timeline, while keeping them obvious/known.

 

 

The problem with building the plugins in is that it means Limetech have to then effectively take over maintaining them since plugins can install components that can break the core functionality.   

 

There have been a number of plugins whose core functionality has later been built into Unraid when they were deemed stable enough.

 

I would think a better (and easier to maintain) option would be to have them detailed in a section in the online documentation.   That would both make them more discoverable, but also leave it to the the users choice as to whether to install them or not.   It would also be the  place that most of these plugins are 3rd party and not maintained and/or tested directly by Limetech.

2 hours ago, itimpi said:

There have been a number of plugins whose core functionality has later been built into Unraid when they were deemed stable enough.

 

That being said, aside from stability, more importantly not everyone needs every functionality and plugins are a good way to keep the OS debloated while letting the users choose what they want or need themselves. 🙂 

These I consider critical and core to Unraid's broader functionality

Community Applications

CA Auto Update Applications

Unassigned Devices

Unassigned Devices Plus

Unraid Connect

 

A setup wizard with information and the option to enable a number of plugins would go a long way to help people get the most from the system. Reading the manual is going to help the 1% or fewer people who do that as they're setting up. 🤣

 

 

It's a difficult list to propose, what is critical for one is not for someone else.

 

25 minutes ago, Espressomatic said:

These I consider critical and core to Unraid's broader functionality
 

CA Auto Update Applications

 

For example, I don't want anything to auto-update without my knowledge.

 

What I did at the time (and still do on occasion) is to go to CA, Plugins sections and sort by Downloads. That somehow provides an idea of what other people use.

30 minutes ago, ChatNoir said:

For example, I don't want anything to auto-update without my knowledge.

 

I don't use it for everything. And it starts OFF when installed. But when someone does want this functionality, it's not exactly something most people would suspect they need to go search for a plugin to achieve, seeing as it's the kind of setting that's available as part of most core operating systems and applications they're used to using.

 

Making things easier means a wider audience. Which means more sales potential. You can see in this very forum people questioning the fact certain "add-ons" don't come preinstalled and whether they might disappear. Docker for instance, via CA. Anyone even faintly familiar with the TrueNAS apps (charts?) fiasco can see why.

 

 

31 minutes ago, Espressomatic said:

 

I don't use it for everything. And it starts OFF when installed. But when someone does want this functionality, it's not exactly something most people would suspect they need to go search for a plugin to achieve, seeing as it's the kind of setting that's available as part of most core operating systems and applications they're used to using.

 

Making things easier means a wider audience. Which means more sales potential. You can see in this very forum people questioning the fact certain "add-ons" don't come preinstalled and whether they might disappear. Docker for instance, via CA. Anyone even faintly familiar with the TrueNAS apps (charts?) fiasco can see why.

 

 

 

I doubt CA is going anywhere, but even if that were the case it's always possible to install plugins via direct URL (as it was before CA existed). I could personally picture a screen with popular plugins being suggested to the user for installation (as part of the initial OS setup), but I wouldn't like to see anything come pre-installed personally.

I'm fine with the way things are for my personal use. I'm just thinking out loud about what can be beneficial for a higher-level customer who doesn't have either our experience, or knack for digging up technical info.

 

There's still a big market out there for the very tech-literate crowd, but the market for more consumer-focused individuals is huge. People who aren't going to even be looking at things like TruNAS, ProxMox, etc. But might be headed toward proprietary solutions from QNAP, Synology, etc. and might not mind stepping up to a simple bolt-together alternative.

On a personal level, I'm more interested in remote management of Unraid systems that I can deploy for others who would love the benefits of what this type of system can offer, but aren't necessarily up to doing even a pre-made Synology solution. So not quite a homelab for those ppl, but a storage and application server that "just works" (with a casual maintenance schedule) - plex, home assistant, etc.

Edited August 8, 20241 yr by Espressomatic

Might be a good idea to explain what you mean by each suggested option for the less knowledgeable.

Is it possible to add changelists to docker app updates?

 

For example frigate has some breaking changes but if I just hit update I have no idea of them.

Its pretty easy to show a GitHub link or text & would be super helpful.

 

 

Multiple Arrays please 🙏 

Im boring so native iscsi support

27 minutes ago, isvein said:

Im boring so native iscsi support

So do you mean being able to enable a share or a device as a target etc from within the main GUI?

7 minutes ago, SimonF said:

So do you mean being able to enable a share or a device as a target etc from within the main GUI?

