VMware Communities : Thread List - ESX Server 2.x - VC 1.x Archives

citrix / ESX

vmpowered — Mon, 15 Aug 2005 16:40:51 GMT

My customer has a Citrix farm running on multiple ESX servers, with Multiple VM's running Citrix.

Unfortunately none of these VM's were check to be Citrix when they were built.

I see in the .vmx record a line that says:

workload = "TerminalServices" in a test vm I built.

My question is and I believe to be yes, can I just simply modify the existing .vmx and reboot the VM, in order for ESX to know what it really is?

Troubles detecting nic settings

softman — Fri, 02 Mar 2007 13:05:03 GMT

I try to find the settings (as ip e.t.c) for the nic i installed under ESX 2.5.4. Some helpful forum members gave me some solution, but now i face the next problems.

Under the root i type the command: /etc/sysconfig/network-scripts/ifcfg-eth0

But every time i get the message:

root#/etc/sysconfig/network-scripts/ifcfg-eth0
BASH": /Permission denied.

Can somebody tell me what the problem is?

My linux knowledge is low

Thanks for any kind of help

VPXClient error - exception that could not be handled

jgassie — Fri, 28 Oct 2005 17:26:14 GMT

Running the vpxclient on an XP SP2 machine is yielding this error:

vpxClient.exe - Common Language Runtime Debugging Services
Application has generated an exception that could not be handled.
Process id=(value varies), Thread id=(value varies)
Click OK to terminate the application.
Click CANCEL to debug the application.

Has anyone else received this error ?

I also get an error on the DEBUG attempt. I'm not certain that the second error is as much concern as the first. The second error states the following:

"vpxClient.exe - No debugger found.
Registered JIT debugger is not available. An attempt to launch a JIT debugger with the following command resulted in an error code of 0x2 (2). Please check computer settings.

cordbg.exe !a 0xb14

Click on Retry to have the process wait while attaching a debugger manually.
Click on Cancel to abort the JIT debug request."

How to enable Telnet on ESX server

ratheesh — Fri, 02 Mar 2007 07:24:26 GMT

I tried starting xinetd service but still cannot able to telnet the ESX serverr from my windows system

SCSI Errors on ESX 2.5.3 Patch 4

Shaha — Fri, 02 Mar 2007 09:02:35 GMT

We have been running ESX 2.5.3 patch 4 with emulaex HBAs connected to EMC DMX 3 for the past 4 months but over the last couple of weeks we have notice the following errors on some of the ESX servers:

Feb 28 19:44:35 UKWATVSH114 vmkernel: 98:00:22:25.848 cpu1)WARNING: SCSI: 5681: vmhba1:0:22:1 status = 0/8 0x0 0x0 0x0
Feb 28 19:44:35 UKWATVSH114 vmkernel: 98:00:22:25.848 cpu1)WARNING: SCSI: 5681: vmhba1:0:22:1 status = 0/8 0x0 0x0 0x0
Feb 28 19:44:35 UKWATVSH114 vmkernel: 98:00:22:25.848 cpu1)SCSI: 2840: Completing reset on handle 417486 (0 outstanding commands)
Feb 28 19:44:42 UKWATVSH114 vmkernel: 98:00:22:33.278 cpu2)WARNING: SCSI: 5681: vmhba1:0:23:1 status = 0/7 0x0 0x0 0x0
Feb 28 19:44:46 UKWATVSH114 vmkernel: 98:00:22:37.278 cpu2)WARNING: SCSI: 5681: vmhba1:0:23:1 status = 0/7 0x0 0x0 0x0
Feb 28 19:44:47 UKWATVSH114 vmkernel: 98:00:22:38.278 cpu2)WARNING: SCSI: 5681: vmhba1:0:23:1 status = 0/7 0x0 0x0 0x0

Has anyone seen this?

CPU0 100% Utilized

sgunelius — Tue, 13 Feb 2007 19:55:05 GMT

This morning we were experiencing some issues that seemed to indicate that one of my VMs (dual vCPU, 2GB memory, running W2K3 and SQL Server 2000) was resource starved. I checked vmkusage and noted that CPU0 had been pegged at 100% since last night around 10:00PM. I checked the top command and noted one process supporting vmware-vmx that appeared to be the culprit and killed that process which in turn took down the associated VM (it was a different VM than the one that was performing poorly).

