VMware Communities : All Content - All Communities

Login is black on Virtual Machine

tom howarth — Fri, 13 Nov 2009 06:41:44 GMT

Matthew

have a read of my blog post on this very self same issue "The Black Screen of Death"

If you found this or any other answer useful please consider the use of the Helpful or correct buttons to award points

Tom Howarth VCP / vExpert
VMware Communities User Moderator
Blog: www.planetvm.net
Contributing author on "VMware vSphere and Virtual Infrastructure Security: Securing ESX and the Virtual Environment”. Currently available on roughcuts

How to tell if your host is 64 bit compatible without a reboot?

jamieorth — Mon, 09 Nov 2009 19:52:21 GMT

Hey Matt -

Just curious, whats the bug that is stopping you from changing the disc size in the 1st place??

Regards...

Jamie
If you found this information useful, please consider awarding points for "Correct" or "Helpful".
Remember, if it's not one thing, it's your mother...

Adding existing formatted luns to a new host

juchestyle — Mon, 26 Oct 2009 17:47:06 GMT

I didn't think I was crazy but it always helps to get a second opionion. I went back to the storage team, they promised everything was right, they checked again and realized they didn't mask the host correctly.

So, they won't be seen yet.

Thank you,

Matthew

Kaizen!

What was the build date of my Virtual machines?

Troy Clavell — Tue, 20 Oct 2009 17:26:20 GMT

something like
http://communities.vmware.com/thread/207709

or
http://www.virtu-al.net/2009/08/18/powercli-daily-report-v2/
http://www.virtu-al.net/2009/08/27/powercli-one-liners-last-10-vms-created-and-removed/

Upsizeing disk in a 2008 Windows server created an unusual issue

VMmatty — Mon, 19 Oct 2009 15:45:14 GMT

Do you have another Windows 2008 server handy that you can test on? I am 99% sure I have done the exact process you describe below using ESX 3.5 and have been successful. If you have the ability I would try testing it again on a dev/test server to figure out what went wrong.

The other option is to use ExtPart from Dell. I've had success from that as well.

http://blogs.kraftkennedy.com/index.php/2009/06/30/expand-virtual-machine-boot-volumes-with-no-downtime/

Business Objects 3.1 Enterprise Server

sgunelius — Tue, 25 Aug 2009 18:35:51 GMT

Matthew,

I'm checking with some old Pfizer colleagues of mine on this one. I'm pretty sure we used Business Objects as a component in our Documentum environment and all of the development & QA servers were VMs.

Cheers,

Scott

VMs not getting enough assigned memory

juchestyle — Tue, 25 Aug 2009 17:49:06 GMT

Hey Guys,

I am leaning towards this being an application issue as well. The only reason it might be an infrastructure issue is because when the linux guy issued the

Free -tm command, the server returned the following:

Total = 2027

Used = 571

Free = 1455

Swap = 1027

Used = 0

Free = 1026

So it looks like the VM is only capable of using 512 megs of ram during this issue.

Question, if the -xms and -xmx parameters were not set correctly, would it keep the VM from using all of its available memory? Am I wrong to think that the VM should have been able to use more memory?

Let me know.

Thank you,

Matthew

Kaizen!

Access denied when trying to connect client cd rom to vm

Joshua Mally — Fri, 14 Aug 2009 13:38:48 GMT

Check this KB

http://kb.vmware.com/selfservice/microsites/search.do?cmd=displayKC&docType=kc&externalId=1004563&sliceId=1&docTypeID=DT_KB_1_1&dialogID=33778099&stateId=0%200%2030754424

If you found this information useful, please consider awarding points for "Correct" or "Helpful". Thanks!!!

-Josh
Trying to learn

How to track changes

MHAV — Sat, 01 Aug 2009 00:22:41 GMT

vSphere brings along the Feature Host Profiles that will help you to track changes done to the ESX-Server.

Regards

Michael Haverbeck

If you find this information useful, please award points for "correct" or "helpful".

DRS Rules not working?

msemon1 — Thu, 30 Jul 2009 15:06:05 GMT

Seems to me there was a problem when creating a lot of rules for DRS. VMware warns not to create too many because of the overhead on Virtual Center. Not sure what the magic number is, however, what you are saying makes sense because you restarted VC and the DRS rules began to work.

Mike

Best or fastest way to get files onto an ESX

azn2kew — Thu, 30 Jul 2009 14:22:03 GMT

Matt,

Doesn't these has great responses to good tools you can use? Answer with thanks note to acknowledge would be appropriate chief!

If you found this information useful, please consider awarding points for "Correct" or "Helpful". Thanks!!!

Regards,

Stefan Nguyen
VMware vExpert 2009
iGeek Systems Inc.
VMware, Citrix, Microsoft Consultant

Settings on VMs change on their own?

azn2kew — Wed, 29 Jul 2009 17:46:07 GMT

I haven't see any of this situation, but to get better ideas who manage or configure changes to your environment, I would look at custom users/roles permission so you can grant them accordingly, and after that make sure you have good logs management where it can ship your logs to a remote log servers like kiwi or something of your choice if not pay for a commercial log management utilities like LogLogic or HyTrust appliance which is pretty good with this part.

Have you check on tasks & event logs to see when its being altered by someone else any chance. Where is your data center guru should have your AC/ and Power redundant all the way, so how you live with 70 VMs shutdown and restarted?

If you found this information useful, please consider awarding points for "Correct" or "Helpful". Thanks!!!

Regards,

Stefan Nguyen
VMware vExpert 2009
iGeek Systems Inc.
VMware, Citrix, Microsoft Consultant

the attempted operation cannot be performed in the current state powered off

azn2kew — Tue, 14 Jul 2009 18:16:27 GMT

Looking at Troy's KB, it seems to be worth looking at it and hopefully you would have better storage management and systems monitoring in the future. If you storage if low on disk space and happend to perform snapshot, it will corrupt your VM and the LUNs itself also influence whatever guests running within that LUN. It would be best practices to have at least 10% of disk space on each LUN for this purpose. I would plan it accordingly or provision new LUN and hopefully you can migrate it and re-balance out.

