Topics created by dave247

no, I am saying that I could see that someone wanted to separate out their devices so each could have its own separate DHCP scope. I am not saying that this was a good idea, or that I would do it, just that I can see how VLAN's could be used to achieve that effect. Again, I am not saying this would be using VLANs correctly

Something to keep in mind is NGFW. Ubiquiti and Meraki, for example, are NGFW.

It looks like much of the market is already starting to cool on the UTM crazy and NGFW is taking off as the "next stage" of popular approaches. Basically a reversal of direction or marketing at least, even from the big players in the UTM space like Palo Alto, Fortinet, Cisco, etc.

@scottalanmiller said in What's with the massive price difference between Dell VAR prices vs xByte prices??!?:

@dave247 said in What's with the massive price difference between Dell VAR prices vs xByte prices??!?:

@scottalanmiller said in What's with the massive price difference between Dell VAR prices vs xByte prices??!?:

@dave247 said in What's with the massive price difference between Dell VAR prices vs xByte prices??!?:

@jaredbusch said in What's with the massive price difference between Dell VAR prices vs xByte prices??!?:

@dave247 said in What's with the massive price difference between Dell VAR prices vs xByte prices??!?:

I think I will go through xbyte even if my VAR tells me they can match their prices...

They can't. That is not how it works.

Well actually one time my VAR matched the exact refurb prices that I was going to get from xbyte like last minute. I got like 5x Dell N series switches brand new for refurb price - which was like $2,500 each instead of the $7k or w/e they were when I was buying them.

They can take a loss if they want, most likely they were just having xByte deliver them directly and not marking them up.

But here is a simple rule... if you need your VAR to match a better price somewhere else, your VAR is making you do extra work and burn other relationships up front to lock you in. You waste time and effort (that's money) and other vendors won't keep quoting you good prices if you don't then buy from them. Tricking you into raising the price from other vendors is how price matching works out for your VAR.

I don't understand this part: "Tricking you into raising the price from other vendors is how price matching works out for your VAR."

The vendor is convincing you to go get good prices from other people. Those people are providing good prices in the hopes of earning your business by being better. Your vendor then convinces you that even though the new vendor treated you better, to stick with them. Your old vendor (the price matcher) now knows that you aren't loyal and will shop around, so they know it's time to "drive to the bottom" and lower prices by lowering service. The new vendor (the one giving you a good price) gets taught that doing a good job for you gets them nowhere and getting you a good quote is a waste of resources, so they will stop doing so and just start quoting rack rates, because you are just wasting their time.

"Shopping around" for prices is a dangerous game unless you are just looking up public web prices. But a VAR, by definition, is not about price but about services. If you care about price, then using a VAR is a mistake every time. You want a pure reseller who isn't marking up to add services to the product. Yes, in IT we need to ensure our prices are good. But you can't get good prices by going around to everyone and getting quotes, it will feel like you are getting good prices, but the products are not directly comparable and the best prices exist only for loyal customers with good relationships.

At this point though, I'm thinking my VAR is really just a re-seller. He doesn't add too much value and when he does give his input, I think it's usually based on a bit of antiquated IT knowledge.. We've bought workstations, servers, firewalls and MS licenses through him but he also does consulting and services sort of like a MSP. My boss has in the past hired him to do consulting and things so we've paid him for "services" and things.... So going back to all the stuff I've seen you post on SW, we were buying stuff from the guy who was also doing the consulting... you know the rest of the story. However, I more or less halted this when I came on. I started asking questions, doing my own work and shopping around for better deals on hardware. I still go to him now and again mainly for MS licensing and the last thing was new Dell switches which I got refurb price.

@scottalanmiller said in Question about weather or not I should enable write-through mode on a RAID1 SSD array:

@dave247 said in Question about weather or not I should enable write-through mode on a RAID1 SSD array:

@travisdh1 said in Question about weather or not I should enable write-through mode on a RAID1 SSD array:

@pete-s said in Question about weather or not I should enable write-through mode on a RAID1 SSD array:

@dave247 said in Question about weather or not I should enable write-through mode on a RAID1 SSD array:

@pete-s said in Question about weather or not I should enable write-through mode on a RAID1 SSD array:

@dave247 All enterprise SSDs have power protection caps so that is an expected feature.

If the raid controller cache makes an improvement or not depends on the workload and the size of the cache.

I think the 3.84TB drives have 4GB cache internally. That's large compared to the RAID controllers cache. What do Dell recommend?

I haven't asked Dell what they recommend. The H730P has 2GB and the H740P has 8GB though.

Give them a call. A little more than 3 years ago I spec'ed a computer with 4 SATA SSDs in two RAID1 arrays and I remember that they were adamant that I get the H730P controller. Machine was an R630.

Of course they were, they're job is to sell you thing, or more expensive things, than you really need.