Yup, create a target from share (fileIO), device and zvol
Same as the plugins do, but with an better gui and native.
The plugin work just fine but the gui is a little backwards

Another thing that would be useful:

Gui for creating docker networks.

Yes its easy to create a simple one in terminal but the syntax for more complex ones easy gets very complex.

On 8/8/2024 at 7:37 PM, Espressomatic said:

I'm fine with the way things are for my personal use. I'm just thinking out loud about what can be beneficial for a higher-level customer who doesn't have either our experience, or knack for digging up technical info.

 

There's still a big market out there for the very tech-literate crowd, but the market for more consumer-focused individuals is huge. People who aren't going to even be looking at things like TruNAS, ProxMox, etc. But might be headed toward proprietary solutions from QNAP, Synology, etc. and might not mind stepping up to a simple bolt-together alternative.

On a personal level, I'm more interested in remote management of Unraid systems that I can deploy for others who would love the benefits of what this type of system can offer, but aren't necessarily up to doing even a pre-made Synology solution. So not quite a homelab for those ppl, but a storage and application server that "just works" (with a casual maintenance schedule) - plex, home assistant, etc.

I agree with the above. I think there are some commercial decisions which could be made here - specifically to increase adoption and market share away from ready-made solutions like Synology and Qnap.

 

I have only been using Unraid for a fortnight, and was tempted by the “Unraid array” option as I have outgrown my Synology NAS in terms of drive bays. The thought of having a 30-drive array was VERY appealing, and I equated this to the SHR/SHR2 format that Synology offers - parity protected, can grown in drive capacity over time, can use variable size drives. When I found out I could only set this up as a single array (and then had to use other pool options), I was a bit disappointed.

 

However, 14 days of playing has convinced me Unraid is the right solution for me and I’ve just bought a lifetime license.

 

I would consider myself fairly technical - I’m not a sysadmin, dev, etc in everyday life, but am an experienced ‘tinkerer’ and home enthusiast. My suggestion would be to create wizards in the system to help with core functionality:

 

Initial config

How do you want to use the system?

NAS-type device (no VM or Docker)

Apps & Services (docker and/or VMs)

All rounder (show me everything)

 

Create a storage space

What is the purpose you’re trying to achieve?

Parity protected storage long-term

Unprotected, fast, SSD storage

Etc

 

This is LESS about the file system type and more about the end goal. Less technical users won’t know the difference between xfs, zfs, btrfs, etc. and most won’t care. They want to know their data is protected against a couple of drive failures, if it’s fast for multimedia creation, if it’s safe for long time storage of documents, etc. 

 

Create a cache

Outline WHY this would help

What drive types are best (SSD or NVME)

Create as a read or write cache (again, Synology offers this)

 

If you’re looking to tempt people away from the other ready-built players, then you have a real opportunity to not only win market share but also educate your customer base. Give simple or advanced options for config, remember that not everyone has been using the software for years and it can be very daunting! YouTube and the community forums have been SUPER helpful for me, but there is an opportunity for Limetech to create some basic e-learning (which isn’t hard these days!). System overview, understanding different filesystem types, docker management, VM management, community applications, etc. help build up knowledge so users become more invested in your ecosystem. You could possibly look at certification options for installers/professional users too?

 

Lastly, could you publish a planned development roadmap? For example, there’s lots of “ooh multiple Unraid arrays will be supported” but no idea of when this will likely be, so people end up drawing their own conclusions which can lead to misinformation being spread and potentially hurting the product. Could you pin down features (and requests you’re progressing from this thread) into a basic roadmap:

V7.0 - major release (clearly outlining features like you’ve done with the beta releases, but AHEAD of release)

V7.1 - multiple unraid arrays, new mobile UI, iOS and android unraid connect companion app

V7.2. - etc

 

I don’t think anyone would be upset if features get bumped to a later release and an explanation given ahead of time, and it’s not something to hold you to account over (well, maybe a little), but it would help manage customer expectations and let everyone understand where their requests feature, plan for any major upcoming changes, etc. This should be fairly easy if you’re using an Agile-based sprint planning cycle or Kanban approach?

 

One final thing… I’d personally like to thank the Limetech team for your hard work, community engagement, and thoughtful development. It shows true passion for your product, and means that it continues to meet the needs of your customers. Lots of companies don’t get this right, but from what I’ve seen so far; you are nailing it!

 

I’d be very happy to chat through with the developers or other Limetech colleagues to give a total “newbie” view or feedback if that’s of interest.

Edited August 14, 20241 yr by Matthewmilesuk
Spelling mistake