If you found this information useful, please consider awarding points for "Correct" or "Helpful". Thanks!!!

Regards,

Stefan Nguyen
VMware vExpert 2009
iGeek Systems Inc.
VMware, Citrix, Microsoft Consultant

Vmotion hangs at 10%

juchestyle — Wed, 08 Jul 2009 20:46:06 GMT

I usually go through as much troubleshooting on my own before I post to the forums. My ESX hosts all run at the max configuration settings of 800megs. Glad that it helped you, that is why I love these forums!

Respectfully,

Kaizen!

Quick Recap of May 28th VMUG

jamieorth — Tue, 16 Jun 2009 14:16:46 GMT

The 7pm start time was a no go for me - I don't know how long the meeting went but I would have been home late after the drive.. I know it is difficult for some when the meetings are durring business hours. I don't know if my current employer would allow me to leave or not - in the past job it was a non issue. I personally like the after work time, but right after work, say a 5:30 start time and have a vendor lined up for some type of food.... then you get out at a decent hour....

Did the Orlando group ever do a Saturday??

Regards...

Jamie
If you found this information useful, please consider awarding points for "Correct" or "Helpful".
Remember, if it's not one thing, it's your mother...

what version allows hot cloning?

norregaard — Wed, 03 Jun 2009 21:29:46 GMT

and you need both VC2.5 U2 and ESX3.5 U2

Extreme Slowness on xp Desktop vm

jamieorth — Sat, 09 May 2009 14:27:55 GMT

Really Matt - its windows!!! (ok, just kidding)

What I would check:

Windows Updates downloading in background
Antivirus
Check the event logs - application and system
Check task manager
reboot

Regards...

Jamie
If you found this information useful, please consider awarding points for "Correct" or "Helpful".
Remember, if it's not one thing, it's your mother...

Host Disconnects

Mikeluff — Fri, 08 May 2009 15:30:57 GMT

Question for you - are you using HP hardware?

Hard drives on Linux going into read only mode

kjb007 — Thu, 07 May 2009 01:07:02 GMT

I saw that article earlier as well, but he's using RHEL 4 Update 7. The kb article says the issue was fixed in Update 5.

-KjB
VMware vExpert

Upcoming Storage Virtualization - Steak and Storage

jamieorth — Fri, 01 May 2009 13:34:15 GMT

Anybody get a chance to go to this?

Regards...

Jamie
If you found this information useful, please consider awarding points for "Correct" or "Helpful".
Remember, if it's not one thing, it's your mother...

VMUG Orlando May 28th 7pm

juchestyle — Wed, 22 Apr 2009 14:44:32 GMT

7pm start time.

There was a request for a later time rather than an earlier time. We decided to give it a try. If this doesn't work well we will go back to an earlier time. Also, parking is better for this location. Please do your best to come out, we promise this will be a worthwile event.

Matthew

Kaizen!

Next VMUG meeting!

juchestyle — Wed, 22 Apr 2009 14:24:09 GMT

The next Meeting is May 28th at 7pm, there is a new thread that details the time, place etc.!

Matthew

Kaizen!

Increase the /opt partition on the fly

harryc — Fri, 17 Apr 2009 20:24:47 GMT

You probably only want to increase space for a few directories under /opt , you can always cut a new partition, or present one from a SAN or NAS, and mount it at that directory.

e.g. /opt/interwoven has filled up the partition that opt is on.

1) mv /opt/interwoven /opt/interwoven.old

2) mkdir /opt/interwoven

3) present a new device (need to run mkfs if not an NFS share)

4) vi /etc/fstab

/dev/sda8 /opt/interwoven ext3 defaults 1 2

5) mount /opt/interwoven

6) cd /opt/interwoven

7) move files back in whichever manner you feel most confortable with.

Caculator

glynnd1 — Fri, 17 Apr 2009 16:11:59 GMT

I've used the Dell Data Center Capacity Planner (www.dell.com/calc) tool to do this in the past - obviously this is only going to give you data on Dell equipment. It is quite useful for seeing difference in power consumed depending on server configuration, eg 2.5" Vs 3.5" HDD.
However the power consumed is going to differ greatly depending upon CPU load, which depending on what your servers do may differ from day to day or month to month.
I would not try and get this 100% correct, but settle for a good guesstimate.

How do we get new vdi's to auto join AD?

Troy Clavell — Thu, 09 Apr 2009 18:48:22 GMT

you create your customization specs within vCenter. When you create a pool in view, one of the questions that will be asked is what template to use and what customization spec. As long as you have all your sysprep files in the right location vCenter takes over in the deployment process

Customization Operation Timed out

juchestyle — Wed, 08 Apr 2009 20:00:35 GMT

Ok, it seems there was another post that mentioned that the network settings were probably not updating. In my case the network settings were not checked. Once I checked them everything worked like a champ!

Matthew

Kaizen!

ESX Hosts not responding then coming back

azn2kew — Fri, 20 Mar 2009 19:59:39 GMT

There are series of steps you can check to resolve this issue so check out this kb from http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=1003409

If you found this information useful, please consider awarding points for "Correct" or "Helpful". Thanks!!!

Regards,

Stefan Nguyen
VMware vExpert 2009
iGeek Systems Inc.
VMware, Citrix, Microsoft Consultant

Add Permissions to see the data store?

juchestyle — Fri, 20 Mar 2009 18:40:43 GMT

Hey Mittell,

In this case I had to do something similar to your suggestion. I created a new role that had just datastore browse. I defined this at the datacenter level and I did not propagate it. Once I did that, everything worked like a champ!

