@scottalanmiller said in Trying to correctly understand core licensing in a vmware environment:

@dave247 said in Trying to correctly understand core licensing in a vmware environment:

@scottalanmiller said in Trying to correctly understand core licensing in a vmware environment:

@dave247 said in Trying to correctly understand core licensing in a vmware environment:

@scottalanmiller said in Trying to correctly understand core licensing in a vmware environment:

Now to make things more complicated, if you didn't want cluster mobility, but only failover, you can move to SA licensing which isn't cheap, but is way cheaper than this. But if you wanted that, the three node cluster doesn't make sense. So based on the example, SA doesn't cover the goals.

So this is the first time I think I've heard of Software Assurance, and I've just looked it up and done a bit of reading. I understood the words that I read, but it's not really clear to me what exactly SA is or how it could help me in this situation... would you mind helping me understand?

SA is a "maintenance plan" for your Microsoft software products that are not SaaS (so MS office, Windows desktop, Windows Server, Exchange, SQL Server, etc.)

Basically, if you buy software the non-SA way, you pay for it and then... that's it. One time cost, and that's all you pay and.. that's all you get. It's very straightforward.

With SA, you get several benefits, but essentially there is one big one that is what SA is really about - you pay a small amount annually and you get upgrade rights. So, as long as you maintain your SA on a product, it changes it from...

Non-SA: "I bought Windows Server 2016!"

to

SA: "I bought Windows Server!"

You no longer care about the "version" that you buy, your SA means you have the right to use any recent version (so right now that's something like 2008 R2, 2012, 2012 R2, 2016, and 2019 in a few weeks.) You can deploy any of them that make sense today, and upgrade anytime that you want. No more "paying for the next version."

You get lots of silly extra like training and "support", but it's all minor. Sometimes certain features like maybe VDI are included with SA only, in those cases it can be a bigger deal.

In the Windows world, SA is often just considered a part of the license cost and most people ignore that there is even another way to buy Windows because it's the cheaper way to keep your Windows running healthy. Paying for each new version is more costly and more complex. But in the extreme small business arena, skipping SA is common and is often a major contributing factor to finding shops with extremely outdated systems because the pain of the full cost for the next upgrade was too much, but SA makes it a small, annual fee every year instead which is way easier to budget for.

So, does that mean if I installed 2008 R2 several years ago, with SA, I can move that server to a 2016 box for free? Otherwise I still don't understand.

Yes. Where "free" = "already paid for SA". So no additional cost.

Note, SA is a subscription. SA is normally sold for 3 year terms, so at this point you would have paid for possibly 2 renewals. Given the price of the SA renewals it would have been cheaper to just not have bought SA and bought 2016 on the new box.

@jaredbusch said in Trying to correctly understand core licensing in a vmware environment:

@dave247 said in Trying to correctly understand core licensing in a vmware environment:

@storageninja said in Trying to correctly understand core licensing in a vmware environment:

@dave247 said in Trying to correctly understand core licensing in a vmware environment:

@storageninja said in Trying to correctly understand core licensing in a vmware environment:

@scottalanmiller said in Trying to correctly understand core licensing in a vmware environment:

You get lots of silly extra like training and "support", but it's all minor. Sometimes certain features like maybe VDI are included with SA only, in those cases it can be a bigger deal.

There are vMotion implications on SA for the non-OS bits sometimes (SQL/Exchange, you need it to move a host, otherwise you will need to high water mark license all the hosts).

I have no idea what this means, especially the "high water mark license" part.

If you do not have SA, then when you move a virtual instance of Exchange, Skype for Business, SharePoint, or SQL from one physical host to another, the receiving host must already have a license to cover the virtual instance. The license does not travel with the virtual machine.

If you have SA the license can move between hosts (Note this is NOT true for Windows Server licensing, just SQL/Exchange etc).

What I mean for High water mark is The number of OSE's licensed for a server needs to be licensed for the maximum number of servers that run there in all vMotion scenario's (Unless you license a host for Datacenter Edition).

Wait, so if I have a VM running Exchange or SQL, I need to have SA in addition to having licensed the cores of each host server? Sorry, I've only just now learned about SA so I'm still wrapping my head around it.

If you have SA on your Server 2016 license, you are allowed to move it around between hosts. So you would not need to purchase licenses for all three hosts.

This isn't true. SA provides upgrade benefits for the server OS, but for sub 90 day license mobility (vMotion/DRS) you have to license all hosts in the cluster to the peak consumption they will use.

What SA on server gets you is SPLA/Azure hybrid use so you could move the VM to public cloud.

SA for SQL/Exchange gives you license mobility.

