The Comms Room

MDMarra

15:01

@RyanRies No, because I'd have to live in Texas

:D

Ryan Ries

Even though this job is stressful as hell, I think probably the only thing I'd leave for is Azure :)

But you never know, my mind can be changed

Travis

The biggest decision I have to make today! What to eat for lunch.

cole

@RyanRies Interesting.

However, the only place I'd probably move to in TX (and have been kinda looking) is Austin.

BigHomie

@MDMarra can any of your sccm goons answer any of these? serverfault.com/…

Ryan Ries

Yeah I feel ya... other than a relatively inexpensive cost of living, we don't have much going for us down here :(

cole

15:04

I hear Austin is pretty cool so

BigHomie

in particular this latest one serverfault.com/questions/583149/…

MDMarra

@BigHomie I just pretend to know SCCM. I mostly do SCSM and SCO work these days. I get my SCCM info from guys that work for me, as I QA all of their design docs before they get to the customer :)

Ryan Ries

As Patton Oswalt described it, Austin is the one little bubble of sanity in this state. :)

HopelessN00b

@BigHomie I took a whack it, got all frustrated and "fuck PowerShell"-y, and then decided to drink a bunch of scotch and take a "nap."

BigHomie

@MDMarra I kinda figured, that's why I mentioned your goons

Travis

15:05

Does Server 2012 still have shitty file server performance like 2008

MDMarra

@BigHomie this is true

My goons don't Server Fault (as far as I know)

I barely answer anything anymore. I get too frustrated

BigHomie

@MDMarra I'll be deploying scom on our desktops soon so expect more questions on main about that

MDMarra

@Travis Define "shitty" SMB3 is fast enough to host SQL Server databases and Hyper-V VMs

@BigHomie woah SCOM to the desktop. That's hardcore.

What's the driver for that? Just super-proactive maintanence?

ewwhite

@MDMarra fanboi

Travis

@MDMarra Since migrating our file shares from '03 to '08 it takes 45 seconds or more for explorer to browse to the UNC path

BigHomie

15:07

@MDMarra yep, need some proactive monitoring, and the terminology has a serious learning curve to me

Nathan C

@Travis DFS or anything?

MDMarra

@Travis That's not normal for any version of Windows. You broke something :)

Tanner

@Travis that's a problem.

BigHomie

@MDMarra are they behind the curve or do you keep the treadmill on 10

Nathan C

@BigHomie 11

BigHomie

15:09

lol

Travis

@NathanC No DFS that I know of

cole

@RyanRies Seems like an okay place for me to be.

MDMarra

@BigHomie Well, we've hired 5 guys in the last 8 months

Travis

@MDMarra Wish I knew what it was.

MDMarra

So we busy busy

cole

15:10

Dayum, 5 guys?

Travis

and now when using the "search" box in Windows 7... it takes forever for stuff to find

MDMarra

File shares arent indexed by windows by design

BigHomie

@HopelessN00b It boils down to calling a wmi method from powershell, but I can't find the object that has the method, and/or don't know why the method isn't showing up on the object its supposed to

MDMarra

you can install the windows search service server-side or add the shares to a library client-side to enable indexing

BigHomie

@Travis is your search service running?

Travis

15:12

@MDMarra Right, but just typing "calc" or "compmgmt.msc" yields piss poor results

MDMarra

@cole SCCM, VMM/Hyper-V, Lync/Exchange, SCOM, and there's an AD guy in the works

@Travis oh, on the server?

Yeah dude, something is busted

Travis

@MDMarra No, workstation

@MDMarra Should be. Never turned it off.

Tanner

@Travis It will still pick through the server files, which is what's slowing down your search for something as simple as 'calc'

cole

@MDMarra Damn dude.

Travis

The only change that was made to my workstation was I mapped an R: drive to a network path...

ewwhite

15:13

Enough of this Windows talk! Where are the Linux people?

Travis

kill -9 @ewwhite

Nathan C

Hmm...a good two dozen people just walked by my desk ...people from a company we do a majority of business with lol

Journeyman Geek

rm -rf

;p

Travis

@Tanner So it has something to do with network shares being slow also...which leads me back to my original problem

All this crap stemmed from migrating a SQL database over to a VM from a physical... I just can't seem to find any bottlenecks... Nothing in performance monitoring stands out as bad.

