@DustinB3403 said in Internal Storage is always better.:

@travisdh1 said in Internal Storage is always better.:

@DustinB3403 said in Internal Storage is always better.:

So this looks like an MD array, is this running on XenServer?

It is. Hopefully I won't have any oddness formatting these as ext3 when that completes.

I was going to ask the same question that @thwr asked, along with what is the hardware set?

WD RED and Seagate NAS drives for the RAID 10, whatever 1TB drives happened to be lying around for the two RAID1 sets. (I prefer HGST drives when buying drives today.)

ext3, because XenServer

Apparently, XenServer is one of those "we don't get storage" vendors.

@DustinB3403 said in XenServer local storage on a T3500:

@stacksofplates said in XenServer local storage on a T3500:

@BRRABill said in XenServer local storage on a T3500:

@scottalanmiller said in XenServer local storage on a T3500:

@Dashrender said in XenServer local storage on a T3500:

@scottalanmiller said in XenServer local storage on a T3500:

@BRRABill said in XenServer local storage on a T3500:

When would you NOT want to thin provision?

And are you saying if it is over 2TB you cannot thin provision?

That's the issue, thin provisioning is limited currently to 2TB or smaller volumes (VHDs.)

Wait, is it 2 TB volumes as in ext3 partitions or is it 2 TV VHDs?

VHDs.

That was another thing people are always clamoring about with XS ... the lack of VHDX.

Which I think is ridiculous that it's not in 7. VHDX is over 4 years old. Even worse, QCOW2 is something like 10 (or more) years old, and has almost a 9 EB limit.

The limitation on the partitions sizes is because of the use of snapshot functionality. Imagine having a 9EB array and partition, and then trying to perform a snapshot on it.

VHDX has the same limit (or similar). No one would really have an image that size, but it would be compatible with every other system. And snapshots wouldn't take any longer. It creates a new file and writes to it and reads from the old one.

I had to try and find a key for QuickBooks here, what a nightmare. Didn't have the login for their online site, and getting a confirm sent by phone was impossible due to them only being able to send text message to the registered phone on the account. Or an automated call. the phone is an ip phone that cant get texts and is answered by an attendant, so neither of those worked.
So I had to call them. Or it, rather. Had some very nice and unhelpful computer voice tell me to talk to it like a normal person. Not happening. Hitting 0 doesn't work. Infuriated, I hung up, and proceeded to search through years of old email to find the key. Eventually found the key in an old file folder in a file cabinet.
Really wish we could get away from it.

Well, didn't I just open a can of worms before unintentionally disappearing (friends computer needed some care last night.)

I only Bash my Linux boxes... just saying.

It's not really "any communications channel". It's not phone, IM, XMPP, email, etc. It's just a file.

@travisdh1 said in XenServer 7 being annoying today.:

@scottalanmiller said in XenServer 7 being annoying today.:

What error did it give with the GUI installer?

Just a blank screen. Didn't bother with trying to work out boot options to get the normal installer working since text mode works.

I've had this. You have to close XenCenter and reopen it. Sometimes moving to other tabs and tabs in other VMs works too. No idea what causes it

@travisdh1 I guess I should note that I got a laugh because the route to go from our site in Killbuck, OH to our host in Walla Walla, WA went

Cincinnati, OH Los Angeles, CA Monroe, AR Chicago, IL New York, NY Philadelphia, PA Atlanta, GA Seattle, WA Walla Walla, WA

Nothing like traversing the entire country 3 times to help with latency, right? Even with that, latency only averages 87ms with the Piladelphia -> Atlanta link being the slowest of the bunch.

@nadnerB said in Smart cars + smart phones = bad.:

@scottalanmiller said in Smart cars + smart phones = bad.:

@nadnerB said in Smart cars + smart phones = bad.:

@Breffni-Potter said in Smart cars + smart phones = bad.:

Nothing wrong with a smart car.

0_1473202814963_upload-69efb14e-4ec6-472c-9547-7136d7e69b96

Fun fact: Probationary drivers in New South Wales can't drive them (state law). They have a super charged engine.

However if you fitted a Suzuki Hayabusa engine... 😉

So New South Wales bases their laws on the tools under the hood, not on the results? Sounds like a government confusing their goals.

Yep. No one uses a power to weight ratio. They have been know to legislate like a they can't tell their collective arses from their elbows.

So naturally aspirated Ferrari or an electric Tesla that will distort your eyeballs are okay, but highly fuel efficient cars are outlawed.

I think I know who paid for that law.