Thank you for the help!

Matthew

Kaizen!

Windows 2008 at 4%???

RParker — Thu, 19 Mar 2009 21:47:19 GMT

Do you have some notes you could share on how you tweaked the 64 bit 2008

Well basically I found out that Windows 2008 has more services turned on by default than did Windows 2003. Like Shadow copy provider, SSDP, UPNP, Windows Audio, Protected Storage. There is a website (I don't remember where, it's been so long) that lists services for Windows XP, Vista, and 2003 that aren't needed for every day use (unless of course you have services that depend on them). I haven't seen one for Windows 2008 yet. So I just fiddled with the services until I found a 'sweet' spot. I looked at my services and tried to figure out which ones I turned off, but now I don't remember all of those that are on by default, so it's hard to say. But I remember thinking the same thing initially on the first Windows 2008 VM, and I was looking at and wondering why they had so many features and services turned on. So to streamline our server (which is also running Windows update services as well - forgot that earlier) I was determined to make it run as clean as possible.

Also on that task manager you can see network traffic and during those 2%-4% loads, maybe there is some services trying to locate things ( like broadcasting for UPNP devices and SSDP) which is a tip off that it's doing something that maybe you don't want it to. So finally I was able to turn most of the services off, and ONLY turn on the ones I needed one by one until it was bare minimums. It took me a whole weekend to work this out.

I had many problems which were attributed to inheriting old domain information, and I couldn't get Windows 2008 member servers to accept the Group Policy. There is a kb on this, but that didn't help. It turned out we had old objects still in AD that were referring to DC's that no longer existed, and as soon as they were deleted, everything worked perfect. I am still up in the air about Windows 2008, something about it just doesn't seem complete. Like when I login (I use the RDP security mode where you type your credentials at the time you connect to the machine -- this is before you actually get the login screen) it takes forever it seems to give me the desktop. Even if I disconnect and log back in, same thing, there is a small delay when Windows 2008 gives me the desktop. It's odd, Windows 2003 never did that. So I think there are still some underlying small things that were changed or updated that don't have the same polish that Windows 2003 has. Things like that.

So I wouldn't worry about things for Windows 2008 that don't really tell the whole story, but like another poster mentioned the overall speed (after the initial screen) even in a VM is amazing. It's very responsive, and despite my earlier posts about not running SQL in a VM and Dual CPU I did experiment (hey I am not above trying to get it to work!) and Windows 2008 does work better than the 2003 counterpart. And even benchmarks may not prove it's faster, but it does seem much more responsive for things. I rather like it, but still there are some compatibility things I run into every once in a while that reminds me 'oh yeah, THAT's why I didn't install in 2008 yet....'. So Windows 2008 still has a ways to go to prove itself. to me.

Anyway, that blog about Windows 2008 running better in VM than Hyper-V was part of my weekend pilgrimage to find the answer to life the universe and everything. And in passing I found a few musings and rants about people overall dissatisfaction with Hyper-V. Maybe saying 2008 ran better on ESX was a jab at MS, maybe there were VM Ware fans, I don't know but I saw it a few times in trying to figure out my problems with 2008 tweaks and settings. So It may not really be true, but I found it funny that people would comment on MS blog forums about Windows 2008 being buggy on Windows 2008 Hyper-V, that's why it caught my eye. But I am not sure that I can actually produce the proof.

No vms listed on host but vm shows on that host from cluster view

sbeaver — Thu, 19 Mar 2009 20:59:38 GMT

You are using the same plugin

When do you plan downtime for work inside your Clusterered Environment?

Texiwill — Wed, 11 Feb 2009 14:56:58 GMT

Hello,

If it is a clusterwide change (i.e. enabling EVC) we did it off hours.
If it is patching ESX, we do it during normal hours and do a rolling upgrade.
If it is a SAN upgrade, we do it off hours.

Even doing things off hours, the only time we had all nodes unavailable was during a total switch fabric upgrade and even then it was limited to the time it took to move cables.

Best regards,
Edward L. Haletky
VMware Communities User Moderator
====
Author of the book 'VMWare ESX Server in the Enterprise: Planning and Securing Virtualization Servers', Copyright 2008 Pearson Education.
Blue Gears and SearchVMware Pro Blogs -- Top Virtualization Security Links -- Virtualization Security Round Table Podcast

HA Error on host

prudhvi — Mon, 09 Feb 2009 12:15:29 GMT

Check the vpxa version on your host: + +rpm -qa |grep vpxa
That will give you the current version of vpxa that you're running. eg: VMware-vpxa-2.5.0-104215
Stop the VMware mgmt service: ++service vmware-mgmt sto
Stop the vpx agent: ++/etc/init.d/vmware-vpxa sto
The vpx agent error can be ignored (++warning: /etc/vmware/vpxa.cfg saved as /etc/vmware/vpxa.cfg.rpmsav)
Remember your vpx version from the first step and use it here to remove the vpx agent.
++rpm -e VMware-vpxa-2.5.0-10421
Switch over to your Virtual Center client and remove the host you just modified (guest vms will remain)
Reboot the host (vm's will go down)
After boot reconnect the host to VC and the latest vpx agent will be intstalled.
Enable HA and the error should disappear.

Best Practice way to remove a lun with a vm on it and reattach it to another stand alone esx?

juchestyle — Fri, 06 Feb 2009 12:41:20 GMT

Hey Donald,

Let me change your order of operations there just a bit.

Add the lun to the new stand alone esx FIRST, BEFORE removing it from the other ESX servers.

This will ensure that you can still work with the lun if the stand alone refuses to see the new lun.

Respectfully,
Matthew

Kaizen!

Sunquest copath application loosing connections on open port

juchestyle — Mon, 05 Jan 2009 20:11:05 GMT

Hey all,

We have a vm with sunquest copath application running on it. After an upgrade from 2.4 to version 4 there seems to be issues with a port, and loosing connections. After they created a new port they are having fewer issues.