HopelessN00b

Tried wiresharking things?

Tanner

15:17

@Travis If you punch in `\\shittyfileserver\somefolder` , does it connect immediately and slowly populate the view, or eventually connect and instantly displays the contents?

BigHomie

@Travis nm

ewwhite

:(

Travis

@Tanner Eventually connect and instantly displays. Then a few minutes later I can punch it in and immediately populates

Nathan C

annnnd....bag of popcorn devoured. stomach demon satisfied

Travis

@HopelessN00b Not yet. @BigHomie Yep

BigHomie

15:18

@Travis and I'm assuming you looked @ disk write times?

Tanner

Yeah, I think @hopeless has it right, bust out wireshark

BigHomie

@Travis the database is on a vhd? ssd?

Travis

@BigHomie It's on a SAN, 600GB 15k SAS

BigHomie

ahh, phew

Travis

RAID 10

10 drives... So IOPS shouldn't be an issue

BigHomie

15:20

are the network shares on the same SAN?

Travis

@BigHomie Yep. All of the VM infrastructure is on the same SAM. Different LUNs

@Tanner What am I looking for in wireshark as an anomaly?

Shane Madden

What kind of VM? Can you easily check what I/O latency it's getting?

Travis

@ShaneMadden What do you mean what kind of VM

Tanner

@Travis I'd look at name resolution first. Is it looking for the server using DNS or NetBIOS? Is it getting a response in a timely manner? If it is, take a look at the SMB traffic and see where it's hanging up.

Travis

@Tanner ok

Shane Madden

15:24

@Travis What hypervisor it's running on

Tanner

@Travis This might be handy if you want a packet capture server side: blogs.msdn.com/b/canberrapfe/archive/2012/03/31/…

and don't want to get funny looks for installing wireshark on a server... not that I've done that... or gotten funny looks...

Travis

@ShaneMadden ESXi 5.0

ewwhite

@Travis UPGRADE!!!!

Travis

@Tanner I'm the only one that can funny look wireshark on a server

@ewwhite To windows 2012? ;)

Shane Madden

@Travis It'll track I/O latency in the performance tab - check and make sure it's not too high.

ewwhite

15:26

@Travis ESXi. There's no excuse.

@ShaneMadden You know this is my new crusade.

Travis

@ewwhite Laziness is always an excuse!

Shane Madden

@ewwhite ? Oh, upgrades?

Nathan C

@Tanner I've installed wireshark on a couple servers...I have the portable one though

Install winpcap on start, removes it when done. no trace on the server

ewwhite

@ShaneMadden People don't upgrade their ESXi.

Shane Madden

@ewwhite They most certainly do not.

Tanner

15:27

@NathanC handy

ewwhite

There's no excuse to be on ESXi 5.0

Nathan C

I have it sitting on my network drive just for that :p

Travis

@ewwhite I like vSphere ?

Shane Madden

@ewwhite I've still got two 4.1 boxes, heh.

ewwhite

@Travis it's a larger issue.

Travis

15:27

@ewwhite No, in all seriousness I did download the latest version yesterday and haven't put it on yet

ewwhite

@ShaneMadden 4.1 is understandable

5.0 isn't

HopelessN00b

@ewwhite I tried, but my manager said "no." Moron. >:/

@ewwhite How do you figure 4.1 is cool, but 5.0 isn't?

ewwhite

@HopelessN00b because there are significant barriers to move to 5.x from 4.1... Logicworks had dozens of ESXi 4.1 hosts in production.

but once on 5.x, there's no excuse not to keep shit updated

Travis

@ewwhite What's wrong with 5.0?

HopelessN00b

@ewwhite If I give you a phone number, will you scream that at the douchebag on the other end of the line?

Shane Madden

15:30

@ewwhite I don't think it's understandable in this case. All the other nodes are the same exact build of 5.1, these two are some cisco setup for the phone system. Not sure why they didn't just put the phone VMs on the main cluster.

HopelessN00b

@Travis Other than it's not compatible with like Server 2012, for example? :p

Travis

@HopelessN00b Pfft... We'll be on Server 2015 before we upgrade ;)

ewwhite

@HopelessN00b yes.

