Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Post
  • Reply
Erwin
Feb 17, 2006

Are there other VMs on the array? If not, can you just yank a cable and wait for a timeout? :haw:

Adbot
ADBOT LOVES YOU

Erwin
Feb 17, 2006

With a two-host cluster that is your only cluster, you should probably just have Admission Control disabled. Its purpose is to prevent you from powering up too many VMs on a cluster, but the idea is that you would just power it up on another cluster.

Erwin
Feb 17, 2006

Question about VMware View pricing. Is it literally only $190/desktop for existing vSphere customers as claimed on this page?: http://www.vmware.com/products/view/howtobuy.html

I figured I'd start looking at View towards the end of the year when I have time because I figured it had an initial buy-in of several thousand dollars at least. Is it really just linearly priced on a per-desktop basis?

Erwin
Feb 17, 2006

Well, we only have about 25 users. Usually at that size things are inefficiently priced, but this makes View very tempting. 500-750 IOPs is not a problem, and a bump in host capacity will cost less than replacing half of those desktops.

I'm not going to switch everyone tomorrow, but it's something to look at before we refresh any hardware.

Erwin
Feb 17, 2006

Alctel posted:


What do y'all do? Our enviroment has 2 ESXi servers with around 20 VMs on

I have it virtualized and replicated to DR. Similar environment to yours, 2 hosts plus 1 in DR, about 35 VMs. I don't think there's a good reason these days to make it physical, but any more than a handful of hosts and you have to be a little mindful of where it is (use host affinity to keep it on a small subset of hosts so you can find it at 3am under stress). Us 2-host guys can let it roam free :clint:

I've heard nothing but bad things about the appliance.

Erwin
Feb 17, 2006

feld posted:

Physical. Nothing more painful than having Virtualcenter crash and having to search all your ESX servers to find it and turn it back on. (I've not witnessed anyone using the high availability thing yet, and considering the amount of RAM they recommend for virtualcenter -- like 8GB -- why would you want that on your VM cluster?)

He said he had 2 hosts, so that negates both your reasons. You have a 50/50 chance of finding it on the first try, and it doesn't need (or, more correctly, use) 8GB of RAM. My vCenter's active memory is rarely over 1GB unless vCenter is actually doing something, and then it might hit 2GB.

Once your host number climbs, just set DRS rules right and you'll still be able to find it.

Erwin
Feb 17, 2006

Moey posted:

How big is your environment? Is your SQL db running locally?

3 hosts over 2 datacenters, 35ish VMs (so tiny). Yes, SQL is local.

To be clear, active memory, as opposed to shared memory is only about 1GB. Total host memory used is 6.5GB at the moment, but that's still not really a reason not to virtualize vCenter.

I guess if you have spare hardware you can give it more thought, but buying hardware vs. just making it a VM is a no-brainer I think.

Erwin
Feb 17, 2006

Rhymenoserous posted:

Though I want to know how to get my hierarchical view back for my VM cluster so I can see what is where at a glance.

Anyone know how?

For a visual of what I'm talking about. I dropped everything into a cluster and this was the result:

Is there any way to go back to



host
-Virtual machine
host 2
-Virtual machine

This is how it looks with DRS turned on and fully automated. I guess it encourages you to think of VMs as just running on a cluster and not worrying about which host they're on.

Erwin
Feb 17, 2006

I feel stupid asking this, but cloning/imaging isn't something I've done since XP, and I hated it then.

I want to make a 2008 R2 template in vSphere. I think I did it wrong since the VM comes up not joined to the domain, even though I set it to do so with the customization wizard. My steps were this:

- Create VM, install windows, VMware Tools, updates
- Change some settings for the local admin account
- Create unattend.xml in c:\windows\system32\sysprep that copies the admin account to the default account (this is not that important to me, but the 'how-to' write-ups I can find all say to do this, so I figured why not)
- Sysprep that bitch, choosing OOBE, and shut 'er down
- Convert to template
- Create VM from template

When the VM comes up, it asks me for language settings, an admin password, and is not connected to the network. So, I'm guessing I did something wrong. I thought maybe I shouldn't sysprep, but I remember trying before without sysprepping and not being able to join the VM to the domain, even though I had "create new SID" checked in the customization wizard. But, I remember in the vSphere class we cloned 2003 VMs without manually sysprepping (VMware took care of that after putting sysprep on the vCenter server).

Anyway, can anyone lay out the steps needed to create a 2008 R2 VM template, and use it to create VMs that join to the domain as specified in the customization wizard?

Erwin
Feb 17, 2006

Mierdaan posted:

I feel stupid suggesting this, but you are changing this here radio button right?



I'm selecting "Customize using the Customization Wizard."