I was curious if anyone else is running this healthcare application and if they ar having this issue. Our application person says that they know of a memory leak in the past, and I think that this issue is the result of upgrading, but I wanted to do due diligence!

Respectfully,

Matthew

Kaizen!

Size of your C drive?

Rockapot — Fri, 14 Nov 2008 10:25:59 GMT

Our VM's (Server 2003) have 10Gb allocated to the system disk.

Carl

Pros and Cons of Qumranet and Red Hat

juchestyle — Tue, 11 Nov 2008 19:55:41 GMT

Ok, I just got out of a meeting with these people. Their largest installation is a beta test of 100 users. They have no market share. They are not on Micro$oft's "Server Virtualization Validation Program." No way to handle bandwidth, not sure how their own software communicates with AD, no application management. They were able to tell us that their protocol is 15 to 20% higher than RDP without any graphics.

When I asked what their largest installation was, they actually said "Wow"

They claim however that they can our perform VMware on the number of vm's running on a host. I found that to be a bold statement considering half the stuff they said they would have to get back to us on.

Since they have ZERO, none, zilch, bagel, market share I see that this post was probably not reaching the 100 users in the world that are beta testing this product.

Ok, I am done.

Respectfully,

Matthew

Kaizen!

How to initiate a Tivoli TSM backup on a vm in the DMZ

juchestyle — Wed, 20 Aug 2008 16:28:40 GMT

Allright, this has more to do with what command line you have to issue on the TSM server, but our backup guy left and we have a new backup guy that doesn't know how to "Execute the schedule" from the TSM server side.

So, are there any Tivoli TSM admins out there that know how to execute the backup schedule when the VM server is "Waiting to be contacted by the server"

I have points to award!

Respectfully,

Matthew

Kaizen!

Windows, Linux, Mac?

ch1337 — Sat, 09 Aug 2008 18:13:35 GMT

Then there would be a possibility to run a guest Mac OS on Linux running on Apple hardware?

(not that I expect it, but I would like that very much)

Reducing Hard Drive Space?

tom howarth — Wed, 25 Jun 2008 16:02:04 GMT

The reply button was missing due to the location the thread was posted in. normal users have no rights to the Virtual Infrastructure roll up forum.

however to answer your question, use convert to convert the guest, during the initialisation process you can reduce the size of your disks, but you knew that already didn't you Matthew

Tom Howarth
VMware Communities User Moderator

Cisco 3120 10 Gig Switches

juchestyle — Tue, 24 Jun 2008 18:31:37 GMT

OK, so apparently it looks like there is a bug in the code that put the switches into a broken mode where they were sending traffic through the channeling port instead of the through the normal configuration. We are waiting to hear back from the analysts on this, and potentially with newer code. This stuff is so new that there aren't any upgrades and we can' t roll back, this is the only code so far.

Good Times!

Kaizen!

A Case Study - Defragging VMs (SQL, Blackberry, VC, Citrix, Applications, Exchange)

bradley4681 — Fri, 13 Jun 2008 03:17:51 GMT

So what was the result? I remember years ago reading some VMware documentation saying there was no need to defrag guests since they were in on large continuous file and because of the architecture of the VMFS

Cheers,
Bradley Sessions

If you found this or other information useful, please consider awarding points for "Correct" or "Helpful".

Virtualization Forum 2008 - Orlando

bradley4681 — Tue, 03 Jun 2008 20:34:52 GMT

Thanks and I really enjoyed the presentation and was very humbled by the fact that some of the items I implemented were still around. Can't wait to get together again, I was telling Steve we need drinks and some VM Nerd talk!

I also need to spend more time in the forums and get my points up to a more appropriate level for my skills, maybe one day I could actually present at VMWorld!

Cheers,
Bradley Sessions

If you found this or other information useful, please consider awarding points for "Correct" or "Helpful".

VMware Virtualization Forum - In Orlando - Thursday May 22nd

wonkawilly — Tue, 03 Jun 2008 17:58:08 GMT

Yes, I know about those...but the promotional emails about the vmware forum mentions that a jumpstart kit is provided with free software..I'm just wondering what was in this kit?

thanks.

How to use FTP to install ESX

azn2kew — Wed, 26 Mar 2008 14:45:22 GMT

Here is a steps by steps guide http://www2.altiris.com/ibmavenue/Portals/0/Installing%20ESX%203-0-1.pdf

If you found this information useful, please consider awarding points for "Correct" or "Helpful". Thanks!!!

Regards,

Stefan Nguyen
iGeek Systems LLC.
VMware, Citrix, Microsoft Consultant

Limititations of sql express?

ksram — Thu, 28 Feb 2008 11:21:50 GMT

http://communities.vmware.com/message/872051

The absence of SQL Server agent in this case is covered by VirtualCenter periodically taking care of statistics rollup process.

Moving Virtual Center from VM back to Physical

juchestyle — Thu, 21 Feb 2008 15:29:26 GMT

Hey all, the saga continues. Here is what we are working on currently and I think we have a better than average approach, but are still running into some issues.

1. Turn off HA on all clusters

2. Remove all hosts from old VC

3. Stop services on old VC

4. Stop services on new VC

5. Copy the database to the new sql instance - (remote verse lite version)

6. Change the ip settings in the new VC to reflect the new ip addresses - very important!

7. Start the services on new VC

8. Good luck - try to bring in your hosts

Will update with more info if I find it.

Respectfully,

Matthew

Kaizen!

Version Mismatch between VC 2.5 and ESX 3.02

juchestyle — Mon, 14 Jan 2008 20:20:11 GMT

Answered

Kaizen!

Error upgrading VI Client to 2.5

juchestyle — Wed, 09 Jan 2008 14:40:11 GMT

Allright everyone, my buddy Patricia and Chris found the answer to this problem and shared it with me so I could share with you!