That's kind of what I was thinking.. though I have had many times where vendors like Dell just don't do that. Sure they want to sell you stuff but it's not good to constantly take advantage of people either as it will catch up to you one way or another. Case and point, I over-spec'd a system a while back and my Dell VAR suggested I go with part X because it was just as good as part Y only a lot cheaper.

A VAR might just want to do a good job. But it's more complex than people say. It's never about "more expensive", but it is generally about better margins. Which are impossible for end users to determine.

The cheaper hard drives might have had equal or better margins than the expensive ones. We say "expensive", but that's dumbing down something that doesn't need dumbing down, and implies that we can easily tell when we are being led for profits based on the resulting price, which is incorrect.

There is every reason to believe your VAR was just doing a good job for you. But the result of a lower price doesn't tell us in any way if they or Dell themselves did something purely in your interest, or also in theirs. And there is always the possibility that the more expensive drives offered something that would have been beneficial to you.

Yeah I see what your saying. Actually the parts in my example were the new Intel processors. I selected a Xeon Gold and he suggested I go with Silver as they also perform well and just because they are "silver" doesn't mean they are a less better version of the Gold. And in actuality, the reason I chose Gold was because I was trying to get more Ghz but I understand that's not all there is to a CPU. It was my first pass at a build.

For me it would depend on the redundancy level of the RAID along with the age of the drives. I'd also make sure they're all the same speed (and ideally size). If you're using older drives then I would at least dedicate one hot spare and have at least another spare offline, ready as a replacement.

If you're not under time pressure then why not just order it and see if the hard drives work with it? If there is a firmware issue then you'll find out right there.

I've seen enterprise level hard drives last in excess of 10 years (although I can't vouch for their performance at that point, it may be affected), so I think it's worth a shot.

@scottalanmiller said in Need some help with SQL Server 2016 Standard licensing (price confusion):

@jaredbusch said in Need some help with SQL Server 2016 Standard licensing (price confusion):

@dave247 said in Need some help with SQL Server 2016 Standard licensing (price confusion):

@jaredbusch said in Need some help with SQL Server 2016 Standard licensing (price confusion):

@dave247 said in Need some help with SQL Server 2016 Standard licensing (price confusion):

@jaredbusch said in Need some help with SQL Server 2016 Standard licensing (price confusion):

@dave247 said in Need some help with SQL Server 2016 Standard licensing (price confusion):

@phlipelder said in Need some help with SQL Server 2016 Standard licensing (price confusion):

@dave247 said in Need some help with SQL Server 2016 Standard licensing (price confusion):

Hi friends.

I am working on building a new physical server to replace one which is running older versions of Windows and SQL server, plus it is almost out of storage space so this needs to be done sooner than later.

This SQL server is running a 3rd party application and they currently only support up to SQL 2016, so that's what I have to install - not 2017. And it's going to be SQL 2016 Standard Edition running on Windows 2016 Server Standard with 16 cores.

I spent a while researching SQL sever licensing to try and get an idea of how much it's going to cost. I haven't dealt with SQL server licensing yet.

First, I assumed that I would still have to purchase SQL Server 2017 core licenses with downgrade rights. So looking on the SQL Sever Pricing page, it looks as though Standard - per core price is $3,717 (2 pack). So if my server has a total of 16 cores, this is going to cost about $29,736 to cover SQL licensing.

Then I checked over on CDW just to get an idea of prices and things and I had the idea to search "SQL 2016" when I found this: SQL Server 2016 Standard - license - 16 cores - with Server 2016 Standard for like $1,900.

Is this even applicable to what I'm doing or am I missing something? It does say in the technical details "BIOS locked (Lenovo)" but I have no idea what that refers to. But other than that, it looks like it's licensing SQL Server 2016 for 16 cores and bundled with Windows Server 2016. Surly this can't be correct... or is it? If it is actually what I would need to be covered, I would purchase it, of course.

Otherwise, can someone help me get an idea of what I should be paying for SQL Server 2016 Standard Edition for 16 cores if not the cost I initially calculated ($29,736)? And I don't think we'd do the server + cal licensing as we have about 80 users and 100 or more systems which would connect to the SQL server.

Simple rule of thumb to ask your Microsoft licensing rep for the following:
First option is license + CALs that allows internal access only with unlimited instances on the server and unlimited cores:

SQL Server Standard License SQL Server Standard User CALs (80 Users)

Second option is per core with a minimum of 4 to purchase:

SQL Server Standard Per Core 2-Pack (2x)

In the Per Core scenario we can license for the number of physical cores to use and delimit that in SQL Studio Management. When it comes to audit, a snip of that setting that only allows the four threads should be just fine.

So if you license + CAL, do you have to cover all users AND computers?

If you license by user you cover users. If you license by device you cover devices.

Well what constitutes as a device? I mean, users use a device to connect to the SQL server... so wouldn't I have to cover both? I don't get it.

