@scottalanmiller said in What's the Best Way to Deduplicate & Organize Files/Folders on a 200 TB NAS?:

@tim_g said in What's the Best Way to Deduplicate & Organize Files/Folders on a 200 TB NAS?:

@scottalanmiller said in What's the Best Way to Deduplicate & Organize Files/Folders on a 200 TB NAS?:

@dbeato would need 256GB of RAM to attempt that with ZFS. That's a lot of RAM on a NAS.

How did you get 256GB of RAM needed?

That FreeNAS article recommends 5GB RAM per 1 TB of deduped data...
Considering he has 200TB of data he'd want to dedup, that's at least 1TB of RAM to start.

This is because dedup on ZFS/FreeNAS is much more RAM intensive than all other file systems. (and also because 200TB is a ton of data)

What caused it to balloon so much recently? Traditionally it has been 1GB per 1TB.

https://serverfault.com/questions/569354/freenas-do-i-need-1gb-per-tb-of-usable-storage-or-1gb-of-memory-per-tb-of-phys

Freebsd zfs page stated up to 5gb per 1tb last time I checked

@tim_g said in Dipping Toes Into Programming:

@matteo-nunziati said in Dipping Toes Into Programming:

isual studio code is your best bet for interpreted languages.

Yes I agree, but to start learning, I'm using Thonny. I like how it debugs through code and how you can step through it as well as it's visual help. I compared the same thing to VSC, and seems like Thonny is better for gettign started.

Yes. As a learning tool thonny is a better start. That's thonny goal in the end!

In my last job a horde of consultancy companies proposed gap analysis to us. That's a good starting point. Then you just need to be prdpared to spend a lot in useless stuff.

@Tim_G imho the best way to learn python is try to rewrite some of your PS scripts with it.

Also if you plan webapps Django is ok. For REST API flask is a better option imho.
My last application in python was done with flask appbuilder. A sort of small django.

Visual studio code is your best bet for interpreted languages.

I see you have made your choice. Btw I strongly recommend python over js for a newbie too. because the non blocking concurrent logic of js tends to blow your brain in the beginning.

@momurda said in Windows on arm this quarter!:

There will be even less applications available here than through the Windows Store for x86, which is a wasteland of malware.

Windows arm pc = linux arm pc. Wipe and reinstall. I don't see other uses in the near future.

Notebookcheck says!

@jaredbusch a bit old. maybe useful?
https://stackoverflow.com/questions/1833654/sql-server-database-schema-versioning-and-update

Also one thing to consider is that some hypervisor level backup solutions will cost in the 500€-750€ ballpark. A single paid version of veeam agent server (which does CBT) costs 130 per guest! And I cite veeam just because it is widespread.

Even with 10 machines this is a bit of a cost IMHO. In my previous work I've spent 500€ for a backup solution which managed 12 VMs - with all the fancy stuff: CBT, dedup, incremental, blablabla and yes: a single pane of plastic.

I do not think that costs are to be kept out of the equation here.

@scottalanmiller said in Hypervisor preference - costs included:

@tim_g said in Hypervisor preference - costs included:

I for one don't have the time to deal with hundreds of agents...

What time do they take? If they are part of your build and automation process, they should take essentially no more time than agentless. It's all automated, right?

Just out of curiosity: if you have not a central management system (if you like it call it a single pane of plastic - not much glass nowdays on monitors) can you programmatically setup/modify/reschedule agent based backups with -just to say- veeam agent free?!

Never tried it.

@scottalanmiller said in IS BASIC programming still in vogue?:

@matteo-nunziati said in IS BASIC programming still in vogue?:

@scottalanmiller said in IS BASIC programming still in vogue?:

@matteo-nunziati said in IS BASIC programming still in vogue?:

VS Code is basically for web languages. if you want serious compiled languages development on MS you need Visual Studio.

Not at all. VS Code isn't for web at all. It's not focused on web tech, languages, or anything else. Most modern languages use web as a main output, but VS Code has nothing making it lean towards web any more than normal VS does.

What I mean is that even if MS "sells" VSC for any language it is quite a PITA to develop stuff in C++/C/C# in it wrt VS.
I prefer Atom or VSC when developing in python, but when I've to code c++ I move to other stuff.
On MS this stuff is VS.