Go to start and run, type %temp%

Keep the window open so you can see when a folder gets added.

Launch the upgrade msi - at some point it will error out, and you will see a folder get added to the window. When this happens, keep the error open, if you close it the folder will go away, open the folder and you willl see the msi executable, launch it directly from there or copy it to your desktop and launch it from there!

Oh yeah! Good times!

Respectfully,

Matthew

By the way, I bet this may work for other applications and not just for the vmware client install, so keep this little trick in your back pocket, it will come in handy in the future!

Kaizen!

Performance Monitoring

RParker — Tue, 13 Nov 2007 14:55:07 GMT

I tried that, but it doesn't seem to work, right out of the box. It requires a LOT of configuration.

adapter type: Flexible

Dave.Mishchenko — Thu, 11 Oct 2007 19:38:29 GMT

The flexible adapter is new to VI3 and is meant to be the replacement to the vlance and vmxnet adapters. See the below thread for more info on this.

http://communities.vmware.com/message/520435

Archiving a VM

ITThies — Thu, 11 Oct 2007 16:09:26 GMT

You can use VCB (or vRanger / ...) to backup and archive a virtual Machine.

vRanger (esxRanger) will automaticaly put them into a gzip archive if you want to. You can put that archiv onto tape or hdd for fong time archiving.

How do you determine if a Physical is ready for VM?

juchestyle — Thu, 27 Sep 2007 18:16:32 GMT

Yes I know! I was pretty excited, even shared it with the family!

1 word - ROCK STAR!

Let me check out these links

Need for Power, US builds X amount of new power plants

Neth66 — Thu, 06 Sep 2007 04:06:04 GMT

Last year at VMworld the when PG&E announced rebates to customers for virtualizing, they said that it's cheaper to give customers money for virtualizing than it is to build new power plants.

Check out this link

http://news.zdnet.co.uk/software/0,1000000121,39284631-1,00.htm

"Risser explained that it was cheaper for his company to reduce demand rather than build new generator capacity, and that increased energy efficiency was the most effective way to reduce emissions. He said that up to $1,350 (£708) per server was available in rebates, up to a cap of $4m per customer. This figure includes money for the servers themselves plus the drop in energy use due to reduced cooling requirements."

Message was edited by:
Neth66

Expanding a vmdk 32 bit vs 64 bit, using diskpart

juchestyle — Wed, 05 Sep 2007 15:58:04 GMT

Laughing, sure, wait until after I test everything to come back and tell me it will work!

Yup, a 32 bit helper vm can expand a 64 bit vmdk.

Respectfully,
Matthew

Encryption level?

desktop — Wed, 29 Aug 2007 15:29:24 GMT

We currently ship ACE with FIPS 140-2 lab-validated AES 128 bit encryption. The AES level is not configurable, but please let us know if there is a reason why that might be important to you.

The use of encryption itself, though, is configurable. We have several levels of application available including Full (AES 128bit), "tamper resistent", and no ecnryption. These settings can be configured per ACE Master. We recommend and default Full.

Balloon driver healing memory leak applications

juchestyle — Fri, 24 Aug 2007 18:01:00 GMT

Allright, so here is what I am thinking.

On a phyiscal machine open this link and see how long it take to break.

V2P that phyiscal, install tools, open link and see how long it takes to break.

Respectfully,
Matthew

Who bought Shares?

king@it.ibm.com — Tue, 14 Aug 2007 17:39:36 GMT

BTW... what is the name of the Italian stock market?

Not an expert on the matter but it should be S&P/MIB

Massimo.

Mom or New version in a vm?

juchestyle — Mon, 06 Aug 2007 20:42:51 GMT

How many servers do you have it managing?

Respectfully,
Matthew

Siemens Apogee - Operating room temp controller application

juchestyle — Mon, 23 Jul 2007 14:09:18 GMT

Boydd,

Thank you for the great advice, we will see if we can make this a reality.

Respectfully,
Matthew

Idea on how to make Forums even better

JohnTroyer — Fri, 29 Jun 2007 00:38:56 GMT

Got it. We'll put it on the list. But just to set expectations, we're trying to stick closer to our vendor's off-the-shelf offering so that we can upgrade more often. I'll bring it up with them, though -- they are used to getting lots of suggestions from us.

Board Warrior Shirts! - Oh Yeah

VMdawg — Wed, 30 May 2007 19:26:17 GMT

If I don't get mine by September, I'm stealing one from the VMstore at VMworld

How do you performance test?

Gabrie — Tue, 22 May 2007 06:38:54 GMT

Hi

I think the only way you can benchmark a VM to a physical system is to measure transaction times, which is quite difficult. Most systems deliver a service, like looking up a record in a database and presenting the results to another application.

Only real comparisson would be when you could execute some transactions from a users point of view on a physical server, time them, then run them on a VM and again time them.

Trouble is that to be able to do this the server normally has quite some connections to other systems and it is not always possible to copy your whole environment to a test lab. Should you run them in production environment, then you would have to watch out for keeping everything in sync. Also not an easy task.

Gabrie

Shares / Reservations / Limits: A case study?

Gabrie — Thu, 26 Apr 2007 10:52:24 GMT

At tsx event in Nice (FR) there was a presentation about cpu sheduling.

Now the presentation is online and even better, there is a great handout that goes with it to explain things in detail !!!

MUST READ !!!

http://www.vmware-tsx.com/index.php?page_id=10

Article: http://www.vmware-tsx.com/download.php?asset_id=36
Presentation: http://www.vmware-tsx.com/download.php?asset_id=39

Gabrie

Snapshot a Black Berry Enterprise Server

oreeh — Thu, 12 Apr 2007 16:35:37 GMT

we are now the proud parents of a happy smiling baby vm!

Are you the father, the mother or both?