Note, this is why I recommend people go talk to Microsoft directly about licensing, and if you work for a MSP use a distributor (like Ingram micro) who will "guarantee" the solution will work and will eat the cost to make it whole. If you are doing any kind of real volume with Microsoft work with one of the LAR/LSPs who do this for a living.

Asking people on the internet who (mostly) have never resold Microsoft licensing or had to help dozens of people through audits is honestly much less productive than talking to a LSP/LAR who this is a core part of their practice.

This is different for Oracle because their sales teams misrepresent the licensing and commit fraud. For Oracle licensing talk to House of Brick, or VLSS.

@scottalanmiller said in Trying to correctly understand core licensing in a vmware environment:

@wrx7m said in Trying to correctly understand core licensing in a vmware environment:

So with SA, what changes in this scenario?

Not a lot, but you can move workloads every 90 days. Which is plenty for a lot of companies.

You get free upgrades to the new version. I learned this the hard way buying Windows 2008 100 days before 2008R2 came out. Couldn't upgrade without re-buying stuff.

@jaredbusch said in Trying to correctly understand core licensing in a vmware environment:

@hobbit666 said in Trying to correctly understand core licensing in a vmware environment:

My God reading things like this confuse me more lmao.

But am I right in summarizing this way:-
If I have a single VM that's "critical" and need to move/migrate it due to issues if all my hosts are licensed with DC with the correct core count and SA.
I can move that VM as often as I want? I.e every day

(Or applied to unlimited number of VMs due to DC)

What if I had all hosts licensed with DC but no SA would I then be limited to the 90day limit?

@StorageNinja can correct me if I am wrong, but if all hosts have DC, there is no limit because there is no limit to the licensing on each host in the first place.

It is only with Standard that you have the 90 day license portability thing. From one host with Standard to another host.

This is correct.

@scottalanmiller said in Has ConnectWise Set ScreenConnect Adrift as Ghost Ship Software?:

@mike-davis said in Has ConnectWise Set ScreenConnect Adrift as Ghost Ship Software?:

I forgot who said that once a product gets to about version 6 it's pretty mature in what it does and you shouldn't expect ground breaking new features in subsequent versions.

Product version mean nothing. If someone said that, they didn't know anything about software. Many products, Windows for example, start in higher numbers. Windows NT started on version 3 to make it sound more mature to specifically trick people who think version numbers would tell them something of that nature.

Veeam did this also. Started at 3.

@scottalanmiller said in KVM Backing and Support:

there is no agentless system that supports the range of apps that shops use

Veeam can hit Exchange/SQL/AD/Oracle/Windows/Linux (OS and file). For a lot of SMB's that's their estate.

@stacksofplates said in KVM Backing and Support:

shove everything into few platforms

but did a bad job of it I guess.

Anyone who uses Microsoft SQL for a log analytic platform did a bad job of it
Hilarious cost scaling issues, and backups with high change rate get fun.

For log analytic situations where data sovereignty isn't a concern, rather than a SMB learn Elasticsearch (which isn't bad to be fair) they could also just use a SaaS provider. SumoLogic, or Log Inteligence (we just launched), Splunk (if they have lots of gold pressed latinum).etc

@scottalanmiller said in I think I am missing something about Hyper-V....?:

@storageninja said in I think I am missing something about Hyper-V....?:

@phlipelder said in I think I am missing something about Hyper-V....?:

Why Hyper-V? We don't have to pay anything extra to build-out a virtualization platform for one.
We've done well with Hyper-V and Storage Spaces.

Storage Spaces Direct is Datacenter licensing only. If you have core dense platforms this gets expensive.

And it isn't production ready, and doesn't have production readiness on its roadmap. And that's right from the MVPs. It's a joke that MS released way too early with no way to get working. In the enterprise space, it's essentially non-existent and those that have used it have been burned big time.

The challenge (from a pricing and packaging perspective) is anytime you bundle something with another product, it's really hard to demonstrate value with it, or track demand.
Are people using it? What are they using it for? If you have a very non-direct sales relationship (Channel heavy) this gets even worse. You end up with a few things..

Roadmaps are controlled by the handful of people who:

1.Are influences (having been at an event recently with a bunch of MVP's i'll caution they are not the typical user)
2. Are the largest accounts you sell with directly (Welcome to "must support IBM Domino as a requirement because the pentagon demanded it).
3. If it's a field you don't have a deep experience with (Storage) you get who you have relationships with (Ex. Application developers). If you are entering storage this is dangerous because these guys make giant assumptions about hardware and networking.
4. Internal customers (Dogfooding is good, solving problems of scale or feature integration that only you need at the exclusion of external customers can be dangerous). Good desiners anticipate this (See Steve Jobs).
5. You hope the market isn't moving against fundamentals (as an example prior to the iPhone, Price, battery length, efficient use of spectrum, and relationship with the phone company where the most important things for a phone company).