I hadn't bothered to check esxtop to gather some additional information, but noted aftwards which VM was impacted and brought it back online with no issues. Now I'm trying to perform some analysis to determine why this VM flaked out and utilized all of CPU0. The VM that flaked was single vCPU, 512MB running W2K3 and a network paging application. This VM has historically demonstrated very low resource consumption. The only pertinent entries I could find the vmkernel log were as follows:

Feb 13 09:03:41 esxvm1 vmkernel: 173:01:45:29.752 cpu0)Host: vm 208: 4884: destroying world from host
Feb 13 09:03:41 esxvm1 vmkernel: 173:01:45:30.093 cpu0)Host: vm 208: 4884: destroying world from host
Feb 13 09:03:41 esxvm1 vmkernel: 173:01:45:30.093 cpu0)CpuSched: vm 208: 12029: WAIT -> ZOMBIE
Feb 13 09:03:41 esxvm1 vmkernel: 173:01:45:30.093 cpu0)CpuSched: vm 208: 11851: removing shares: alloc=1000, base=2105
Feb 13 09:03:41 esxvm1 vmkernel: 173:01:45:30.093 cpu0)World: vm 208: 1196: world isn't running, scheduling reap
Feb 13 09:03:41 esxvm1 vmkernel: 173:01:45:30.093 cpu0)Host: 3693: irq 4 vector 0x29 (host 0x41)
Feb 13 09:03:41 esxvm1 vmkernel: 173:01:45:30.093 cpu0)IDT: 1096: 0x29 for host
Feb 13 09:03:42 esxvm1 vmkernel: 173:01:45:31.094 cpu6)World: vm 208: 421: Starting cleanup via module table
Feb 13 09:03:42 esxvm1 vmkernel: 173:01:45:31.094 cpu6)World: vm 208: 423: refCount=3
Feb 13 09:03:43 esxvm1 vmkernel: 173:01:45:31.486 cpu6)World: vm 208: 439: cleanup done
Feb 13 09:05:55 esxvm1 vmkernel: 173:01:47:43.671 cpu0)World: vm 213: 894: Successfully created new world:

We're still running ESX Server 2.5.2 (build 13057) and have pretty allocated most of the available memory on this host. We don't run any jobs on the VM or COS during the timeframe when CPU0 was pegged and none of the VMs show a corresponding jump in CPU consumption during this timeframe. Can anyone suggest where else I should look in the logs or how I can diagnose what happened? Thank you in advance for your response.

Scott

Cannot logon to MUI or VC but can connect via putty

jpmfresno — Wed, 21 Feb 2007 16:56:41 GMT

I the following error when trying to connect to MUI:

"Unexpected response from vmware-authd: 511 Error connecting to /usr/sbin/vmware-serverd process. "

In VC the ESX server is greyed out and I can't connect to it.

The vm's are running and I don't want to restart ESX. Any suggestions?

Thanks in advance!

Virtual Center Incorrect Licensing in Quad Core System

CalvinMorrow — Mon, 26 Feb 2007 18:15:29 GMT

I just installed and started evaluation of Virtual Center 1.41 to manage Virtual Server (1.01) on Linux. We have a brand new Dell 2950 server with 2x Intel Quad Core Xeon processors which show up incorrectly in Virtual Center licensing as 8 Processors, 8 Cores.

Is this a known issue, and are there any workarounds? We're looking to buy the VMware Server Small Business Bundle, but it would seem that in the current configuration, we would have to purchase a very high processor count (8) to fit the licensing.

Relevant Specs:
Virtual Center 1.41 (33425) on Windows Server 2003 SP1.
VMware Server 1.01 on Ubuntu 6.06 x64
Dell 2950, 2x Intel Xeon E5335 (2.0 GHz - Quad Core)

ESX 2.1.2 upgrade to 2.5.4 necessary?

RigasW — Wed, 28 Feb 2007 09:18:45 GMT

Hi everybody,

we have been running ESX 2.1.2 (build 10921) for about two years without any problems.

Do you see any reasons why we should upgrade to release 2.5.4?

Performance, Features, Security?

Thanks in advance for your help.

Rigas

ESX 2.5.4 Patching procedure

fraber01 — Wed, 28 Feb 2007 21:33:35 GMT

I'm running ESX 2.5.4 build 32233 and I'm wondering if i need to apply all 5 patches one after the other OR if the 5th patch contains all of the previous patches?