Something Cool, Something Fun!

grasshopper — Mon, 26 Mar 2007 16:38:01 GMT

One thing both managers and techies can relate to in some way are BSoD's (blue screen of death).

Scenario:
- You're back in business

There are several ways to induce a Blue screen on demand, or cause other mayhem as you see fit. While this is a simple concept, it's one of the biggest benefits to everyone. No 2AM pages; no Sev One conference bridge; No outage.

When is the next meeting?

sgunelius — Mon, 26 Mar 2007 15:45:09 GMT

Haven't seen anything on the next meeting yet. Is everyone that busy with VI3 migrations/implementations or is everybody spinning up for the next VMworld?

Scott

Significant networking degradation on 100 Full Vswitch

grasshopper — Wed, 21 Mar 2007 19:28:34 GMT

Communication between two vms on the same host, as I am sure you
already know, shouldn't even leave the host.

On the same vSwitch, traffic stays internal. However, different vSwitches on same host, and traffic goes out to the Pswitches first.

If you have 2 vm's on the same ESX host, the transfer
rate should go through the roof.

Based on my first statement above, I'll interpret this to mean 2 VMs on the same vSwitch. I once assumed that performance would always be better in such a scenario as well. However I was shocked to find that performance _could_ actually be worse on the same vSwitch. It has been documented somewhere, but that was a lot of beers ago.

If I remember correctly, I think you should be using
1000 Full on the pswitches, not auto.

From what I understand, the IEEE standard calls for auto on both sides in Gb networks. However, I have observed 'quirks' required deviation from that standard under certain circumstances, so YMMV.

Back to the issue at hand though. While I don't have an answer for your network ailments, you may consider performing some pings tests to calculate your MTU and determine if you are fragmenting. Just google MTU or take a look here.

You might also consider running tcpdump on your host in addition to the network sniffs you are taking. Since you have an SR open, VMware can walk you through the setup for using tcpdump.

Idea for next meeting

KrazyD — Mon, 05 Mar 2007 20:56:20 GMT

I'm also new to this user group due to a new job and relocation. Was one of the few inital members of the Atlanta User Group and was an active participant.

I don't know what your past topics have been. So if I suggest something that has already been covered, just ignore it.

I gave a presentation back in Aug 2005 on Troubleshooting and Performance on 2x which the members at the time found to be beneficial. I also had prepared to present some of our "optimized" advanced settings that worked in our SAN environment, but I ran out of time.

At one meeting last year, we had a "round table" format which people could move around to different tables to hear experiences and ask questions from subject matter experts. Topics at the time that I can recall included ESX 3 upgrade experiences and gotchas, VM snapshots for DR, scripted deployments (host and vm's) using technologies such as Alteris.

A couple of other meetings have included virtualization futures (not just VMware). One meeting included a live demonstration of a ESX 3 upgrade.

Anyone presenting usually included specific details about their environment (number of hosts, number of virtual machines, and whether used just in dev/test or in production).

I come from a large production environment (160 hosts, up to 1300 guests at one point) and did a lot of scripting for the hosts and was trying to teach some of my former teammates before I left how to do some. For example, pushing out the Daylight Savings Time update for 2.5.2 and 2.5.3.

Again, just some ideas to get the juices flowing.

My water bottle broke

mikeskomal — Mon, 26 Feb 2007 21:44:54 GMT

You should wash your water bottle on occasion. Only the Lord knows what corporations pump thru their air ducts that eventually settle into your water. I think it's something that keeps us awake. Sometimes anyways.

When do resource Shares kick in?

juchestyle — Tue, 13 Feb 2007 21:41:48 GMT

Eh, you lucked out this time!

R,
MKJ

Rename a vmdk, the directory, to stay consistent with internal name?

Vliegenmepper — Fri, 09 Feb 2007 14:34:25 GMT

Take a look at the ESX Console Manager. With that you can easily change the filename of the .vmdk and the rest gets auto-adjusted.

edit: link http://www.esxguide.com/esx/

Performance Monitoring in 3.x like 2.5x

epping — Thu, 08 Feb 2007 08:39:19 GMT

i agree the performance graphs in VI3 are a step backwards compared to VC 1.4

lets hope VMware are watching this tread and do something about it, i really dont want to start looking at 3rd party tools.

Virtual Solution Box

kix1979 — Tue, 30 Jan 2007 19:55:59 GMT

I've played with it. It's got some cool features.

What is the default buslogic or lsilogic when creating a shared disk

juchestyle — Mon, 29 Jan 2007 19:39:16 GMT

While I know my sense of humor isn't cutting edge, I thought that it was coming through on some level?

I am sure this is a buried horse by now, but for anyone else that isn't clear.

The gui shows that you have a lsi logic; but it is not an lsi logic, it actually is a bus logic. This will be confirmed when you try to power on the vm. It will stop and ask you if you want a bus logic or if you want to switch it to lsi logic.

You feel me brother?

PS. It was actually longer than an hour, but I will settle!

Building a cluster - why Zero out disks?

juchestyle — Mon, 29 Jan 2007 15:49:50 GMT

Maybe we will get someone from VMware to chime in and shed some light?

Respectfully,

VC Corruption after Microsoft Updates

juchestyle — Fri, 19 Jan 2007 16:07:44 GMT

I upgraded the vmfs to 3 I just didn't change the size of the grub partition. I think something filled up and killed my esx. When I logged in with putty it told me there was an issue that didn't save the configuration on last shutdown, but I ignored the issue. Again, my curiosity was to see what would happen if I didn't follow the rules.

Respectfully,

BartPE / Ghost 8 / 3 Partitions on 1 drive

juchestyle — Thu, 18 Jan 2007 15:43:19 GMT

The only other thing that needed to be done is make sure the drives are in the correct order. The cdrom was in the second partitions spot and vice versa.

Thank you for all the help, you guys rock!

