finding and breaking external excel links

Apart from the great tips all over the internet such as this one

http://tjshome.com/blog/249/guide/excel/accounting

I find very common when using conditional formatting and copy/pasting from one sheet to another that this link gets created with a reference to the original file. Break links doesn't really break them. Find them by clicking on

Home ribbon, Conditional Formatting, Manage Rules

change the "Show formatting rules for:" drop down to "This worksheet" and look for anything referencing a file instead of a rule, like in this case:

Change the formatting and you are set.

Adding Dell's and HP's download source URLs to VUM to get updates automatically

This is a short and sweet one. By default VMware adds some URLs to VUM: 

Here is the URL to add Dell (very useful if you install the OMSA vib, for example):

http://vmwaredepot.dell.com/index.xml

Don't forget to click on Apply and Download Now if you are in a hurry :)

Check the patch repository, sort by vendor and you will see new vib packages. I do believe it only shows packages applicable to your hardware

HP's is http://vibsdepot.hp.com/index.xml

some gotchas with drac, java and firmware updates

RAC0218: The maximum number of user sessions is reached

this happens when several people have not logged out their sessions - or may have forgotten about them. The obvious answer is to do it from the OS and Open Manage Server Administrator (OMSA) - but what if that is somehow unavailable (esxi, for example). Two levels of solutions without resetting the drac settings:

1) if you are lucky, establishing a SSH session will allow you to login. issue the drac soft reset command to regain web access (racadm racreset soft, press enter, wait for pings to go away and come back)

2) if the SSH session tells you it's also out of sessions (in a message similar to " Server sent disconnect message type 11 " you can still reset the drac without power cycling the server, but need physical access to the server. Hold the i button on the back of the server approximately 20 seconds (until the drac stops pinging). This is a hardware initiated drac soft reset.

***

Upgrading drac firmware fails at 30% (seems more common on generation 11 dell servers (r710, r610) and early versions of firmware)

What solved it for me was finding a server that still ran internet explorer 6. You have to remember that new browsers were not supported until X moment of time and firmware revision, so always test with browsers available at the time the firmware was released, or different browsers. A friend of mine swears firefox always works.

***

Java, you pain in the butt.

The best version of java for accessing Dell stuff (compellent support actually told me this tip) is version 7 update 45, both x86 and x64 versions. It still prompts if you want to upgrade java or do it later, and still runs everything.

If you don't like this solution, and want to use the latest Java, you can do two things:

1) add the drac url to the java trusted sites (Control Panel, Java, Security tab). this is probably the best way to go.
2) lower Java security to the minimum from Control Panel, Java, Security tab

***

After upgrading the drac's firmware, you get

JAR resources in JNLP file are not signed by the same certificate

The solution to this is to open Java in Control Panel and delete the Temporary Internet files :p


I will add as I find from my own experience.

Reset domain administrator password in directory services restore mode

Ran into this when I booted my lab DC, which I hadn't used often. It wouldn't take my domain administrator password! I remembered this was a brand new DC and I hadn't created any other domain admin passwords - and I just didn't want to spend the time rebuilding it.

Luckily, after rebooting the VM, pressing F8, and going into Directory Services Restore Mode  (DSRM), my pw did work there (do remember that it boots into a login screen where you now can use a local administrator, so change the user to .\administrator ). The only problem is I didn't know what to do next - how to reset the domain\administrator pw.

It took me a while to find it in Google, but here it is. Marc Weisel did an excellent post.

Head over there and read it all - but if you want the summary, here it is:

You basically create a service that will reset the pw on the next boot. The pw has to comply with your domain policy, and you must also change the user if you know you renamed it.

The commands are:

1) Create a service to reset the pw on next boot
2) reboot
3) delete the service

sc create ResetPW binPath= "%ComSpec% /k net user administrator PA$$w0rd94" start= auto

sc delete ResetPW

vcsa 5.5 with domain authentication

If you have joined the vcsa to a domain in the authentication tab of your vcsa administrative console, you do not need to add a STS SPN; you can use the machine account to add the identity source of that same domain.

What you do need is to go to your AD DNS and make sure the forward and reverse entries for the new VCSA and your AD have been created, and your VCSA has the proper DNS settings.

pics will come, but I have seen google lead me to believe I have to use a SPN, or that I have to add it as a AD LDAP source. None of these are needed if you already joined the VCSA to the domain; you could use them for additional domains.

Nested ESXi in a physical ESXi quick sheet

First of all, hardware has to meet some minimum requirements

- your CPU should have VT (google the CPU, see the advanced technologies at the bottom, example: http://ark.intel.com/products/52214)
- enable CPU VT option in BIOS (if you just did a BIOS update, check it again)
- this machine should have enough RAM for the physical ESXi and the nested ESXi (especially if you want to run more than one nested host to test all features of vcenter) and the VMs. I'd say 16GB is conservative, but definitely doable.

Recommendations so the VMs perform well
- SSD (if you have both SSD and spinning, move VMs you aren't using as much to the slower disk)
- at least 1 cabled gigabit
- plenty of ram

if the physical host is 5.0 and you are deploying nested 5.5

- choose rhel 5 x64 bit as OS when creating the nested ESXi VM
- I honestly only put a 1GB disk. if you need the logs, you can send them elsewhere using syslog.
- I choose vmxnet3 - intel should work too
- make sure you give it more ram (the default for rhel 5 is 1GB and the installer would fail)
- add the string vhv.allow = "TRUE"  to /etc/vmware/config in your Physical ESXi 5.0 host (with vi through SSH, for example)

F5 replacement useful commands

some F5 commands - v11.

-Shutting down that F5 box:

plagiarized from here , SSH and run

Actually if you want them to shutdown then power off use the following.

shutdown -hP now

h means halt.
P means power off.

- Restore another's device full configuration (even all network and mgmt ip address, all vlans, virtual servers, etc) without running into licensing issues (useful for RMA or DR). This is the condensed form of this article. I assume you can access the new device through web/ssh by the mgmt IP.

1. Through SSH/SCP, copy the .ucs file you want to restore (default is /var/local/ucs)
2. leave a SSH session open to the new mgmt IP
3. turn off the other device (see above)
4. in the SSH session, run tmsh and then load /sys ucs <path/to/UCS_file> no-license. You can "complete with tab" when putting the filename to make double sure.
5. once the file is processed, your old mgmt IP will begin responding, and your new mgmt IP will go offline
6. change cables from old to new device
7. check. Sometimes a vip needs to be disabled and enabled for stuff to work (arp?)
8. think highly of yourself

- Clear a device to factory settings (except mgmt IP)

I assume you figured out how to get the mgmt IP setup right. Now go here . Really. It's short and sweet.