13

Q: I'm tired of answering questions with: "Upgrade the firmware" or "Run updates"

...or sometimes both! ESXi 5.1 crashing (bug fixed by updates) Installing XenServer 6.2 on Proliant DL160G6 freezes during "Installing from base pack" (firmware) Disconnected Network Adapter on VMware - only power cycle solve this (firmware) Unable to view sensor data in ESXi 5.x on DL380 G7 ...

HopelessN00b

Fair enough. I think what I need help doing is creating PSODs and crippling bugs/crashes on our older ESXi hosts. "Told you we need to upgrade..."

Travis

@ewwhite So you're saying I should run updates and upgrade?

MikeyB

15:34

@ewwhite "But we have 500 machines all running the same firmware, why does only this one have trouble?"

HopelessN00b

Hmm. Would that be an acceptable SF question? "How can I create bugs/crashes on ESXi? [old version]"

BigHomie

@HopelessN00b yeh if enough people upvote it

Chris S

problem with a program... look at code.... THE HORROR. It's littered with different ways of checking for errors, but they all just goto error with no description of what failed. Makes me want to sacrifice a developer or two.

HopelessN00b

An alternative method is to panic the ESXi server's kernel, forcing a Purple Screen of Death (PSOD) and a resulting hard failure. You can accomplish this within the ESXi console by using the following command:
vsish -e set /reliability/crashMe/Panic 1

Hmm. I'm very tempted to try that.

ewwhite

@HopelessN00b Yeah, you can create PSODs... Send an NMI to the host from the ILO

@MikeyB That's troubleshooting

HopelessN00b

15:40

@MikeyB "Because your mother. Who cares? Upgrade."

ewwhite

@HopelessN00b test the NMI approach... it'll crash your servers :)

HopelessN00b

@ewwhite Yeah, I'm tempted. Then again, I'm outta here as soon as I can get out, so I also... don't care anymore, at this point.

Seems like just pointlessly creating work for myself at this juncture, I suppose.

BigHomie

@ChrisS haha error in <program>. If this continues, contact technical support.

ewwhite

@MikeyB Ever use turbostat?

HopelessN00b

@BigHomie The worst are [error]: contact your system administrator. I AM MY SYSTEM ADMINISTRATOR, BITCH!!!!

10

BigHomie

15:44

@HopelessN00b yeah that one too.

Dan

@HopelessN00b More to the point "I AM THE SYSTEM ADMINISTRATOR AND I HAVE NO CLUE WHAT'S HAPPENING"

HopelessN00b

@Dan That would be why they're the worst. If I'm contacting anyone, it's gonna be the developer(s). With a blunt object.

Dan

@HopelessN00b I'd prefer to contact them with a rocket propelled grenade

HopelessN00b

@Dan Expensive, and hard to get through customs, though.

My steel toed boots, on the other hand, get through customs every time.

Chris S

@BigHomie Almost... ERROR: Could not initialize FFTW. Bad wisdom?

BigHomie

15:50

yeah I don't get why I can't talk to developers at $ShoddySoftwareCorp for support, especially tech support can't friggin help. Does anybody know how the software works?

Jenny D

@HopelessN00b "I'm not a bastard, I'm the bastard. And that's Ms Bastard to you."

HopelessN00b

2

cole

lol

HopelessN00b

I suspected as much, but have to try every now and then.

(Missing a free lunch with a vendor 'cause I'm on the phone with our current webfilter vendor's support team.)

ewwhite

@HopelessN00b BARRRRRRACUDA

MikeyB

15:55

FC1 server, SSH accessible from anywhere, no fail2ban, root ssh password allowed, root password of 'pizza1', what could go wrong?

@ChrisS Fast Fourier Transform... Wizard?

HopelessN00b

@ewwhite As far as webfilters go, I like them. At least it's not the cloud crap we have now, which I have to support, but have no control over. :(

MikeyB

@ewwhite no: git://github.com/ajaiantilal/i7z.git

ewwhite

Client just had multiple switches reboot overnight..

won't give me enable password (ProCurve) because it's the same as their domain password.