That is never how Microsoft CALs have worked.

ok, I finally re-read the overview.. makes sense again. We have a pretty even user/device ratio with slight fluctuations in both over time. I suppose we'd just do user CALs..

There is almost no reason for anyone in the normal, day to day, business world to use device CALs.

Agreed, this is super specific niche stuff normally reserved for manufacturing shift work.

We have a few clients that run two or three shifts across one or more facilities. A shared device by two or three peeps per day is about the only time we've ever deployed Device CALs.

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

@dave247 said in SAS SSD vs SAS HDD in a RAID 10?:

delete

You can delete your own post. it is only visible to you and a mod, until mods purge it.

0_1537364565805_a84f8999-46fb-4f1f-bace-f8cdef7bd9ea-image.png

ugh, I swear that wasn't there the last time I looked... I can't delete it now though since it's been over 2 hours.

@storageninja said in Cisco SmartNet:

@dave247 it’s basically a people management platform. Can have a system do a sales manager can tap calls and recordings and do all kinds of metrics they integrate to the CRM. I’d argue avaya is more powerful, but call manager isn’t something you Casual replace with an open source PBX.

Now call manager express (its little cousin) is a basic PBX with unity for voicemail. If you went call manager instead of express I assume someone had some fancy needs.

We actually had an old Avaya PBX for the last 7 years or longer. It was quite simple and did the job.. CCM is a fucking nightmare with how many menus, sub-menus, sub-sub-sub-sub-sub-sub menus and such.. not to mention all the servers and sub-server-applications involved. You'd think it was designed for companies with thousands of employess, not under 100. OH WAIT.

@scottalanmiller said in Proper NTP server usage?:

@dbeato said in Proper NTP server usage?:

@obsolesce said in Proper NTP server usage?:

@dbeato said in Proper NTP server usage?:

@obsolesce said in Proper NTP server usage?:

I just stood up a 2016 DC. I did nothing at all to it, and by default it uses the PDCE as the w32tm /query /source.

I haven't had a need to stand up a 2016 PDCE, just regular DCs.

I'm going to stand one up in a lab to see what the source is by default.

I could have sworn it was time.windows.com and not CMOS. That was 2012 R2 though, I'm curious now.

It has always been CMOS first, that's why all the systems that lose their time over time are due to that. Also any VM prior to booting to the OS regardless or not they have Guest Services enabled, get the time from the Host BIOS.

That makes sense. The PDCE I set to use ntp.org very well may have said CMOS before I changed it. But regardless, when you join a pc or server to the domain, it automatically is set to use the PDCE as the time source.

Yes, in a domain all computers get the time from a DC.

They SHOULD anyway.

Yeah, that's important to note, should is the keyword.

@phlipelder said in How do you have your disks & folders set up on a virtual Exchange 2016 install?:

In a virtualized setting there's no real reason to configure more than one location for the Exchange databases and logs. Unless, there's a separate high IOPS location that the logs would be stored on?

@jt1001001 Why thankfully going away?

Virtualized would be the same as physical. Basically, there is rarely a reason to separate them no matter what. In special circumstances, yes, but generally, no.

@scottalanmiller said in Anyone running Exchange 2016 on Server 2016?:

This is MS' bread and butter. While I've not tried it, if it is officially supported I'd have little concern.

0_1527484744157_430b78c9-7eee-4e5f-bc84-fcbdfdf77a3e-image.png

@emad-r said in Trying to correctly understand core licensing in a vmware environment:

@dave247

maybe this was answered here, but I was wondering if someone can correct me on this

If i put WIndows server VM with 2 cores on Hypervisor any brand which has 8 cores.

Do i need to license windows server VM for 2 cores or 8 cores ?

There are no cores at the VM level, those are vCPUs. Windows is licensed by cores, not by vCPUs. Cores are always physical, and the minimum is 16. No matter how you configure the VMs and vCPUs, you have to license 16 cores at a minimum.

A single vCPU could use sixteen cores or more to power it, the licensing is by the available power, not the configuration.

@dbeato said in Anyone running SonicOS 6.5.0.2-8n?:

@dave247 said in Anyone running SonicOS 6.5.0.2-8n?:

We run a SonicWall NSA 3600 where I work and I am staring to look into upgrading to the new 6.5 firmware (6.5.0.2-8n). I have heard of some issues with the last two updates, and wanted to get more input if anyone has any to give..

I am not, I am using the 6.2.9 on the production Sonicwalls we have. I would recommend to test it out if possible.

Unfortunately I can't realistically test it out. Even if I had an extra, non-production unit, I doubt I could effectively detect issues since production factors would not be present enough to fully test. I suppose I could just always roll back if necessary... but I think I will probably be waiting a few more releases.. that or just move to a different UTM all together..