@DustinB3403 said in Talking security during the server talk, question for you all.:

Are you going to be discussing Kali linux, or just touching on it?

Security is a very broad subject matter.

As this is supposed to be server, just talking about the basics to secure a Linux server. I'm thinking just cover the basics of "I've got my widget configured, what should I do before exposing it to the world?"

Kali would be bullet point, something like "Run the Kali hacking tools on it"

@JaredBusch said in ER-X static routing:

I would use a source and destination NAT rule to force it.

Thanks. I'll have to wait till after lunch or tomorrow to get that setup. Redoing the server over there as well, so no remote access at this point 😞 (I want my jumpbox back!)

@stacksofplates for HC3 Move it is actually in partnership with Vision Solutions. The full name is HC3® Move Powered by Double-Take®...Double-Take Move being the product we are OEM-ing for that.

We have played around with virt-v2v and virt-p2v, but haven't integrated them in the product yet.

@scottalanmiller said in AMD, trying to mount a comeback.:

@hobbit666 said in AMD, trying to mount a comeback.:

wow 32cores that's going to be expensive for licensing when everyone starts doing it based on cores not cpu's 🙂 Like Microsoft SQL and Datacentre

Only a few places do that, most are not based around cores. Most enterprise OSes don't need licensing at all. Remember that even 32 cores is tiny for big iron where they are way past these numbers long ago.

Oh, I'm reliving memories from 1997-2002 here, and the licensing associated with IDEAS, CATIA, and Pro-E. Most of those didn't have per-core licensing, but I remember one did. Just thinking about all the licensing servers in use at that place can cause me to have convulsions.

@travisdh1 said in Ubiquiti ER-X Initial Firmware Update:

So I uploaded the 1.9.0 firmware file with scp (the ER-X doesn't have internet access yet, not final install.)

add system image ER-e50.v1.9.0.4901118.tar

Complained about not enough space available. Fine, uploaded through the gui, and it worked fine. ???

I'll have to tackle the duel WAN tomorrow, after verifying I got all my documentation correct.

Well, it's setup and being used. I'm have a feeling I'm going to have to limit http and https traffic with the QOS settings on it tho. Two 700k/300k DSL connections don't go very far 😞

I also need to get the static routing figured out. I tried adding a static route to our web host (viviotech.net), and it still sends traffic out of whichever of the WAN ports it decides is better. Bet I forgot a NAT setting somewhere, now that I think about it?

The boxes are amazing for the price.

Apparently I figured out my issue with the dual wan traffic shaping. Had to turn on stickiness.... just goes to show you how little I've dealt with networking.

@travisdh1 If you know anyone that needs an IT guy that also has a Class A license, let me know!

Good information for me as I'm in the process of learning XenServer (although its 7)

My kids are still a little too young for this, but we will definitely be taking them when they are older.

@travisdh1 said in Friday afternoon account corruption.:

@Romo said in Friday afternoon account corruption.:

@travisdh1 Zabbix is a monitoring solution you're probably thinking of another product, probably Zentyal I would guess.

I bet you're right. I think it's a good thing I wasn't trying to fix things Friday, I would've just made things much worse!

"I'll jus treinstall Zabbix here and get everything back up and... hey what the crap is this?!"

Nice deal!

@thwr said in Raspbian and Gluster, annoyance.:

@travisdh1 said in Raspbian and Gluster, annoyance.:

@thwr said in Raspbian and Gluster, annoyance.:

@travisdh1 said in Raspbian and Gluster, annoyance.:

@thwr said in Raspbian and Gluster, annoyance.:

said in Raspbian and Gluster, annoyance.:

So, apparently the version of raspbian that DIY temp sensor is on, the version of gluster in the repository is 3.2.7. Which just happens to be the last version before they broke backwards compatibility, so the client on the pi can't connect to the gluster servers running 3.7.9. I'm building from source currently, and if it works, I'll post the condensed version of how I got it working.

I have a cross compiler VM somewhere, but I won't have a chance to look before the next weekend.

I think the problem is that the Pi2 is 32 bit, and the programming expects 64 bit now. So, in theory, a Pi3 would work.

Erm, can be yes. My CC build is also just 32bit, pretty sure. And the gluster stuff is 64bit arm?

The errors I was able to read during compile were complaining about variables defined as int64w, which wouldn't work so well on a 32 bit platform I don't think.

I can define a long int (64bit/8 bytes on most platforms) just fine on any 32bit CPU. You can even do that on a 8 or 16 bit uC. So probably the failure is somewhere else?

Possibly, let's see if I can find that compile-time error here.