Respectfully,

Cold Migration - from 2 different local datastores to 1

kix1979 — Tue, 19 Dec 2006 22:36:01 GMT

Kix cannot read... You are correct, they can both get moved to one datastore. Default leaves them where they are, but you can tell it to go wherever on a cold migration.

Cold Migration taking forever - ???

kix1979 — Tue, 19 Dec 2006 22:29:46 GMT

I am trying to find one of those things myself, what
did you call it, oh yeah a wife. They are nifty!

I like the new dating service called VMworld 200X. Lots of oppertunities there

Are you running Kronos (time management app)

wcrahen — Mon, 11 Dec 2006 23:06:59 GMT

There is no reason they shouldn't, unless they are telling you that you need to run their app a specific hardware brand and model. You should be able to work with them and show them that it runs fine in a VM, which it does. If ever needed you may need to reproduce an issue on a physical server. Sometime they do require a modem on their communicaiton server, so this may not be a good fit.

VCP Questions?

juchestyle — Thu, 30 Nov 2006 18:16:46 GMT

Rockon,

I agree with you completely. There was a backstage conversation that spurned this on. Already going through the pdf's like a madman with a cause!

Regards,

Partition Magic: Can it resize a system partition?

juchestyle — Thu, 30 Nov 2006 16:19:30 GMT

Great information guys! I awarded points based of off good information, and specifying 2003, and the other tool called gparted. I didn't use gparted but would like to look it up and understand it better.

Again, thank you for your help, mission accomplished with EASE!

Respectfully,

How to tell if two cpu's are being used in RHEL

RobMokkink — Mon, 20 Nov 2006 07:23:48 GMT

do a cat /proc/cpuinfo
That will give you all the info you need.

Load balancing between two physical switches?

juchestyle — Thu, 19 Oct 2006 14:16:56 GMT

For the sake of understanding how things work. Lets just think of this as one ESX server. (a side curiosity would be if there is any difference in how 3.0 works verse 2.5x)

Inside this one ESX there are two virtual switches, but we are only concerned with one of those right now, ie the virtual switch that is designed for the vm's. There is another virtual switch and two nics for vmotion and the SC; but again, this thread is not aimed at that. I want to know how functionality works for the vm's.

There are 6 nics, and all of these 6 nics are in one virtual switch in that one ESX.

3 of the 6 nics go to one physical switch, and the other 3 nics go to a different physical switch. These two physical switches are designed for fault tolerance so that if they lose one, the other is there working.

My belief is that the 1 ESx will utilize all 6 nics on both physical switches for its inventory of 20 vm's that are running on that one ESX.

I don't believe that only 3 of those 6 nics will be utilized on only one of the physical switches; thus leaving the other 3 nics in standby, or not utilized.

Can anyone confirm, deny, or explain one of those belief's?

Respectfully,

Adding more local storage

acr — Tue, 17 Oct 2006 22:43:19 GMT

I know with 2.5.x there was no way of expanding the LUN, you just got extra space and used that.. thsese may help..

Of course you could consider, and many dont.. Spanning LUN's

http://www.vmware.com/community/thread.jspa?threadID=21119&start=0&tstart=0

http://www.vmware.com/community/thread.jspa?forumID=21&threadID=12629&messageID=126704#126704

http://www.vmware.com/community/thread.jspa?forumID=21&threadID=11162&messageID=107278#107278

http://kb.vmware.com/vmtnkb/search.do?cmd=displayKC&docType=kc&externalId=930&sliceId=SAL_Public

Hope they help, but i suspect you may already know them..

ESX Host disconnects when brought into vc2

ZippyDaMCT — Fri, 13 Oct 2006 20:24:21 GMT

have a look at this thread, it might offer a direction for you.

http://www.vmware.com/community/thread.jspa?threadID=48469

Regards

Zippster

Best way to import vm from vmfs2 to vmfs3

mittell — Fri, 13 Oct 2006 14:37:51 GMT

Add both hosts to a Virtual Center 2.0.1 server, then use the VI client to cold migrate (i.e. VM powered off) VMs between hosts and datastores. This is done over the network via the VC server so gets around the fac that the storage isn't visible to both hosts.

Console Redundancy

Besan — Wed, 11 Oct 2006 06:02:01 GMT

Hi
just to let you know.

VMware does not support nfs-mount in Service Console, support told me. The only nfs-mounts they do support are nfsmounted ESX Storage. The rest is a customer problem.

Dell Open Manage Problems?

servo — Mon, 09 Oct 2006 23:07:22 GMT

You are welcome and good luck.

Changing from 192 memory to 800 mb for the service console?

dtux101 — Thu, 05 Oct 2006 18:56:07 GMT

you can reassign the service console memory allocation using vmkpcidivy -i or from the MUI

vmkpcidivy -i option is easier - type the name of the boot u wish to change - normally esx (whatever option you select at lilo to boot the esx)

first option is to assign service console memory

hope this helps

Problems with the ladder!

jasonboche — Thu, 05 Oct 2006 02:24:11 GMT

(in Eric Cartman's voice)

"OK, that does it, I'm going home!"> QUESTION: On average, how many posts do you BIG BOYS
get to on an average day?

Not many during the summer. Fall/winter/spring is when I'm usually more active. Last winter it was not unusual to get a few hundred points a day. The quantity of posts doesn't raise the status much, it's the quality, since the quality posts with answers are what (should) get awarded with correct or helpful answers which are worth 10 and 6 points respectively.

Dual Core verse dual processor?

king@it.ibm.com — Sat, 30 Sep 2006 09:17:24 GMT

This is an interesting thread. And fortunately didn't get into the religious wars that I have seen in other posts .... which is good.

