what did you learn today? (part 2)

[Rick25]

SandyTech[/url]":1ng2e2t3] Sysadmining while sleep deprived, never a good idea.

So very true. If you feel a disturbance in the Force before clicking Apply, back way and come back at it after some sleep

 

[Rick25]

xcrunner529[/url]":a6t7ptgw]Am I missing something or can you really only do 64 days of daily backups to disk with DPM? That's absurd. The goal was to eliminate tape...

At the bottom of http://technet.microsoft.com/en-us/libr ... 57886.aspx 448 days for disk. Hence why we're not using the license and sticking to Commvault. That and our tape library isn't apparently supported.

 

[Rick25]

That some vendors are scared of a 3-phase PDU putting out 208v to the equipment despite the equipment clearing being marked that it'll run from 110-240v.....so much for Universal AC FML

 

[Rick25]

If you move the circuit you'll never have a problem. Leave it where it is and it's almost guaranteed that Murphy shows up and it somehow fries the whole panel.

 

[Rick25]

llib[/url]":26hy5d59]
Yep - we run about that hot in the summer, also. And Rick, the idjits probably don't know that almost all 3-phase PDUs provide single-phase 208v on the C13 or C19 outlets. They're probably thinking "Three-phase!?! OMG!! We'll be all burnin' up an' shit!" (Well, actually it's phase-to-phase, but they still call it single phase...)

Thanks llib for pointing that out. I may need to do a better job of clarifying to them that it's single phase 208 output from the APC PDU in the rack. You're probably right where they're hearing 3-phase and 208 in one breath and not realizing that not "3-phase 208" into their gear.

 

[Rick25]

I've also learned to pull the NIC (physical or virtual) on the box instead of shutting it down. That way you can't get bit by the "I know you're about to get rid of me so Murphy and I agreed I don't have to restart" bug. Less an issue on the virtual side but hey apply the same rule to both.

 

[Rick25]

That 720TB (Raw) in 12U of space and 30A(5 on each PDU phase x 2PDU) is really cool.

 

[Rick25]

Dell MD3260 with two expansion chassis.

 

[Rick25]

That I can drop a UPS to install a new subfeed (FYI have it done in the factory, way easier) and keep everything up....ok one thing dropped cause apparently I mucked up tracing that cable.

 

[Rick25]

Here's a good article from MS http://blogs.technet.com/b/askds/archiv ... etion.aspx

In the normal scheme of things you'll never run out, the MS domain at the time of that article only had handed out 8 million.
It's more a FYI and the hotfix to prevent sloppy applications from catching you unaware.

Even better is a way to find out exactly where you're at

dcdiag.exe /s:dcname /test:ridmanager /v

 

[Rick25]

afidel[/url]":1bvtfmbl]And the official email came out from HP today about the access changes, time to figure out who I'm going to move my business to. Lenovo might be interesting with the IBM purchase, will be watching how the Dell moves impact enterprise support as well.

Linkage?

 

[Rick25]

sryan2k1[/url]":1cjfnf23]

internalhostname1notregistered.domain.tld
internalhostname2notregistered.domain.tld

There is nothing wrong with that.

Yep, as long as your not trying to use IP or servername.local. Those are what they're cracking down one.


TIL how to deal with Exchange 2010 and quarantined mailboxes. Powershell would have been nice but we tracked it down.
http://techjournal.318.com/microsoft-ex ... -exchange/

 

[Rick25]

As long as it's a FQDN for a TLD that you own you're good to put anything in front of it.

Here's the blurb in details

Background

The CA/B Forum is a collaborative effort between Certificate Authorities (companies like DigiCert® that issue publicly-trusted certificates) and web browsers (companies like Mozilla or Microsoft that facilitate secure connections).

Because of these new requirements, Certificate Authorities (CAs) must immediately begin to phase out the issuance of SSL Certificates for internal server names or reserved IP addresses and eliminate (revoke) any certificates containing internal names by October 2016. In addition, the baseline requirements prevent CAs from issuing internal name certificates that expire after November 1, 2015. After 2015 it will be impossible to obtain a publicly-trusted certificate for any host name that cannot be externally verified.