This is all made worse by if you are a public company you can't legally discuss roadmaps specifics without an NDA and in public (lot of paperwork).
So what do you do?

You give away the product until it has market share and people find it stable/usable (Microsoft Communication services vs. Lync in pricing as an example).
You pray that the early adopters you get are indicative of the rest of the market
You hope your PM's are damn good at anticipating where the market is going and not where it was. A classic example was blackberry ignoring 4G and having the Gaul to lecture Verizon that their 4G network would fail and 2.5G was good enough rather than build a 4G phone. (https://www.amazon.com/Losing-Signal-Extraordinary-Spectacular-BlackBerry/dp/1250096065)
Motorola took a risk with Droid, and Blackberry was "turned off".

@matteo-nunziati said in KVM Backing and Support:

A nas IS cheap don't bore too much about space. Just backup.
is there a cheap solution with centraluzed management of backups? Cross platform?

For primary backup target I"m becoming less in love with Cheap NAS's at sacle.

1. Restore performance is terrible if it's anything but a few files or a small VM.
2. At large scale I've seen data integrity errors (especially ones that do non-MDRAID implementations).

@scottalanmiller said in Barracuda vs Meraki - firewalls:

@jt1001001 said in Barracuda vs Meraki - firewalls:

I won't do manage sd wan from the carrier ever again! Ask around here how much I love carriers!

OH yeah, such a bad idea. Just a VPN that you don't control and pay a fortune for.

SDWAN is a hell of a lot more than VPN tunnels...

1. Link bonding. Mix MPLS/Cable/T1/4G etc.
2. Per packet routing. Have the same session use multiple links depending on requirements...
3. Jitter management for the above. Can strategically use buffer bloat to make 2 similar segments match on one way latency (Inflate the lower link to match the higher one). This reduces the need for packet re-odrering (expensive from a compute basis).
4. Per segment monitoring. Latency isn't symmetric. Using things like 2 party clock synchronization and packet tagging you can measure with packet stamps in real time the one way latency of a link (Critical to keep jitter under control).
5. Automated rules engines with multiple factors that can even handle encrypted. Massive centrally collected rules engines based on destination IP, Ports, CNAME on SSL Cert for encrypted, packet headers if not encrypted etc.
6. Packet loss mitigation. For bulk media store and re-forward to avoid TCP retransmits keep throughput up. More sensitive real time protocols that are narrower can benefit from duel transmit (send packet down both links) as well as parity injected into packet streams.

Reason why you would pay a 3rd party....

1. Management of hardware for disparate links. Someone to deal with all the 4G cards, Cable Modems etc.

2. Management of billing. Having to sort and verify through billing for 5 different carriers

3. Awareness of options and scale. WAN resellers who do this for a living tend to aggregate a lot more demand and can get better pricing, as well as already have the fiber maps and quote tooling backend hooks for the tier 1 players so they can quickly identify the best options for each site.

Here's the dirty secret about not wanting to deal with someone else reselling someone else's links.... You always are. Thats how the internet (and wireless Networks) work. Everyone is leasing lines and transit from everyone.

@jaredbusch said in Barracuda vs Meraki - firewalls:

Bullshit. Those things all together are what makes it SD-WAN instead of some random guy trying to say all of his Ubiquiti ERLs with IPSEC tunnels are SD-WAN, because they are not.
@StorageNinja is exactly right on this.

SD-WAN in general simplifies management at large scale (and nore than just the devices, but also the links), optimizies performance in ways BGP, Shaping DSCP alone can't (and with a 1000x less work, no need for bespoke optimizations), enables unlimited choice on the WAN links without unlimited management hell as you try to deal with 10 different CLEC providers, and brings costs down for CPE gear (Virtualized, x86 rather than proprietary ASICs).

@emad-r said in Nested hypervisors?:

Yup go figure for value, Virtualbox has no intention of doing this amazing feature since 2011

Virutalbox is kinda useless as basic things like USB 3 drivers require a 5000$ license.

ESXi has supported nesting longer than all of them, and has a few production use cases for hypervisors on hypervisors.

@scottalanmiller said in Nested hypervisors?:

This would work, as only one vendor supports the virtualization of another hypervisor (and only for a niche use case) that I'm aware of.

Most vendors specify that X hypervisor must be there, and specify nothing of what is beneath it.