What's wrong with C# on VSC? I don't do much any more, but I prefer VSC for that over legacy VS still.

I like VSC more too, but debugging is way more powerful on VS (watch windows, remote debugger and similar) and when you go c++ the difference is more and more relevant.

Interpreted languages on VSC are probably even better tahn on visual studio.

Also consider that a bit more of complexity (dependency build, libs and so on) will let you do a lot of "handmade wiring" on VSC while VS has dependency build order built in.

I would surely use VSC for Python because most of the dependencies are a pip install away and I tend to build smaller projects with it. Surely VS is the place for bigger projects or lower level languages.

Also: try debug embedded devices with VSC... GOSH!

biggest issue with KVM: you have to write your own backup UNLESS you prefer agent based backup.
Otherwise it would be my choice (it has been mine for around 1 year in my previous job).
I've deployed hyperv, mostly because of the uncertain situation around xenserver. Otherwise both are OK.
VMWare asks too much for my needs. I can simply get free stuff for my needs, so any price for an hypervisor is too much.

@scottalanmiller said in IS BASIC programming still in vogue?:

@matteo-nunziati said in IS BASIC programming still in vogue?:

VS Code is basically for web languages. if you want serious compiled languages development on MS you need Visual Studio.

Not at all. VS Code isn't for web at all. It's not focused on web tech, languages, or anything else. Most modern languages use web as a main output, but VS Code has nothing making it lean towards web any more than normal VS does.

What I mean is that even if MS "sells" VSC for any language it is quite a PITA to develop stuff in C++/C/C# in it wrt VS.
I prefer Atom or VSC when developing in python, but when I've to code c++ I move to other stuff.
On MS this stuff is VS.

@scottalanmiller MS propose c# as a Platform Language, c++ as low level Language and now is in the web vagon with a ton of typescript. VB form the most remains as an app Scripting Language.
VS Code is basically for web languages. if you want serious compiled languages development on MS you need Visual Studio.

@scottalanmiller said in Do we dislike Ubuntu:

@bbigford said in Do we dislike Ubuntu:

@scottalanmiller said in Do we dislike Ubuntu:

@bbigford said in Do we dislike Ubuntu:

@scottalanmiller said in Do we dislike Ubuntu:

@bbigford said in Do we dislike Ubuntu:

@scottalanmiller said in Do we dislike Ubuntu:

@bbigford said in Do we dislike Ubuntu:

No thanks on Ubuntu. I prefer Fedora for workstations and CentOS/RHEL for servers. I haven't tried Fedora on servers but have needed more stability and less bleeding edge features. Curious to start trying it out as a server instance in a lab though.

Fedora is not bleeding edge, that's not the right way to think of it. It's just current, if anything, I'd say that it is probably the more stable of the two. CentOS is "old", not "stable". The two are not the same thing. CentOS is for specific cases where you need to maintain unchanging libraries - not something you ideally want. We specifically want Fedora for better stability over CentOS.

I chose the wrong verbiage, having read back through what I wrote. I meant bleeding edge as in latest features and consistently up to date. Where things couldn't change due to requirements where things might break, I've used CentOS. Not necessarily that it is more stable, just that things wouldn't break as easily for that reason.

I think that it makes "meaningful" breaks way more likely. CentOS does basically all the same changes as Fedora, just saves them up to make it far more painful when you have to deal with many at once. It's not like Fedora makes "more" changes, it just makes them in smaller amounts more often which protects you in many ways.

I can understand that. Having small things break, or rather maybe smaller things often rather than big breaks rarely, comes down to preference. Would I rather deal with small "noise" or a late night here and there with big stuff? If the small stuff does not effect end users to a point it is unnoticeable, but the big stuff takes down a service, those are big differences.

It's not about just breaking things, that shouldn't happen in either case. In both cases you deal with that through testing. What we are talking about between the two is whether your developers fix things "as they go" or if they save things up for years and then have to do major fixes to fix not only not keeping up as they went, but years of accumulated technical debt that could have been avoided.

I should say I haven't had any stability issues beyond some odd graphical program issues with Fedora, ever. So stability isn't really the right word for me to use there. I wish the major releases were a little longer. Certainly not as long as CentOS because those releases are far too long in my opinion, but around 1 year is just a little quick for some healthcare environments. 1 environment I managed had around 70 or so CentOS 6/7 VMs. I don't want to upgrade those every year. Every other year or every 3 years would be okay. At the same time, I'm not going to run a version or 2 behind.