Nathan C

ESXi keeps telling me "Network uplink redundancy lost" but I have two NICs active....

ewwhite

@NathanC Right-click on host and "Reconfigure HA"

Nathan C

16:05

@ewwhite Grayed out

ewwhite

@NathanC is this a single server?

Nathan C

@ewwhite I have two hosts, but both are independent other than being managed together on vcenter

single server yes

I reset it to green and it hasn't come back...so..

ewwhite

@NathanC Oh, make sure both uplinks are active for the management port group

MDMarra

Override Switch Failover - BLASPHEMY!

:)

Shane Madden

All kinds of policy override blasphemy going on there!

ntrrgc

16:10

I received this yesterday on the SUSE security mail list lists.opensuse.org/opensuse-security-announce/2014-03/…

Tell me I am wrong. Did they have an outdated Xen affected by 47 known vulnerabilities for years in one of their Enterprise versions?

HopelessN00b

@ewwhite So... "fix it your damn selves, then?"

ewwhite

@MDMarra no it's not. You're supposed to pin opposing port groups to different adapters.

MDMarra

@ewwhite Only if you don't buy a server with enough NICs to bind management and vMotion to different physical NICs :)

ewwhite

@MDMarra Multi-NIC vMotion.

MDMarra

Whats that mean

ewwhite

16:13

@MDMarra No Bitchassness

2

A: Network configuration for VMware w/ vMotion and a single switch

This is good as-is. You won't need a special VLAN for vSphere replication. For small installations I often go with 4-pNIC solutions. Below is a 2 x vSwitch setup with VM traffic on vSwitch0 and storage on vSwitch1. Public, management, private and vMotion networks are on their respective VLANs, ...

MDMarra

Right, I meant you should have a pair for management, a pair for vmotion, and then however many NICs you need for data

Shane Madden

@ewwhite For active/standby?

ewwhite

@ShaneMadden Yeah.

Shane Madden

@ewwhite Ahh. Why no active/active?

MDMarra

You're talking about having management and vMotion on the same vSwitch over the same physical nics, but pinning active/standby to opposite physical adapters for each task, right?

ewwhite

16:15

@ShaneMadden Determinism... If I have 4 NICs, two dedicated to storage, two to management and vMotion, I typically want to be able to predict what traffic is flowing where.

MDMarra

@ewwhite My point was that NICs are cheap! Get 2 more ports and have dedicated vMotion interfaces!

ewwhite

@MDMarra Colo space, PCIe cards and switch ports aren't necessarily cheap.

HopelessN00b

@MDMarra Ports may be cheap, but so are my employers.

ewwhite

But my own installs are 2 x 10GbE most of the time anyway.

MDMarra

Switchports aren't cheap in a colo, but if you can afford to license vSphere, you can afford a NIC with two more ports

Shane Madden

16:17

@ewwhite For troubleshooting purposes?

jscott

@MDMarra Russian keygen is only USD $50.

ewwhite

@ShaneMadden depends on the environment. Most clusters don't have a tremendous amount of traffic coming out of them. A small installation with 2 to 3 hosts needs resiliency across port groups, but isn't passing too much traffic. vMotion is the only thing that will peg a link. So I use the same vSwitch as management and VM traffic for that.

Nathan C

@jscott torrent keygen is $0

I've done vMotions over the same link as data and it didn't appear to affect anything

but, I've only done that a couple times to bounce a couple servers around when we moved...

Shane Madden

@ewwhite Ahh, gotcha. Yeah, I like to be able to have multiple VMs in the same port group using different physical NICs to spread the load. And to know that nobody's gone and screwed up the switch port config on the standby uplink.

ewwhite

That's what the LACP is for ;)

I'm guilty of using LACP as much as possible now

just can't do it on the smaller sites

Shane Madden

16:23

Depends on the load. In a lot of cases, active/active with balancing on port id is nearly as good as LACP. Without the pain in the ass.

MikeyB

@ShaneMadden I don't see why people perceive LACP as a pain in the ass. They must be using shitty switches?

pauska

active/standby management and standby/active vmotion on the same port group is common practice

Shane Madden

@MikeyB Well, yes. But when you're comparing load balancing that requires config on the switch against load balancing that doesn't care what the switch is doing.. pain in the ass is relative. :)

MikeyB

@ShaneMadden OK when you're doing VMware, yeah, may as well let vmware handle it.

ewwhite

@pauska exactly!