Erwin
Feb 17, 2006

Mierdaan posted:

Okay. I have everything set in a customization specification and have definitely forgotten to change that when deploying and the guest comes up all uncustomized :saddowns:
I've forgotten that in the past too, but sadly this seems to be another issue.

Maneki Neko posted:

Silly question, did you use WSIM to create your xml?

That stuff seems like it changed quite a bit between 2008 and 2008R2. We don't use the customization stuff built into VMWare at all, but just have a generic sysprepped 2008R2 image sitting as a template, used WSIM to generate everything it worked like a champ.

I copied and pasted an existing unattend.xml from some "how to make a 2008 R2 VMware template" blog and substituted my own company name, etc.

So, it looks like the network is not connected because of this maybe: http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=1020078 That might actually explain the entire problem, and certainly explains why VMware can't get it onto the domain. I guess I'll install the hotfix linked in that article onto the template, but I've already sysprepped it and babby here doesn't understand the implications :smith:

Erwin
Feb 17, 2006

Erwin posted:

2008 R2 template problems...

For future reference, the solution was this VMware KB plus NOT sysprepping (VMware does this). I lost out on copying my local admin account to the default account, but I'm not a sperglord who turns off IE security and changes the start menu to classic and other dumb things, so no big deal.

Erwin
Feb 17, 2006

Why do people ever touch the browser on a server?

Erwin
Feb 17, 2006

adorai posted:

depends. A lot of times we have vendors in that need to "do research" and I am not a fan of giving up my PC for 3 hours. Share the console session in VMware, and you can continue working. In my experience if you RDP to the server and then share a webex session inside of RDP, they lose mouse control if you minimize the window.

I have a laptop that I usually use for this, but why not make a Windows 7 VM, then use the browser in that for Webex, then RDP to the server that the vendor needs?

Erwin
Feb 17, 2006

We just deployed Veeam about a month ago and it's not without its headaches, but support has been decent in general. Perhaps they've changed in the last month?

Misogynist posted:

Did any techs get fired over repeated passive-aggressive comments made in your last support case?

Did the tech really get fired? It sounds like he should have.

Erwin
Feb 17, 2006

This is a dumb question but I don't think the VMware Update Manager documentation explains how to upgrade the VM hardware from 7 to 8 on your Update Manager VM. I feel like if I try to remediate it, it'll just explode halfway through :ohdear:

Also, what about the vCenter VM? I feel like vCenter has to be running for Update Manager to do its thing.

Erwin
Feb 17, 2006

Ugh, of course. Like I said, dumb question :downs:

Erwin
Feb 17, 2006

I have no opinion other than that we're at 8GB/core and not worrying about ballooning or other memory management at all is nice. Our current hosts can go up to 16GB/core, but I, too, would rather scale out (and plan to add hosts soon).

Erwin
Feb 17, 2006

So I'm starting to juggle data around so I can upgrade my datastores to VMFS5. Is there any reason I wouldn't want to combine, for instance, two 2TB LUNs/VMFSs/datastores that are on one RAID set into one 4TB LUN/VMFS/datastore?

edit: I'm storage vMotioning VMs off of datastores, deleting the datastores, and creating new ones, not necessarily because it's required, but I have the room so why not.

Erwin
Feb 17, 2006

Serfer posted:


Go vmware!

I think some of their UI messages are written by high school interns:



"Oh some file or directory stuff is happening. You know, like some things, man."

Erwin
Feb 17, 2006

Misogynist posted:

Pricing is typically considered to be vendor-confidential privileged information, and it's rather damaging for vendor relations to share it.

Only because the way storage hardware (and most IT resources) is sold is bullshit.

Erwin
Feb 17, 2006

Misogynist posted:

That's why you give them a budget up front. I used to do this with recruiters all the time and it saved me a lot of heartache.

And when you give them a budget of $100k, that hardware they just sold to another guy for $75k will be shown to you for $95k because they "really sharpened their pencil and got you a great deal."

Erwin
Feb 17, 2006

Martytoof posted:

I was hoping to shrink the VMDK associated with this particular VM. For some reason I thought that was a thing I did through the Tools icon :(

There's a menu and settings (including shrink) available through the icon for VMware Tools on vSphere/ESXi VMs. Maybe you're remembering that? (I didn't know Fusion didn't have those options).

Erwin
Feb 17, 2006

Frozen-Solid posted:


What's the best course of action here?
Fire the guy to start with. Everything he said is wrong, and if he misrepresented himself as having VMware knowledge, you should not pay him.

What you said is right, multiple hosts can (and are supposed to) access one data store, and the VM is only registered on the host on which it is running. vCenter just takes care of de-registering/registering automatically during migrations.

