Regarding size, I did a regular debian 10 netinstall. And when running the xoce installation script after the installation, there was an additional 230 MB of packages and data downloaded.

That means a total of 350+230 = 580 MB have to be downloaded to install the OS and build XO from scratch.

Debian iso and packages downloads very quickly as they are sponsored by a CDN (Fastly) that delivers their files.

Anyway starting from nothing, it took about 7 minutes to download everything needed, install debian unattended and build xo with the script and have everything up and running.

Some of the steps I did manually like create the VM because we haven't fully automated everything yet. I only took actual running time for the server in account though.

@PhlipElder said in NVMe and RAID?:

They are a perfect board for our cluster storage nodes with two built-in 10GbE ports. An AMD EPYC Rome 7262 processor, 96GB or 192GB of ECC Memory, four NVMe via SlimSAS x8 on board, and up to twelve SATA SSDs or HDDs for capacity and we have a winner.

Just a side note - 4 NVMe drives is a typical Supermicro config that they have on a plethora of motherboards and chassis. So your config is not unusual at all and you could buy one off the shelf from Supermicro. Supermicro is not HPE or Dell - they probably have 20 times as many models, maybe more. And they cater to OEM system builders.

Maybe just use wordpress.com? I think it's free.

@dbeato said in On prem Exchange hardware questions.:

@JasGot said in On prem Exchange hardware questions.:

@scottalanmiller said in On prem Exchange hardware questions.:

SSDs... yes, unless you have a special case with extreme amounts of storage (like 30+ TB) combined with little need for performance, you should just use SSD and never have spinners even on the table.

Read Intensive? Write Intensive? or Mixed use?

Write intensive, that is what Exchange does most of the time.

No, read intensive is what you want. It's Dell's description of the write endurance of the drive, not the workload per se.

Read intensive SSDs normally have about 1 DWPD for 5 years. DWPD is drive writes per day so if the drive is 480GB you can write 480GB of data each day for 5 years straight.

Put another way, you can write 876 TB of data to the drive during it's 5 year lifespan.

Mixed use is usually 3 DWPD and write intensive 5-10 DWPD.

A mail server will never need anything more than read intensive drives.
2 drives in RAID1 will do the job in this case - without breaking a sweat.

@siringo said in Windows Servers Archived onto Linux, suggestions/help pls:

I have a client that wants to repurpose 3 Windows Servers that are no longer used but contain data that needs to be kept.

The data can be read by specialised software that is installed on the servers.

I'm thinking of running up a Linux box and having these servers run as VMs on the Linux box so the data can still be accessed.

Remember licensing. MS Server is licensed after physical cores on the hypervisor, not the VMs. So moving from a physical server to a virtual might require higher licensing costs.

Same goes for the software that's on these servers. You have to know what it is and how it's licensed.

This how you remove linux partitions and other partitions in Windows, for instance on an USB drive.

Start a command prompt with Admin rights

For instance by typing cmd in the windows start menu and right click on cmd.exe and select Run as Administrator.

Start diskpart

Start Windows disk partition manager:

diskpart

Show all attached disks:

list disk

Select which disk you want to remove partitions from:

select disk 1

Remove one partition

To remove one partition first list what partitions you have:

list partition

Then select the partition you want to remove:

select partition 1

And then delete it:

delete partition

Note that it will delete the partition immediately.

Remove all partitions

To remove all partitions on the selected disk do:

clean all

Finish

To exit the program:

exit

Note: There is no need to save or write changes, as all operations are performed straight away.

@JasGot said in How To: for files:

@Pete-S said in How To: for files:

Don't be sloppy, keep the case consistent.

Pretty code is his problem not mine. "/M" is from the man pages.

Why would you alter his code (/s ) without knowing if he wanted subs or not? He never addressed that. Further, the issue was deleting, not going too deep.

See if it works? Are we debugging his code? Or teaching him how to write pretty code?

You could have made your remarks without quoting me and would have been successful without being offensive.

You're right and I apologize. It wasn't my intention to offend anyone and I while I was writing in a hurry, I could have worded it differently.

@gjacobse said in How To: for files:

That said, from what you are saying; is my ‘failure’ was omitting the /M which specifies the ‘searchmask’. And because of that omission, it didn’t have the comparative variable, and deleted everything.

There were two things that tripped you up.

• The lack of /M to just find *.tmp files
• Using del *.tmp as the command

First forfiles is a for-loop that executes whatever command you want for each file it finds.
As @JasGot mentioned you will get one hit for each file it will find.

With /D -15 you are specifying that you only want to find files older than 15 days.
With /S you are telling forfiles to look in all subdirectories as well. You may or may not want that.
With /M you can specify a filter which matching filenames you want to search for. Without /Mit will find all the files regardless if it's named 123.tmp or installation.txt

Your initial command would find all files in all subdirectories regardless if it what name it had. That's why you needed /M.