But I'm a fan of the distributed switches now.

pauska

16:26

2x10gig is bad practice unless you're doing CNA

you really really don't want vmotion traffic to hit the same port as storage traffic

ewwhite

@pauska NIOC

Oops... need to update it after changing vCenter.

MikeyB

@ewwhite vNICs with reservations?

ewwhite

pauska

@ewwhite meh..

If you already have 6x10gig then I don't see why you would put storage in the same group

it's so much safer to separate it from other traffic types

ewwhite

Oh, that's 2 per host in this 3-host cluster

pauska

16:36

oh

well ok, better than nothing I guess

monitor your switch port buffers! :)

Nathan C

Cool. I figured out how to lockdown all the doors in the building with our security software. :p

ewwhite

@pauska 5400zl

HopelessN00b

@NathanC now you just need to figure out how to do that remotely. "Sorry, boss... tried coming into work today, but all the doors are on lockdown. Let me know when you fix it so I can come back to work."

Nathan C

@HopelessN00b I can do it remotely :P

All the doors are controlled by software on one of the servers....just need to be connected via VPN lol

HopelessN00b

There you go... never have to come into work again. Hmm. Wonder if this why they don't let me admin the security systems...

Nathan C

16:48

lol

Unfortunately it logs everything, so it'd be stupid to attempt something like that anyway. The amount of power the software grants is pretty awesome though

I almost wonder if we could somehow patch the security system into our alarm system (separate vendors) so if the security alarm trips it'd lock all the doors. :P

MDMarra

This is sad

I googled an exam number to see what certification(s) it applies to and the first 10 pages of google results are braindumps

Travis

I applied hotfix KB2550581 and rebooted...it seems I have no delay in accessing file shares now... so .!. @ewwhite and your upgrades!

On a side note, @ewwhite, I'm burning esxi 5.5 now to install ;)

@NathanC IT people can lock doors. Smart IT people can lock doors then wipe the logs without any trace :)

Nathan C

heh

Jacob

@MDMarra What's a braindump?

MDMarra

test dumps

files with the actual exam Qs and As in it

Jacob

16:56

@MDMarra How is that possible?

Travis

I'm always semi-aroused by @MDMarra's avatar

MDMarra

@Jacob What do you mean?

@Travis ;)

Jacob

@MDMarra Do people try to memorize it to write it out later or something?

Travis

@Jacob Probably leaked tests just like leaked OS releases

MDMarra

@Jacob Sometimes. Other times, corrupt Prometric/Pearson employees sell the exams to scum sites like Pass4Sure and then people pay for them. Then after they pass they post the paid-for files on free sites and bam - 10 pages of google results full of cheats

Jacob

16:58

MDMarra

There are even services where for $1000, someone will take an exam for you. Their friends (or someone on the take) will work the sign-in counter at the testing center so they dont check ID.

Jacob

The Simposons wriiters must be on the inside

Travis

@MDMarra For $1,000 I will take any test you want!

MDMarra

haha

Jacob

@MDMarra That's a shame.

MDMarra

16:59

Fairly sure the punishment for getting caught is losing all certs from that vendor and being banned for life

Travis

*no warranties or guarantees expressed or implied

MDMarra

So that would rule you out for ever working for a partner :)

Jacob

@Travis Pretty sure that @MDMarra can handle it by himself. :)

Travis

@Jacob Pretty sure any test @MDMarra wants to take probably would not be one I could pass

MDMarra

Nah, I don't have impressive certs

No MCM/MCA or anything. Just a couple of MCSEs, which aren't super prestigious anymore (partly because of all the brain dumps) and a VCP 5

HopelessN00b

17:03

@MDMarra Um, easier approach is photoshopping up something to put over a proper government ID (like on a transparency, appropriately cut to size). Not hard, and it's not like the people who do check IDs do a thorough job.

MDMarra

Some require a photo - VMware and Cisco for example

Jacob

@ewwhite Can NewRelic monitor individual services?

MDMarra

They keep a photo of you on file

@HopelessN00b Not to mention that people doing this for profit probably take dozens of exams per week under various names

HopelessN00b

@MDMarra Point being that a fake ID, especially one that's not validated by an actual professional, is almost trivial to create these days.