You keep using the term vSphere, but it sounds like you don't have vSphere licensing, just ESXi hosts (not your fault, their naming conventions are confusing). You really should splurge for vSphere licensing though, you'll be very happy. Sell your body if you have to.

Erwin
Feb 17, 2006

Frozen-Solid posted:

We just don't have an actual vCenter server, because we only have 2 hosts.

You've confused everyone :) Why does only 2 hosts mean you can't have a vCenter server? You know it can be a VM, right?

Erwin
Feb 17, 2006

It is sold separately (you don't want a vCenter license every time you buy new host licenses). The real question is, what kind of reseller allows someone to buy vSphere licensing without a vCenter server? A bad one, that's what kind.

Also, yeah, Essentials Plus. Again, bad reseller.

Erwin
Feb 17, 2006

Bitch Stewie posted:

The words "Mission Critical" and cheap NAS or SAN don't sit too well with me.

What I would look at is a pair of solid boxes with RAID10 DAS and depending on budget either run a VSA

The words "mission critical" and VSA don't sit too well with me, only because it's 'new' like SRM replication, which doesn't work very well, and it's an "add-on" like VDR, which doesn't work at all.

I mean, it works for a few weeks, then your backups become corrupt and support spends a week looking at it before telling you you just have to delete all the backups and start over, and that "you should use some other software along with VDR because VDR isn't really meant for enterprise production." This was on something small, like 20 VMs at the time, with VDR 2.0. Piece. Of. poo poo. It's basically VMware's proof of concept for their backup APIs. Definitely buy 3rd-party backup software.

Erwin
Feb 17, 2006

Maybe look into the used market? You probably won't get a warranty or support, but it's better than not having a SAN, and you can get something like an EMC AX4 for under $5k (or, you know, a better brand than EMC). I'd sell you an EMC AX100 for little more than the cost of shipping, but it hasn't been on the HCL since 3.5.

Also, FWIW, I have a Promise VessRAID for our test environment. It was cheap (sub $2k for an empty 16-drive-slot box) and I can buy hard drives from Newegg or Amazon, as long as they're on Promise's compatibility list (the standard WD and Seagate drives are). I wouldn't put it in production, but maybe I would if I had no budget.

Erwin
Feb 17, 2006

Set up a VM with two network cards. One on the isolated VM network, one bridged. Install PFSense or something, and make it the default gateway on the VM network.

Erwin
Feb 17, 2006

Syano posted:

The hate against Veeam has come up in this thread recently and since we are on it again I have some new hate to throw its way as well. We recently had a problem where our nightly jobs would fail with some obscure error about CTP files. I sent in a support ticket and after two days of no response I posted my error to their community forums soliciting help only to have my post denied and being told I needed to read their 'rules' or whatever which apparently state you aren't support to post support items on their forums. Anyways, long story short: It took 2 WEEKS to get the problem solved. Thats 2 WEEKS of not knowing if my backups would be usable or not during failure.

You mean CBT files (as in, it couldn't use changed block tracking)? If that's the case, the backups were usable, they just took longer. I had the same issue, and I just had to install the latest patch to fix it. It did take them quite a while to get around to telling me about the patch, though.

Erwin
Feb 17, 2006

We'd like to move our DR location from our property to a 3rd-party location. Are there companies that do hosted ESXi? Or is it better (cheaper or easier) to just colocate our own hardware? We're only looking at 1 ESXi 5 host, plus a storage array if colocating.

Erwin
Feb 17, 2006

Bob Morales posted:

Shouldn't be any different than co-locating any other server.

Isn't there something in the EULA that doesn't allow people to re-sell VM's on ESXi?

We have a (Dell R710) server @ Rackspace that runs VMware ESX 4.0.0 and we can add/remove VM's as we please (actually, we have to call them to do it and they charge us more a month for each one) but that's the 'Private Cloud' product and each VM is managed by them.

That sounds awful. I just want an ESXi host in someone else's building that is connected to my vCenter server, and that I do all the configuring on (I don't want it managed). I'm just trying to figure out if there are companies that rent out a physical box with ESXi installed, or if it's better to colocate my own hardware.

I'd rather pull my existing hardware back to my production location as an extra host and then rent a host somewhere.

edit: sorry, didn't mean to sound like a dick. I mean it sounds awful for my purposes to pay per VM. I want to have an ESXi host attached to my vCenter server to which I can replicate VMs with Veeam for failover, and run a few full-time offsite VMs. A Google search says people do host ESXi, so I guess I'm wondering if anyone has experience in hosting or colocating an ESXi host under vSphere licensing.