These baseline requirements are also being incorporated into global auditing standards. They were included in the WebTrust and ETSI auditing standards for CAs on Jan 1, 2013. Once the requirements are adopted, browsers will require certification from auditors that a CA meets the baseline requirements prior to renewing their root certificate.

What is an Internal Name?

An internal name is a domain or IP address that is part of a private network. Common examples of internal names are:

Any server name with a non-public domain name suffix. For example, www.contoso.local or server1.contoso.internal.
NetBIOS names or short hostnames, anything without a public domain. For example, Web1, ExchCAS1, or Frodo.
Any IPv4 address in the RFC 1918 range.
Any IPv6 address in the RFC 4193 range.

 

[Rick25]

YIL that RAID 0 and Raid 1 are terms are there to help you identify the hard drives in a Raid set aka either drive 0 or 1....I had to read the email from a support desk twice....vendors

 

[Rick25]

Go with cry...or at least softly pound head on the desk. They weren't reusing the terms for a product. The desktop does have a RAID1 config and it lost a drive. When we asked for the replacement we got an "explaination" about the "differences" between the two.....please just ship me the drive.

 

[Rick25]

TIL that VMware View Capacity Planner seems to be GA for everyone, not just restricted to Partners anymore

http://www.vmware.com/ca/en/products/view-planner

 

[Rick25]

TIL not to touch servers on Friday.

RDP to a server to get an IP address for a AnywhereUSB console.
Launch the config utility and it comes up fine
Launch the viewer and the server reboots....and doesn't come back
Sits at the splash screen, tried last know good with the same response
When I do Safe Mode it's freezing at ACPITABL.DAT

Mounted the drive on another server and currently running chkdsk against it.

No idea what caused it and the two VM backups that I restored boot the exact same way.

 

[Rick25]

Thanks, gave that a shot after the chkdsk didn't fix it. Still showing the same behavior. It was a 1 vCPU and as far as I know no one has touch it. It was worth a shot.

[edit]

Got it fixed.

Booted off the 2003 install disc and picked Recovery console. System spent some time looking at the disk. When I got to the
point where I had an option to log into a Windows instance I hit Enter thinking it would default to the only instance. Nope, that was for exit.

On the next reboot I took a chance and didn't boot from the CD, Windows happily loaded and everything is good. Rebooted the system normally and
it came back up. Took a snapshot just in case, hopefully this working version will get backed up properly over the weekend.

 

[Rick25]

On the Boot tab in msconfig (does it exist in Windows 2008?). There's two options, Boot log and OS boot information.

I'm betting the second one causing the files to show on the screen during a normal boot they way they do if you're booting into Safe Mode.

 

[Rick25]

PaveHawk-[/url]":1uws2ntn]TIL, keeping it simple works out better.

After much soul searching and pondering about that other role, I upped and went and got another job somewhere else. Signatures/paperwork to be done this week, then I start on the 7th of April.

Its been 4 weeks since I was let go, and I'm easily at my most relaxed state, something I havent seen for 10+ years. Problem is, my brain is so relaxed that its actually hard to study for VCAP5-DCD.

Congrats!

 

[Rick25]

M. Jones[/url]":1c88nqo8]

sryan2k1[/url]":1c88nqo8]
In other news, I forgot to enable jumbo frames on a new 3750x stack, and was pulling my hair out trying to figure out why this one VM wouldn't mount any of it's disks located on a Synology NFS share.

I recommend against jumbo frames. If you choose to go that route, look forward to more of this.

We've run into something like this lately with ESXi 5.5 hosts and Promise storage. The first Host mounts the iSCSI volume fine and can create the datastore. Second host can't. Digging into the logs it shows that the Host is seeing the Datastore as a snapshot and so it doesn't mount. If you drop Jumbo Frames down to 1500 it identifies it correctly and happily mounts. Based on the support call it seems this is a known issue and they just haven't gotten around to patching it yet.

 

[Rick25]

That the hardware requirements for Commvault 10 media agents are crazy expensive.

 

[Rick25]

sryan2k1[/url]":3pfl6ll1]

Rick25[/url]":3pfl6ll1]That the hardware requirements for Commvault 10 media agents are crazy expensive.

Hm? We're on 9 and going to upgrade to 10 soon.