MDMarra

Right. But say I sit for a VMware exam. They take my picture at the test site and keep it on file.

HopelessN00b

17:06

@MDMarra Fuck VMware anyway... for what they charge for that rip-off class you need to get a VCP, even asking for a photo seems especially low.

MDMarra

So unless the same exact ringer takes the test for the cheater for all VMware exams for the rest of time, it wont work with a fake ID

@HopelessN00b I got my VCP without the class :D

So I wouldn't know

HopelessN00b

@MDMarra spfxmasks.com

MDMarra

hahaha

HopelessN00b

@MDMarra How'd you pull that off? And more importantly, is it reproducible?

Jacob

@HopelessN00b Seriously? As much effort as setting up that would take just learn the material. :)

Bob

17:08

@MDMarra Unfortunately requires Win8 clients, though.

Lucky enough to not be on XP anymore. Mostly.

Jacob

@Bob How long have you been awake man?

Bob

@Jacob Uh. Afternoon-evening nap? :P

HopelessN00b

@Jacob Yeah, but with a proper silicone mask, you can do other things, too. Like get a second passport so you can get to a non-extradition country if needed. :)

MDMarra

@HopelessN00b VMware runs a promotion for partners of a certain status from time to time. There's a promo code that employees can use to be certified by passing only the test. The thought is that if you're a partner and have whatever the status is, you'll have had adequate on-the-job training by certified engineers from inside your organization

ewwhite

@Jacob Naw.

HopelessN00b

17:09

Ah, damn. Nothing that's likely to apply to me anytime soon, then.

ewwhite

Unless it's Postgres/programming framework, etc.

Jacob

@ewwhite What do you use to monitor processes ?

Nathan C

ewwhite

@Jacob mmonit.com/monit

Nathan C

Sad truth...

@ewwhite "Did you get that alert?" "I'm monit!" ...wait..

Bob

17:12

So. Dell replaced one of the RDX drives. Here's hoping it works.

Joel E Salas

@MarkHenderson openvpn is pgood

Bob

What kind of fucking backup software bluescreens the fucking server when fucking dismounting fucking drives? Gah.

Joel E Salas

@Bob Why's it dismounting drives?

Bob

@JoelESalas Limitations of backing up onto RDX => creates a virtual drive container thing => needs to mount pre-backup and dismount post-backup

That's all handled by the backup program. Supposedly.

Joel E Salas

RDX

| Section2 = | Section6 = | Section7 = }} RDX, an initialism for Research Department Explosive, is an explosive nitroamine widely used in military and industrial applications. It was developed as an explosive which was more powerful than TNT, and it saw wide use in World War II. RDX is also known as cyclonite, hexogen (particularly in German and German-influenced languages), and T4. Its chemical name is cyclotrimethylenetrinitramine; name variants include cyclotrimethylene-trinitramine and cyclotrimethylene trinitramine. In its pure, synthesized state RDX is a white, crystalline so...

Chris S

17:14

@MikeyB fftw.org ~~No idea what the "W" stands for~~ "Fastest Fourier Transform in the West"

Bob

@JoelESalas Except when it's not. It clashes horribly with certain network drivers commonly found on laptops. OpenVPN running + sleep => bluescreen :(

@JoelESalas Heh. I wish I had some handy.

RDX, also known as QuikStor and RD1000.

Jacob

@Bob Really? Never seen that happen.

Bob

@Jacob Had two HP laptops that did that.

HopelessN00b

@JoelESalas I like that idea... RDX drives. Better than FDE - unauthorized access scrambles the drive (and the access-er) across 10 square city blocks.

Bob

Quite a lot of info (but no fixes) available if you googled it. I just remembered to disconnect before suspending.

Jacob

17:17

@HopelessN00b C4 is better. You won't explode from sneezing.

Bob

@JoelESalas Basically, some standard for ruggedised removable hard drive cartridges and caddy.

Worst decision ever.

Joel E Salas

@Bob Gotcha.

I'm going full commodity with our storage

Bob

Should've gone with LTO. Or even consumer USB HDDs, those were more reliable -__-

Travis

My company paid for a VCP class ...it was a waste of money.