Beyond the performance hit on the IO path (lot of interrupts make this not efficient) this is the same as running on non-supported hardware.

If a vendor uses Oracle DB and says "We use Oracle" and you run Oracle on a x86 emulator on a Raspberry Pi, I can see you trying to claim a loophole, and I can also see the vendor not wanting to support you.

@dustinb3403 said in Nested hypervisors?:

An app vendor saying "you have to use X guest on X hypervisor" is a weird requirement. Unless there was a very specific feature/function that the guest somehow can can flex by using a specific hypervisor, and is the reason you as the customer engaged the app vendor.

It comes from a few cases....

1. VDI often hooks the hypervisor.
2. It comes from the fact that customers often ask the app vendor for performance and configuration advice, and to be fair it's kinda nice when everyone doesn't live in a silo at their ring.
3. It boils down to who gets blamed for an outage. An Application vendor is often expected to produce or support a RCA, and if the storage platform is Ceph running on BSD and they can't provide it the customer management may throw the baby out with the bathwater (It's not right, but it happens enough).
4. Nested Hypervisors have a lot of CPU overhead issues (Good luck finding a NUMA boundary through it), and storage latency is rarely consistent without taking extreme measure. Both of these can represent themselves as application layer problems. The complaints start at the app layer, as do RCA's on issues. They carry a lot of costs if people are constantly calling them and they are constantly having to investigate and say "it's your wacky Jenga pile of hypervisors".

If we lived in a mythical land where people didn't hold application vendors accountable to performance and availability this might work. Sadly that's not how things work. I can't tell you how many times people yelled and blamed and replaced "Shitrix" when the problem was a bad storage config.

@nerdydad said in Looking for firmware:

Definitely not vSAN. All in one datastore and not expanded.

You have an external RAID Array I hope? This HBA does raw pass through well...

@scottalanmiller said in What are necessary/worthy/affordable tools for SMB?:

@storageninja said in What are necessary/worthy/affordable tools for SMB?:

@scottalanmiller said in What are necessary/worthy/affordable tools for SMB?:

And one key difference, SMBs tend to use homogeneous desktop environments. They "choose" Windows. Enterprises almost never do. They almost always deploy desktops as needed. Windows, MacOS, Linux, etc. They don't just force one to everyone. Somewhere, some must, but it's rare.

Work for a 30K man company. I can have our VAR spec out whatever I really feel like and my boss will approve for a computer. Currently using the "standard choice" MacBookPro. When you use MDM API's and agents for management you don't really need to focus so much on a corporate "Image" (which Microsoft has been going away from as the SMB's tend to use it more). Our trusted stuff is behind SSO portals, or VDI generally.

I used to be able to do that at the non-profit with Windows and MacOS, but they wouldn't let me have the Chromebook that I wanted - even though it would have been cheaper and more useful than the Macbook they made me get

Hijack Anyone sees value in an i9 beyond cooking things on my laptop? I do some video render jobs but not enough to care (or I wouldn't just throw at my desktop if they became a chore).
Also is 32GB of RAM seriously useful if you don't run VM's locally?

I've noticed the biggest slowdowns on a Mac are

1. Running the SSD low on space
2. Putting the McAffee Virus with automated scanning of all files on access and no exceptions.

@pete-s said in GeekBench Results for Cloud Servers:

As you said it might be the best cost performance ratio for cloud deployment.

ehhhhhhh. It's the best cost for selling someone a modernish core for hosting that lacks any real scaling issues, or licensing per core/socket concerns.

@emad-r said in GeekBench Results for Cloud Servers:

Thought about Network testing and came up with this:

That's just a testing of peering.

@dustinb3403 said in SAS SSD vs SAS HDD in a RAID 10?:

OBR5 is the standard if you are going to be using an SSD

A URE isn't the only failure or corruption mode on a SSD. You can have drives that are not dead, but you want to shoot (firmware acting squirrel and you get 500ms). Also, 16TB SSDs that have deduplication and other data services in front of them can take a LONG TIME to rebuild (making that 7+1 a non-fun rebuild).

Throw in people using cheap TLC and QLC (crap write speed and latency after the DRAM and SLC buffer exhausted) and I wouldn't say as a rule RAID 5 for traditional RAID groups of SSDs is always a good idea. If you have an SDS layer that wide stripes across multiple servers, and limits the URE domain to an individual object this is a bit more controlled. If I have a small log file that writes in a circle a lot (My Casandra/REDIS systems) erasure codes may not be worth it has given the volume of ingestion.

I'm a bigger fan of RAID 5 on SSD in systems where I can pick and chose my RAID level on a single object, LUN etc so I can break up the write outliers that are small.