Secondly, forfiles will execute whatever command you want for each and every file it finds.
Since your command was del *.tmp it would delete all tmp files every time it found a file.
What you really wanted was to just delete the file it found.
The @file will have the name of the file so del @file would only delete the file that was found.

@EddieJennings said in Script for Creating VMs from Template VM in KVM:

@travisdh1 said in Script for Creating VMs from Template VM in KVM:

@EddieJennings said in Script for Creating VMs from Template VM in KVM:

@Pete-S said in Script for Creating VMs from Template VM in KVM:

Not the exactly the same thing but you might want to look into how to create a VM from scratch.
Meaning a script that will set up a VM with vCPU, memory, storage, network etc and then boot it from iso and have it do an unattended install, create what users you want and install the packages you need.

That's one of the next things I'm looking into.

@EddieJennings Also remember about things like kickstart in RedHat based operating systems. In Fedora/CentOS/RHOS you can use a kickstart file to automatically select all the install time options for the OS. A short time later you've got a fresh server and all the time it took you to setup was running the creation script on your hypervisor.

One of the things I'll need to figure out going the Kickstart route is setting the hostname what I want it to be at the time of installation. Likely not difficult to do, I just have to figure it out. Or perhaps, I can just truly take the approach of just making a clean minimal install, and then later configure to whatever specific thing I'm wanting the VM to do for my lab / testing.

Inside the kickstart file you'll find something like this:

network  --hostname=centos8-4.example.com

We use debian as our goto and then it's called a preseed file. The only real thing that can be tricky is to tell the installation what kickstart/preseed file you want to use. You can do it in different ways. If you don't want to rely on dhcp/tftp/pxe etc you can roll your own iso file. I think the kickstart file can also be mounted as a drive that the installation will detect when it starts.

I think the best approach is to make an automated installation with same basic settings and some of those will get changed later in the installation. For example you can use a fixed hostname that is later changed from ansible.

Look and feel matters so everything customer facing is done by pros, not by the employees themselves - unless it happens to be the business of the company.

A real business also have procedures for almost everything. So how to handle a customers request/complaint/whatever is in 99% of all cases not a spur-of-the-moment kind of thing.

@scottalanmiller said in Topics of Systems Administration:

If you were writing a book on systems administration, or you wanted to read one, what topics do you feel would be best to cover? This isn't like specific to an OS particularly, but more general. If you were thinking about a book or guide to overall systems administration or engineering, what would you cover or want covered?

Examples:

• Taking Backups
• Scheduling Reboots
• The Role of the Filesystem

Big picture.

It's easy to get lost in details. I would want to delve in to what the job actually entails and how one would go about organizing it.

And then from big picture down to technical solutions and then how-to's.

@frodooftheshire said in Vertically wall mounting Dell R640 Chassis:

Hi Guys,

I have a client that's moving into a new building. The IT closet was supposed to be large enough to house a proper 4 post rack, but now it looks like things have been downsized considerably and I now have to look at wall mounting the R640. Historically I always use 4 post racks with Dell's rack rails so I'm a little out of my element and I'm not seeing a lot of people talking about vertically mounting these servers.

I've been looking at two different racks at racksolutions.com:

1. https://www.racksolutions.com/wallmount-rack.html

2. https://www.racksolutions.com/wall-racks.html?gclid=Cj0KCQiAk53-BRD0ARIsAJuNhpvM0JnX3ZfdWs8ccBpNznxFSfk9mokI6HvUTDQCJiAJ9q3jPsrTo6AaAgJAEALw_wcB

The problem is...neither one of these solutions show up as compatible with Dell's servers. Has anyone wall mounted of these servers before? Can I just use some cage nuts/m16 screws the the generic bracket (option 2)? I just know that the front fascia is designed for Dell's rails - don't recall if there is space for screws on either side.

Sorry but I think you can dismiss the entire idea. Servers are made to be horizontal for proper cooling. The racks you're been looking at are switch racks. They can't handle the weight of a server.

Look for hosting the servers in a colo. Or move to other smaller servers, like Dell tower servers.

We've been using Logitech C920 and they've been good. But we got them before corona.

@travisdh1 said in Light weight Distro for VMs:

@DustinB3403 said in Light weight Distro for VMs:

Ubuntu server with cockpit and KVM is probably as light weight as you'd need. The types of services you're talking about wanting to run don't need any sort of high performance.

But why replace an rpi with a VM, other uses planned for the rpi?

Fedora Server minimal install is a good lightweight base. Ubuntu server minimal might work as well, but not as easy to use imo.

I don't like most distributions that are made to be very lightweight because they are hard to work with. IE: Puppy Linux, forget running 90% of the software your used to having available with it, at least not without great deals of unnecessary complexity.

I agree with @DustinB3403, why put this old power hog in place of the Raspberri Pi? It'll cost you way more in just power use.

Some of the really light weight stuff uses busybox and boots from compressed images and what not.

Otherwise a minimal install of debian is the least resource demanding of the major distros.