Joel E Salas

@Bob don't knock what you can do with a large sum of USB drives..

HopelessN00b

17:19

@Jacob RDX comes as a plastic explosive, too. Stable as C4.

Jacob

@HopelessN00b I'm thinking of T4 then.

Dammit @HopelessN00b I'm a Sysadmin, not a chemist!

2

Joel E Salas

@Jacob Also not a miracle worker?

Jacob

@JoelESalas No because if I was I'd be the first person I helped.

HopelessN00b

@Jacob Probably not. T4 is a synonym for RDX. :)

Jacob

@HopelessN00b See above.

Bob

17:21

@JoelESalas Probably also want to avoid literally knocking them.

Jacob

I know more physics them chem

Bob

Those things

They don't even eject properly :\

ewwhite

11

Q: Are RDX removable disks a good replacement for LTO tape?

Over a period of three weeks, I experienced six complete failures of LTO-1 and LTO-2 tape drives at client sites. Some had failed mechanisms. Others lost the ability to write reliably. These were HP Ultrium 232, 448 and 460 drives. Most of these units were deployed between 2006 and 2008, so the t...

backup hard-drive tape

Bob

@ewwhite It's not on USB2, thankfully.

I haven't actually tested speeds, though.

HopelessN00b

@Jacob I'd think that sysadmining would leave you more inclined to arcane knowledge in tools that you can use to dispose of lusers permanently.

Bob

17:24

Oh, they're up to 1.5 TB now.

Jacob

@HopelessN00b Oh I do... Firearms.

Matt Bear

@Jacob yay guns!

Nathan C

We use rotating seagate passport drives for off-site backups

Matt Bear

I have guns.. I like guns, I want more guns, but this cracks me up cause I know this guy IRL

►

Nathan C

a nice rugged case to go along with it...and indeed, I accidentally dropped one and it still works just fine without any problem

Bob

17:26

@ewwhite Did you end up using them?

HopelessN00b

@Jacob Yeah, but guns fall short as a luser disposal tool really quickly as the density or number of lusers rises. Hence, explosives. You should really be prepared to deal with hordes of lusers, as a future-proofing technique, if nothign else.

Jacob

@HopelessN00b I don't dispose of anybody.

Bob

@Jacob Right. wink nudge

HopelessN00b

@ewwhite Alright, so what's the advantage of RDX drives, over, say, just using a bunch of external hard drives? (Or USB keys/SD cards, for that matter.)

Bob

@HopelessN00b Supposedly, they survive falls better? :P

Much bigger than USB flash/SD.

Nathan C

17:31

I use these: wdc.com/en/products/products.aspx?id=600

HopelessN00b

@Bob <shrug> I've picked up flash keys as big as the largest SSD model they offer.

Bob

@HopelessN00b We're using 1.5 TB RDX drives. For the same price, you'd get... 256 GB on flash?

It was difficult enough getting enough to cover a week's worth of daily rotations.

ewwhite

@HopelessN00b ruggedized.

I'm still dealing with asia-exotics.com and not being able to get email to my clients.

hiss

Chris S

@MattBear Well... didn't see that one coming.

Dennis Kaarsemaker

17:51

@ewwhite that domain could be a brazzers subsidiary or a produce company. Either way, I'm not clicking.

Jacob

@DennisKaarsemaker It doesn't work.

Nathan C

@ewwhite Maybe the NXDOMAIN?

ewwhite

@DennisKaarsemaker Exactly... It's a produce company in Bangkok.

Nathan C

# nslookup asia-exotics.com 8.8.4.4
Server: 8.8.4.4
Address: 8.8.4.4#53

** server can't find asia-exotics.com: NXDOMAIN

Dennis Kaarsemaker

no MX records either

not even registered, says whois

ewwhite

17:57

oops, no "s"

From: Mail Delivery Subsystem [mailto:[email protected]]
Sent: Tuesday, March 25, 2014 1:09 PM
To: [email protected]
Subject: Delivery Status Notification (Failure)

Jacob

@ewwhite Nice website.

Dennis Kaarsemaker

no A record for the domain, some mailservers don't like that

hm

ok, rerunning dig does give an a record

Transcript for