@dave247 said in Disaster Recovery as a service companies that support IBM iSeries / AS400 systems?:

@scottalanmiller said in Disaster Recovery as a service companies that support IBM iSeries / AS400 systems?:

@dave247 said in Disaster Recovery as a service companies that support IBM iSeries / AS400 systems?:

@scottalanmiller said in Disaster Recovery as a service companies that support IBM iSeries / AS400 systems?:

Some important things to keep in perspective...

The average CEO is incompetent. The average business loses money and will go out of business. The average business that survives doesn't do well. The average IT person is incompetent. The average of any position, in any field, is incompetent.

Averages are quirky things. The average is crap. So that "most" companies and IT people are like this, shouldn't really make us upset or worried. No one worried about this stuff is average or looking to be average.

LMAO. So I am actually correct in constantly thinking that I am incompetent at my job and should technically be fired.

LOL. Assuming you are AVERAGE then you are incompetent and should NOT be fired because incompetence is all that is normally expected 😉

But like I have said elsewhere, if you are taking time to post on a professional forum, that already puts you above average. And as I said here, if you are worried about being average, that puts you above average.

None of us are average. That can make this harder, because we start to feel like the average is pretty good. But it's the average of a highly self-selected group. It's like trying to determine who well the average person would do in hand to hand combat by evaluating the results of Roman arena gladiator fights. The average gladiator is nothing like the average person, they've already self filtered heavily.

Well thanks. I do trey very very hard to make the wises choices with everything for work. I mean, I worked a decade in retail and hated it, and now I have an IT job where I'm in charge of an entire environment... it's a very massive learning opportunity and I've learned and done so much.. but yeah, there's a lot of times where I feel like I'm not qualified, at least until I learn and do the thing and succeed at it.

That's a different effect. IT is a HUGE field. No individual should ever feel like they have enough knowledge to do the job at any significant level. It's just not realistic. That's why, no matter how small a company is, they need a team of experts. IT is just too big. Most only need to work a few hours a year, but they are all needed. The best CIOs still hire consultants or in larger companies internal IT experts, for nearly every functional area.

I think you will begin to hear more about Workspace ONE powered by Airwatch rather than Airwatch as a separate product (https://www.vmware.com/products/workspace-one.html). There's a large amount of work being done around managing Windows 10 devices too (moving to Unified Endpoint Management instead of just traditional MDM, things like ability to manage BitLocker keys).

There were some platform announcements yesterday as well - https://blogs.vmware.com/euc/2018/03/workspace-one-intelligence-march-announcement.html.

But I'd stick with @StorageNinja's advice on being patient for a response.

A big rule is... no fluff. Keep it clean, simple, and on task.

@dave247 you might have a look at this thread. I think it could be useful for you.

https://social.technet.microsoft.com/Forums/windowsserver/pt-BR/745577a2-d035-4a4e-80ce-ea2a68370e19/replacing-domain-controllers-reuse-ip-and-server-dns-names?forum=winserverDS

@dave247 said in Quickest way to set up Zabbix to monitor managed switch metrics??:

@coliver said in Quickest way to set up Zabbix to monitor managed switch metrics??:

@dave247 said in Quickest way to set up Zabbix to monitor managed switch metrics??:

Well after adding the rest of my switches, Zabbix decided to take a crap... the configuration cache was topping out. I increased the Cachesize from 8 to 16 but things seem to have stopped working anyway... after another day of futile struggling, I decided to just shut down the system and give LibreNMS a try.

That's bizarre... I've got almost 100 servers half a dozen hosts and dozens of switches reporting to our zabbix server.

I know.. At first I figured it was because I was using the 4 different templates and after I added all 8 switches, it was being over-loaded. But then I realized that it's probably normal to have hundreds of systems sending data to Zabbix, so that shouldn't be the issue.. well it's par for the course for me.

I just got LibreNMS up and running and it's working like a charm. Got all my switches added and I'm watching beautiful graphs going before my eyes. It just seems to work. Zabbix was a mess...

Glad you got something to work! 😄

@thwr said in Question about switch configuration in relation to server NIC teaming:

Switch independent works pretty good. Also take a look into SET (Switch Embedded Teaming), which moves teaming one layer up. It's basically the next logical step.

Meh, my bad. SET requires Windows Server 2016

@dbeato said in Can't figure out the cause of LAN broadcast traffic spilling over into WLAN zone on SonicWall:

@dave247 said in Can't figure out the cause of LAN broadcast traffic spilling over into WLAN zone on SonicWall:

Problem solved. Turns out I had to manually remove VLAN 1 access from the configured ports. Apparently this must be done for all general mode ports on Dell N series switches.

Yes, because that is the default PVID of the ports.

Yes but I had changed the PDIV of the port from the default value to 300. So, I thought that VLAN 1 access would stop and only VLAN 300 would be allowed.