The question I would ask myself is if a simple form is really what you need.

In a lot of cases you have some kind of business process behind that form. And you'll get more value out of the whole thing if you can add that into the process as well.

So questions I would ask would be:

• What happens with the information in the form after it is filled out?
• What happens if the form is incomplete (is it even possible)?
• Is there a situation where the form would need to be filled out partially and then later adding the missing pieces?
• Could the information in the form make something else happen in the organization?

As I see it there are three options for a SaaS solution:

• simple forms
• low-code app builders
• business process managers

If you need to track the information in the form after it was initially filled out you are probably better of with something more than a simple form.

In the zoho ecosystem the applications to look at would be zoho forms, zoho creator (low-code apps) and zoho orchestly (bpm).

@scottalanmiller said in Access 2003 in a 2021 World???:

@Carnival-Boy said in Access 2003 in a 2021 World???:

@scottalanmiller said in Access 2003 in a 2021 World???:

This... I need to make a video on why buying an ERP is almost never the right answer. Almost every customer that we have that we manage an ERP for is sorry that they didn't spend the same (or less) money building something that they could control and customize.

I'd like to see that video. I've never heard of a company being sorry in the way you describe. I'd be interested to know why your customers feel like this.

I don't know how you could possibly build something for less - developing an ERP system from scratch is a massive expense. I think it would be ridiculous to even think about it.

I'll add that to the video list But the fast economic basics are that off the shelf ERPs typically require a staggering amount of customization at rates that normally exceed bespoke development (meaning per hour cost is higher because it's neither fun nor rewarding nor using good tools) and COTS ERPs are normally bloating with tons of unneeded components that most customers will never need. And making a bespoke system means ground up design for the needs of the customer, lower development costs, and only making what is needed.

So your thought that an ERP would be staggering to make.. I totally agree. But what gives the bespoke approach a chance is that they tend to only need to remake maybe 20% of what goes into a standard ERP. Every customer needs a different 20%, of course, that's why there isn't one ERP to rule them all.

This is why many ERP vendors only have a handful of customers. Every little industry has a bunch of specialty ERP players that have just a few customers and yet still make money.

I'll also add to that a lot of companies still have a "standard" ERP but have other systems that ties into the ERP. For instance a manufacturer may use SAP as their ERP and they might plan their production in SAP but actually executing and keeping track of the production is not done in SAP and they'll use their own custom made solution for that.

PS. I wrote "standard" ERP above because in the case of the typical SAP clients the ERP is usually heavily customized. Every company buys into the solution thinking they only need some "minor" customization but 10 years and lots of $$$$$ later, they realize that's not the case.

Change your keyboard settings to the English US international. You'll get some additional AltGr+key combinations for Spanish and other languages.

Yes, you can copy the files directly if you want - assuming the drive isn't encrypted.

For Firefox for instance:
https://support.mozilla.org/en-US/kb/restore-bookmarks-from-backup-or-move-them

@Dashrender said in Alternative to FTP:

That said - I'd be looking at - why does it need to be local? any way to have this potentially hosted and do the work on the file in that hosted location? You're much more likely able to get great large bandwidth at low cost into a hosted solution (or colo) and not affect the people on your local network.

It's possible that its video files, raw image files or something of that nature that they want to start working with immediately. And that is going to be local files.

If it's the same clients doing the same kind of file transfer on a regular basis and you want the minimum amount of time from the start of the transfer to the file is local, I agree with @scottalanmiller that sftp would be the easiest. Have the clients install Filezilla and use that to transfer the files.

@Danp said in Need help with Autohotkey Windows:

@Pete-S Seems like someone already created the script for you -- https://www.computerhope.com/tips/tip224.htm

Edit: The above is based on this -- https://github.com/pmb6tz/windows-desktop-switcher

Thanks @Danp !

I had a look and it's complicated because you need DLLs and whatnot just to figure out which Desktop you are actually on. But then the script just runs Ctrl+Win+arrow to switch to the right Desktop.

So I borrowed that little bit but the rest went into the bin. I just run enough Ctrl+Win+Left_arrow to make sure I'm on the first desktop and then go from there.

This is what I have now (it assumes 8 virtual desktops present and uses Win+F1 to Win+F8 to switch between them):

#NoEnv  
#Warn  
#SingleInstance Force
SendMode Event

SetVirtualDesktop(DesktopNumber)
{
   Sleep 100
   Send #^{Left 8}
   if (DesktopNumber>0) {
      Sleep 100
      Send #^{Right %DesktopNumber%}
   }
}

#F1::SetVirtualDesktop(0) 
#F2::SetVirtualDesktop(1)
#F3::SetVirtualDesktop(2) 
#F4::SetVirtualDesktop(3)
#F5::SetVirtualDesktop(4)
#F6::SetVirtualDesktop(5)
#F7::SetVirtualDesktop(6)
#F8::SetVirtualDesktop(7)