Sent you a PM

 

[Rick25]

TIL Qlogic is still shipping fibre switches that require Java 1.4.2 for their GUI interface.....

 

[Rick25]

euri[/url]":3rc3s2it]TIL I passed the VCAP-DTA! Yay!

Now the only thing I have left on my to-do list for this year is VCDX-DCV submission (and defense if accepted). Welp.

Congrats! Nicely done.

 

[Rick25]

VMworld registration is open

I'm a little ticked though, I'm one of the first 400 people to register and all of the hotels close to the Moscone Center weren't showing as available when I got to the hotel booking.

 

[Rick25]

Scarily I've seen that in real life. For some reason the SCCM OS deployment categorized all the devices as Unknown and started to push out OS installs. We lost about 3 servers (nothing critical thankfully) and a whole bunch of desktops. Very happy that I was at my desk when a friend called to ask "Hey what's this screen mean"....and so was the guy who was working on the OS deployment tasks in SCCM. We use MDT now.

 

[Rick25]

PaveHawk-[/url]":1unk5amw]
I studied, sat and passed VCAP5-DCD.

Congrats, that's awesome news!

 

[Rick25]

It's always a little funny (and sad) when you see a job posting and they are requiring more years of experience on a product then the product has been released.

 

[Rick25]

andyfatbastard[/url]":e6wqo0zi]Anyone here ever use Atlassian Jira as a helpdesk/bug tracking system? Is it any good? Any obvious pitfalls or big advantages to it?

Unless they changed it, I believe end users (not just helpdesk agents) count as a user.

 

[Rick25]

TIL that consolidating a 250GB snapshot into the parent VM takes a very long time. 29% after 2hrs.

 

[Rick25]

AngelZero[/url]":2qszsvkj]

Rick25[/url]":2qszsvkj]TIL that consolidating a 250GB snapshot into the parent VM takes a very long time. 29% after 2hrs.

Used to make alarms in client sites: warn whenever a snapshot has been around longer than 30 mins, alert at 45. How's the application doing during the consolidation?

Thankfully it's a Proxy server, we just took it out of the DNS round robin yesterday and it'll go back in once it's complete.

 

[Rick25]

Good trick. This was a bit of an outlier cause it was snapped prior to the last round of Windows Updates. Person didn't know that it was a) an appliance b) based on linux

 

[Rick25]

I had one where an application on the client site was having problems connecting. They "proved" it was the network by submitting a ticket that had a screenshot of the error that was taken from the HQ side through their active RDP session.

 

[Rick25]

ferzerp[/url]":fwh1hitt]So far I'm 2/2 on VMware exams with no prep (other than the mandatory class). VCP-DCV and VCAP-DCA so far. Next up is VCAP-DCD I guess. The certs were never important with my organization, so I never bothered, but some things have changed so I figured it was time. The VCAP was scored within a 4 hours of completion despite saying it would take up to 15 business days.

Congrats!

 

[Rick25]

VMworld Schedule Builder is live!

 

[Rick25]

euri[/url]":2t3ias7v]

Rick25[/url]":2t3ias7v]VMworld Schedule Builder is live!

RUB IT IN WHY DON'T YOU

Sorry! It's ironic that now that you're part of VMware you can't go.

 

[Rick25]

daldrich[/url]":47agavxj]We are in the process of moving users home folders to a new server, which means changing their mapped drive. Apparently whoever set their AD Profile up original used \\servername\users\username instead of \\servername\users\%username%. Which means I have to change each one individually. sigh.

Would it break anything if I did just change it to the variable?

%username% gets flipped to username when you hit apply.

 

[Rick25]

24hr day - check
Fiber cut - check
2 48 Gig port cards failed (in same chassis at same time) - check

 

[Rick25]

TIL that if you create a USB install disk of the OS X Mavericks install it won't work by default.

You can boot from it and it starts but then gives you an error about the installer can't be verified and may be corrupted.
I tried creating it three different ways and got the same result. At that point I failed over to Google.

Apparently the date isn't right by default and that's causing the issue. So booted off the USB, instead of doing "Install OS X" I launched Terminal. Issued the "date" command and was told it was Jan 2001. Fixed the date and launched the Installer, works like a charm.