Erwin fucked around with this message at 20:38 on May 8, 2012

Erwin
Feb 17, 2006

Moey posted:

It has folders in there with the names of the VMs, then the orphaned snapshots inside. But when I look at the actual VM within vSphere, it doesn't show that it is using that datastore at all (and there are no snapshots for any of these VMs).

There is also seemingly a full copy of a VM in one of the datastores, that has a 0kb VMX file, but the VM is perfectly fine running on a different datastore.

What my old boss did to some of this stuff, I have no idea...

Edit:

gently caress it, time to blow this stuff out.

If you edit the settings of each of those VMs, are the .vmdk files just the name of the VM, or is there a snapshot number attached? EG myvmname-0000007.vmdk. If that's the case, even if snapshot manager doesn't show any existing snapshots, it may be riding on that snapshot. Doing 'consolidate' in ESXi 5 fixes this, but I think the behavior is different in 4?

Someone posted a powershell script a while back that shows all snapshots, and it did show the snapshots in situations like this, even if snapshot manager didn't. You may want to run that before blowing anything away.

edit: or make an alarm for snapshot size over like 1k, and it should pick that up (but not guaranteed).

Erwin
Feb 17, 2006

luminalflux posted:

Well yeah but I'm getting tired of all these trips to the colo facility on the other side of town, and one host is running vms that aren't on shared storage for performance reasons (but downtime for these is begrudgingly tolerated). Guess which host has an iLO password tat nobody remembers? :v:

Are you licensed for storage vMotion? Also, what in the world shared storage do you have that under-performs local storage?

Erwin
Feb 17, 2006

Corvettefisher posted:


On our front end we have some nice Cisco C210 servers which perform well, all CPu's are intel,
All Enterprise Plus
C1 - 2 x 4 CPU's 64GB ram <= half of VDI, 1st Domain Controller, File server
C2 - 2 x 4 CPU's 64GB ram <= half of VDI, 2nd Domain Controller, Vcenter/View Server
C3 - 2 x 4 CPU's 24GB ram <= Pool reserved for HA/DRS, Standby
C4 - 2 x 6 CPU's 48GB ram <= Client Servers host
C5 - 2 x 6 CPU's 48GB ram <= Client Servers host


Is this the way it is now, or the way you're going to do it, or both? Why so many small clusters? My intuition would be 2 clusters, grouped by CPU count, and setting host affinity if you really want to keep that segregation in tact (not sure why you would, though). My reasoning would be more HA resiliency.

Edit, never mind I'm dumb.

Erwin fucked around with this message at 18:15 on May 18, 2012

Erwin
Feb 17, 2006

luminalflux posted:

Is there a way to extract performance metrics from vCenter to another graphing tool? I'd like to take the graphs I can see in vSphere Client under the "Performance" tab and get them into something like Graphite or Munin. I looked a bit at SNMP, but vCenter only has traps, no performance counters at all.

You can export to Excel, and from there go pretty much anywhere (you might be able to do CSV directly). Of course, it's a manual process. Also, you could probably query SQL directly.

Erwin
Feb 17, 2006

Frozen-Solid posted:

I'm assuming this is all some kind of VMWare magic with it caching read/writes? Just how long can a VM survive without access to the SAN?

Our test environment SAN locks up sometimes when you put in a new disk or replace a disk (Promise VessRAID, seriously don't buy it for production). One time it was almost 5 minutes. VMs obviously didn't like what was going on, but nothing locked up and everything was pingable. Even though it wasn't production, it was still several minutes of butt clenching.

Erwin
Feb 17, 2006

HalloKitty posted:

To be fair I don't really think CPU provisioning is too critical, and I'd give everything at least 2, and just use reservations to make sure each VM gets a guaranteed minimum. RAM of course I would prefer to provision rationally based on physical amounts. If that's 2008 R2, though, 1GB is a bit low, I would personally say. 8GB for a DC doing nothing else at all though? That's probably excessive for almost any business.

You really give every VM 2 vCPUs, or am I reading that wrong?

Erwin
Feb 17, 2006

HalloKitty posted:

No, I meant as a minimum generally, I didn't mean all VMs get a fixed number or something

A minimum of 2 is what I thought you meant. That's really stupid.

Adbot
ADBOT LOVES YOU

Erwin
Feb 17, 2006

Speaking of allocating vCPUs, can anyone speak to the best way to do virtual sockets vs virtual cores in vSphere 5? My assumption has always been 1 vCPU, 2 vCPU, 4 vCPU would be 1 socket 1 core, 2 socket 1 core, 2 socket 2 cores respectively, but it may not actually make a difference. I was at a VMUG today and someone asked that question and the presenter didn't know.

My only thought is that two sockets would allow the VM to straddle physical CPUs in the case of NUMA?

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply