Posts made by NetworkNerd

I think the costs associated with more labor, more upkeep, more overall administration of a solution moving forward often get overlooked when decision makers begin to see how much money comes out of the bank account. The long-term costs are often not communicated to the big boys well, not communicated at all, or just flat out ignored.

@ajstringham said:

@NetworkNerd said:

Maybe I could use the HP DL 385 G5 I have with the MSA 70 and run something like FreeNAS on it for another datastore option?

That's one option. Something that would be good for, to be honest, would be your StorageCraft server, if you go with that.

I was thinking use it as a NFS datastore and a place to store the VM files for my current physical machines like was mentioned above. That saves money on a secondary backup solution.

Maybe I could use the HP DL 385 G5 I have with the MSA 70 and run something like FreeNAS on it for another datastore option?

@thecreativeone91 said:

I would keep the file server as a windows file server on a VM. I personally don't like a NAS being a file server. They are great backup devices but will bog down with many con-current connections.

That would be my preference as well.

@ajstringham said:

Also, you talk about backing up the remote site offsite. How much data we talking here? Syncing the Drobo, or files in general, to the main site over a 10Mb pipe is gonna take forever. What data is there at the remote site that isn't at the main site?

There are at least 4 VMs at the remote site that are not at the main site (file server, DC, Spiceworks server / Veeam Proxy, Engineering software workstation). It's probably 500 GB of data but much less when we talk about the size of the backups themselves.

When you say sync, keep in mind I cannot sync the Drobos as I do not have the right models for that.

@ajstringham said:

@thecreativeone91 said:

@ajstringham said:

Also, you talk about backing up the remote site offsite. How much data we talking here? Syncing the Drobo, or files in general, to the main site over a 10Mb pipe is gonna take forever. What data is there at the remote site that isn't at the main site?

Not if He syncs them first then just does incremental.

That's why I asked how much data he has. Still, a 10Mb pipe is pretty small. Even running it unthrottled off-hours won't yield much in the way of transfer.

I think the WAN Acceleration piece from Veeam will help in this endeavor but am not sure how much.

@Dashrender said:

@NetworkNerd said:

The servers that are physical are running Server 2003, so it is time for them to go. But, one of them (main file server) I cannot just P2V or rebuild easily.

You can't P2V only because you don't have the needed storage on the ESXi hosts.

Do you have an ESXi Essentials package? In otherwords, do you have the ESXi host license? If so, you could (over a long weekend) image the file server to a cheap NAS or other available storage, then rebuild that server as a ESXi host, and restore the image. This gets you into using Veeam to backup that server.

Yes, we have the Essentials Kit license, but technically we are using all 3 licenses for the 3 hosts I mention here. We'd have to get another Essentials Kit license (cheaper than Essentials Plus). Actually if I am running the storage part from a NAS, I could use RAM and processor from one of my existing hosts.

Keep in mind this is a DC as well. Aren't there bad things that frequently happen to people who P2V DCs?

@ajstringham said:

For the 2003 file server, I'd say replace it with a NAS and incorporate its roles into existing servers, or spin up a new VM if you need to. How much storage we talking here?

It's about 700 GB worth of files on that server.

@scottalanmiller said:

Nothing compares to actually attending. It's held in Barcelona, my favourite city ever.

I'm sure being there is best, but when you can't, just watch it online and pretend you are running with the bulls.

Warning...this is one of those projects that was entered into where at the beginning we decided to virtualize everything slowly as we went, and as you can see, we are still not finished. And the company is pushing everyone to not spend if we do not have to spend.

Right now I have 3 ESXi hosts (2 at our main site and 1 at a remote site, all local storage with 10K SAS and OBR10). All are getting backed up with Veeam. Here's the current layout:

Remote Site

• Veeam backups of ESXi box to onsite Drobo 5N (30-45 day retention)

• File servers backed up with Crashplan Pro for offsite file backup

• No good offsite backup of Veeam backups

• Time Warner 10/10 fiber for internet

Main Site

• Veeam backups of ESXi hosts to Drobo B800i (30 day retention), copied to removable HD and taken offsite (2 week rotation of offsite drives)

• There are also 3 physical servers that are being backed up with an old version of BackupExec (no longer under maintenance) to the same removable hard drives I mention above. The 3 physical machines are the BackupExec server, a file server / DC / web server / print server, and a DC / file server / FTP server.

• We're backing up one of the physical servers mentioned here with Crashplan Pro also.

• Charter 50/50 fiber for internet

So I am looking to change things up a bit and need some advice. I have license to Veeam Enterprise Plus and can do BackupCopy jobs, but I also have those blasted physical boxes.

The servers that are physical are running Server 2003, so it is time for them to go. But, one of them (main file server) I cannot just P2V or rebuild easily. There are so many files entrenched in various places in our company intranet mapped to those file paths, it is truly a Hurculean task to even think about doing. Let's combine that with the fact that I do not have the storage on my 2 ESXi boxes at the main site to absorb the file server mentioned here, and we have a mess.

Option 1

1. Beef up storage on the remote site's Drobo and run Veeam BackupCopy jobs from the main site to the remote site. This is a lot of data (probably 100 GB of backups per day) over a slow link.
2. Beef up storage on the Drobo B800i. Run Veeam BackupCopy jobs from remote site to main site Drobo for an offsite solution.
3. P2V the BackupExec server, continue to use it to backup physical machines at main site, and run backup jobs to a removable HD connected to the ESXi host.
4. Push for a new ESXi host / more storage in existing hosts.

Option 2

1. Beef up storage on the remote site's Drobo and run Veeam BackupCopy jobs from the main site to the remote site and do not put them on removable HDs at all. This is a lot of data (probably 100 GB of backups per day) over a slow link.
2. Beef up storage on the Drobo B800i. Run Veeam BackupCopy jobs from remote site to main site Drobo for an offsite solution.
3. Renew BackupExec / look at StorageCraft or Unitrends to backup the physical machines at the image level.
4. Push for a new ESXi host / more storage in existing hosts.

My goals are really to make sure the remote site has a better offsite backup and to do what I can to protect this behemoth of a file server that cannot go away yet. I'd love to rid myself of the burden of using the HDs and taking them offsite.

I have one spare HP server GL385 G5 with a MSA70 attached and a bunch of storage that I fear I may be forced to use for any growth, but I do not believe the gear is on the VMWare HCL.

I should also add that RTO is 1-2 hours at the main site, probably closer to 8-24 hrs at the remote site.

So if you are among the many who cannot experience VMWorld Europe, I just received a notice from VMUG that VMUG Advantage members can see it for free. To clarify, VMUG membership is free, but the Advantage membership gives you discounts on VMWare course learning and exams. I thought I would share.

@JaredBusch said:

@NetworkNerd said:

It's Trutops Inch from Trumpf in this case.

I have that installed at a client (not in a VM granted) and do not recall a USB dongle being required. But the install was quite a bit ago. maybe I lost those brain cells.

The final install did happen with me on a remote connection and a conference call with the Trumpf support people. It is entirely possible the user plugged in a stick to license.

In the past it has always been USB stick required, but in this particular case, we got a brand new laser from Trumpf and the software that goes with it. I was pretty surprised when there was no USB HASP to plug into the server since that has been the trend.

@scottalanmiller said:

@Dashrender said:

I didn't know VSA used a VM on each host to do it's job. How does it control the disk beneath the other VMs?

You can build your own VSA to see how it works. You can do it with Linux or BSD quite easily. You build a virtual NAS (which is what VSA means) and use DRBD (Linux) or HAST (BSD) to make the cluster work. You share the storage to the local machine via NFS. Now you have a VM that can provide storage for the other VMs locally.

Quite easily to SAM is not so easy to the person who is semi-familiar with Linux.

@Dashrender said:

@scottalanmiller said:

@NetworkNerd said:

I thought the VSA had to be setup a certain way from the beginning but was near impossible to add to the cluster later (because a certain amount of storage on each host was to protect against another host failing) whereas these software solutions would be able to install in an existing environment non-intrusively and allow you to add hosts / more storage at any time.

Yes, but they are all VMs.

I think NetworkNerd is saying that you can't (his and my understanding) add VSA after the fact because the underlying disk that ESXi is using is already partitioned off, so there won't be any free space, or most likely not enough, to implement VSA after the fact?

I didn't know VSA used a VM on each host to do it's job. How does it control the disk beneath the other VMs?

Yep - that's exactly what I meant.

@Bill-Kindle said:

@NetworkNerd said:

The install was successful, so it is now time to activate our shiny new software. But there's one glaring problem here when I went to enter the license activation key. I'm sure someone can tell me

I've worked around an issue similar to this by using a USB > Ethernet adapter from Digi.

However, I didn't see what software you are using?

It's Trutops Inch from Trumpf in this case.

I went ahead and proceeded with activation despite the warning, and the software seems to function as expected for now. We'll see if that is also the case when the Engineer who needs to use it starts playing with it.

In the past we have had HASP keys that had to be plugged into the server / workstation, but in this case, all we got was an activation key (no piece of hardware to go with it). But overall, when faced with using a HASP key for this kind of software in conjunction with ESXi, passing the USB drive to the VM has worked beautifully for us.

We have a large number of lasers in our manufacturing shops, and many times when we get a shiny new laser or some other machine we get new offline programming software. In the last couple of years, I have been making a large effort to virtualize all of these installations because it is extremely painful to transfer the program from one machine to another.

So, just last night, I was building a VM to use for this very purpose and came across something really interesting. The software itself runs fine in a VM and is supported. The install guide says it is fully supported. But, Engineering software normally comes with some kind of fun activation process like using a USB HASP, forcing the install of some type of license server, etc. In this specific case, there is a license manager application that has to be installed in order for you to use the software. Since this is a single workstation installation and will not grow past that, you can install everything on a single machine for simplicity, which is exactly what I did.

The install was successful, so it is now time to activate our shiny new software. But there's one glaring problem here when I went to enter the license activation key. I'm sure someone can tell me what it is.

So why support the installation on a virtual machine if I have to have a physical machine to run the license server? This makes no logical sense to me.

I thought the VSA had to be setup a certain way from the beginning but was near impossible to add to the cluster later (because a certain amount of storage on each host was to protect against another host failing) whereas these software solutions would be able to install in an existing environment non-intrusively and allow you to add hosts / more storage at any time.

Infinio sounded cool but will only work for NAS or SAN from what I remember - no local storage or DAS (at least not right now).