I guess that the answer to the base question is... it depends (easy answer). First it should be noticed that 4-socket systems support (typically) more memory slots and more PCI devices than 2-socket systems. So most of the time the customer is forced to use 4-socket boxes for this reasons. On the other hand other customers have choosen to go onto the 2-socket paths simply because of the VMware licensing schema (i.e. 4 cores at the price of 2......).

As per what is better 2 single core on 2 sockets or 1 dual-core on 1 socket ... well I guess it depends as I said. Assuming the same clock speed and the same amount of memory .... it all boils down to the workload you are trying to run. If you are "core bound" than it would be pretty much the same as you will rarely get outside the cores and being them on a single die or two won't make much of a difference.
When you are "memory bound" it's when it might become tricky. Dual-core processors typically share the memory connection (be it Intel FSB or AMD Hypertransport). So a 4-socket single core AMD system for example would have 4 HT memory paths while a 2-socket dual core AMD would have "only" two. So, assuming a strong and solid NUMA support in your software stack, ideally a 4-way single core AMD would outperform a 2-way dual core AMD system. Right ? Not always. There are circumstances where a dual-core system would outperform a single core AMD system even with memory bound applications: a 2-way VMware virtual machine for example would be hosted/contained on a single socket with local memory (which is good) while a 2-way VMware virtual machine would be hosted/distributed across two sockets with sparse memory (which is not as good).
I am not even getting into different cache policies such as discrete exclusive caches as in the current AMD opterons or shared cache designs as in the new Intel Woodcrests ....

So as you can see it might become very complex and very very dependent on the workload you are running. Most of the time we waste our time debating on processsor technologies while what you could do in software can potentially be order of magnitude better or worse than what you achieve on the silicon.

Clearly dual-core cpu's are going to cannibilize some of the higher-end market. No doubt about this and this is driven by the fact that, for half the licenses costs you achieve pretty much the level of performance that was available on single core systems (perhaps slightly less but yet you pay half the licenses !!! No brainer).
While I don't have any crystal ball .....on the other hand I would say that if you look backwords you will see that Intel/AMD made huge steps within the core itself to make the next generation processor twice as fast as the previous generation..... and I have never seen a discussion such as "what i used to do 3 years ago on a 4-CPU (single core) system I can do today on a 1-CPU (single core) system". Software improves and demands more powerful hardware. It happened to be more Mhz in the past it is now more cores. As VMware will introduce new functionalities such as fault tolerant hidden virtual machines (which requires to run on different physical hardware), scurity layers in the hypervisor etc etc and CPU vendors such as AMD will make easier for OEM's to create 8-socket and above systems ....... I guess that will be an interesting time to see ......

Well I guess I have digressed a bit ..... what was the question ?

Massimo.

The future of Virtualization (at the speed of thought)

DrAxeman — Sat, 30 Sep 2006 05:01:57 GMT

You're thinking of Softricity.

Alex Weeks

http://www.vi411.org/2006/07/19/ms-developoing-a-vmotion-competitor.html

Are you using ATA's

juchestyle — Tue, 26 Sep 2006 21:14:19 GMT

Oh, I completely agree. I don't think for a moment that the ATA is corrupting or failing, just doesn't make sense to me. Furthermore, we have 3 hosts on the ATA for over a year. To me that means that we have three years worth of experience on the ATA and this to my knowledge has only happened once. Therefore I don't think we need emergency changes to frantically move all vm's from the ATA to SAN storage. I agree, the problem had to do with a single point of failure, ie one path.

On a side note, we do have two paths on the SAN. I just don't agree with the reasoning of the decision, good decision yes, but the reasoning behind it is faulty to me.

Respectfully,

ESX 2.53 crashed

juchestyle — Wed, 20 Sep 2006 17:55:32 GMT

Dominic,

Any kind of test or command we can run to test our theory? We have opened up a ticket with vmware and ftp'd them the big dump file. I would like to be proactive on this as much as we can.

Respectfully,

Can not Terminal Service into a 2003 Server

Rumple — Mon, 18 Sep 2006 19:32:26 GMT

Its close, but not quite it.

In our case the server is running perfectly normal, but its as if the TS service hasn't started (its appears to be running, but hasn't created a network socket connection until after you apply the registry key or restart the server).

Everytime this happens a reboot always fixes it.

Unix support

juchestyle — Sat, 09 Sep 2006 13:17:52 GMT

Crap!

Documentation on Standards

rewh2oman — Wed, 30 Aug 2006 13:08:33 GMT

Hi Matthew

Don is correct in his thoughts. I have no form, I simply document as much as I can about an ESX Host.

such as:

*ESX hardware configuration, down to the motherboard slot population
*ESX build document and configuration; passwords
*VirtualCenter build document and configuration (including SQL or Oracle connectivity documentation)
*License documentation
*SAN configuration, paths, cabling, etc.
*Network configuration, topology, cabling, bonds, VLANs, etc.
*Guest VM inventory, including virtual hardware used, networks, memory, cpu, guest OS, default ESX host, static MAC requirements, VMotion policy, etc.
*In some companies, chargeback, cost centers, ownership, etc.
*Maintenance planning, standard operating procedures for upgrades, VMotions, VM build standards, etc.

Mixed Environments?

juchestyle — Tue, 29 Aug 2006 17:28:16 GMT

Great answers guys, love the answer from Sbeaver about vmtn and dev environment!

Dpomeroy, great answer too! I subscribe to your ideology as well. I don't think we have a big enough environment yet to farm out Dev and Prod yet. And I agree that competent vm'ers and management should feel comfortable putting the barriers in place to keep things from impacting production. Though I don't think that things like that have been done here yet.

I am holding onto the idea that mixing vms of different component needs is the way to maximize your virtual investment, irrespective of dev/test and production.

Anyone Disagree?

Respectfully,

What are we worth?

Jasemccarty — Tue, 22 Aug 2006 20:59:05 GMT

I'll be there with bells on.

I may not live up to my college years (being called Bacardi), but I'll make an effort!