Why do you want any length over the "absolute minimum", though? What's the benefit to any length at all? There is a minimum time needed for testing and support, I'm not saying to shorted that. But even Fedora holds updates after that point for up to six months to hit their "cycle". I don't want even that, every day that those updates aren't released is a day that we might be making technical debt for no reason. I understand why they do it in six month releases, they want enough time to get the whole system into a state where people can target it for package announcements. And I'm okay with that. But why would you want it longer, rather than shorter?

I use bleeding edge: pip install in venv is the way I code for the most (when I can use python). the remain of the system is of no use to me. if upstream changes the abi/api or the interpreter or the gcc (this is especially true with c++ whose iso/ansi is a moving target) I've to re-check all from scretch.
Instead I can take time to develop and switch basic palatform with enough time to test. Last time opencv libraries had to be recompiled due to c++ STD library changes.
For a one man show in R&D 6 month are nothing: you are in the middle of devel and platform is changed again . GOSH!
just to say: R&D project of a customer which is a win shop it started with xp, now it is win 10... 6 months are nothing!
it is ok if you are a team... not for one person imho.

@black3dynamite said in Do we dislike Ubuntu:

@matteo-nunziati said in Do we dislike Ubuntu:

I'm quite opposite: when developping stuff I really like to reduce the amount of maintanance of the OS. Of course too old is something I don't like, therefore I find a good balance by using ubuntu LTS and recently I've done a short test on opensuse leap too.

Yes you have no support but I don't search for it. what you have to do is installing security fixes, and I trust both ubuntu e opensuse enough to set automatic updates for fixes only. On the other hand I got stable API and ABI to develop against.

Nice to me.

I would also like debian but they have a quite short release cycle. To me ubuntu is a sort of LTS over a debian base (ubuntu is something like 75% debian - and when I install ubuntu minimal I got something like 100% debian - on the source everything is recompiled).

Debian LTS
https://wiki.debian.org/LTS

I'm aware of it, but currently it is a best effort solution not a proper one.

I'm quite opposite: when developping stuff I really like to reduce the amount of maintanance of the OS. Of course too old is something I don't like, therefore I find a good balance by using ubuntu LTS and recently I've done a short test on opensuse leap too.

Yes you have no support but I don't search for it. what you have to do is installing security fixes, and I trust both ubuntu e opensuse enough to set automatic updates for fixes only. On the other hand I got stable API and ABI to develop against.

Nice to me.

I would also like debian but they have a quite short release cycle. To me ubuntu is a sort of LTS over a debian base (ubuntu is something like 75% debian - and when I install ubuntu minimal I got something like 100% debian - on the source everything is recompiled).

@dashrender said in Buying new business desktops - what do you like?:

OK, to the end to my question seems to be - stick with status quo.

i.e. Dell Optiplex or HP Elite series.

Of course a few other names were tossed around, but the general consensus wasn't some unheard of (by me) brand or whiteboxing, etc.

+1 for optiplex

@emad-r said in NAS alternative on the cheap:

@matteo-nunziati said in NAS alternative on the cheap:

@emad-r said in NAS alternative on the cheap:

I removed 1 HDD from my test environment and it was very easy to detect and handle.

how did you remote it? usually with soft raid you would not eject a drive without purging it from RAID before...

in virtual environment you can do whatever you want.

ah! sorry I've misunderstood: I was sure you had remove a disk from a physical OBR10!

@emad-r said in NAS alternative on the cheap:

I removed 1 HDD from my test environment and it was very easy to detect and handle.

how did you remote it? usually with soft raid you would not eject a drive without purging it from RAID before...

stay away from watchguard. personal experience. really convoluted.
also more expert people here warn against UTM as a general rule.
And yes, if you need an UTM same people seem to agree on palo alto.
I've not much experience to evaluate the options.

Firewall
VPN
these are available with every router/firewall I'm aware of. Don't need an UTM.

Intrusion Prevention / Intrusion Detection
almost 40 and still I ignore what this is.

Virus Protection
what kind of virus protection? this can be implementated via a VM...