VMotion gets 95% then hangs

KNardi — Tue, 12 Dec 2006 16:21:25 GMT

We recently migrated to VC 2.0, but after this past weekend we had some issues and had to move everything back to our 1.4.1 VC Server.

Now when I try to VMOTION the servers get to 95% and hang....

Some servers go to 80% then hang.
They have been in the running state for about 45 minutes.

I cannot cancel the operation and they are just sitting there trying to VMOTION.

Please advise
thanks

ESX 3.0 reboting and rebooting

goofy — Wed, 28 Feb 2007 05:44:12 GMT

Hi,

I triing to install ESX 3.0 on a pentium m motherboard.
The install process is going ok, but when it is rebooting its rebooting all the time after uncompressing the kernel...
Any body have a solution?

Nicolas

Could not Read from Connection

Tim Franz — Mon, 26 Feb 2007 09:27:27 GMT

Hello @all Users

we have a critical Problem with our Virtuell Center Version 1.4.1 Build 33425.
When we start the Console in Virtuell Center pop up a errormessage "Connection Error Could not Read from Connection"
Have anybody a idea to solve this problem.

Thank you for your help

Oracle RAC on Linux Cluster-Across-Boxes

multirotor — Wed, 28 Feb 2007 09:55:28 GMT

Is there a recommended configuration for this like there is for a Microsoft Cluster ?

I know that VMware recommends booting from local storage and using LSI Logic instead of Buslogic for MS Clustering.

We have been running this RAC setup for over one year.
Recently, we have been encountering strange SAN controller reboots and I want to remove everything which is not supported out of our configuration.

We have always booted both VM's from SAN VMFS volumes and we have always used Buslogic HBA's in the VM's...

Is this only not supported with Windows VM's or does this count vor RHEL3 too ?

Thanks !

DVD Problem

PatB — Mon, 12 Feb 2007 20:18:03 GMT

I can't get my virtual machines under ESX Server 2.5.1 to read DVD-R disks.

I've tried to research this in the ESX Server 2.5 documentation and throughout this forum, but can't seem to find anything pertinent.

ESX Server 2.5 documentation references the device as a DVD/CD-ROM, but whenever I try to read a DVD in the attached drive, it displays a message saying "Windows cannot read from this disk."

CD's are fine. The only problem is with DVD's.

The host is a Dell PowerEdge 2850. It has a DVD drive, not a CD drive.

All vm's are running Windows Server 2003.

Thanks for any advice you can give me.

ERROR "Thin/TBZ disks cannot be opened in multiwriter mode"

nidhi_opsware — Tue, 27 Feb 2007 20:37:28 GMT

Getting this error when I try to Power On a newly created VM.

Please suggest what's wrong with this VM.

/Nidhi

Error in WinSCP

foster29 — Fri, 23 Jun 2006 15:27:51 GMT

So i just spent 4 hours copying over a 40gb file using WinSCP. at the completion of the end I recieved a error message and now the image is giving the file extension . Filepart

any ideas? do i need to run the conversion?

Slow keyboard repeat rate

foster29 — Fri, 16 Feb 2007 21:57:52 GMT

Has anyone experienced a slow keyboard repeat rate? If so is there anyway to increase it?

I've adjusted what I could through the control panel/keyboard but had no impact.

thanks

4 ESX 2.5.3 35703 Servers Crashed at the same time

bradyrandolph — Sun, 11 Feb 2007 07:08:16 GMT

All-

I upgraded these four servers from 2.1.2 to 2.5.3 via the 2.5.3 upgrade patch and all four ASR'd at 12:36 this morning. Has any one else seen this yet?

ESX 2.5.2 Emergency CD/DVD

innu — Mon, 26 Feb 2007 06:00:13 GMT

Hi,

Currently i am running ESX server 2.5.2 on a HP DL380 hardware and I have important windows 2003 server configured on this hardware.

Currently everything is working fine, without any issues. I am planning for disaster recovery methods. For the virtual m/c i have taken measures by using Acronis TrueImage, but I also want some sort of emergency CD/DVD for the ESX server itself.

What if the ESX server doesnt boot? or Crashes.
What are the recovery methods available, I know its build on Linux kernel, but i was wondering any emergency CD/DVD that is available for ESX server to boot it in case of boot failures and then perform fixes.

Please provide suggestions.

Thanks

load balancing in the esx server

atsushikataoka — Tue, 27 Feb 2007 02:36:06 GMT

Hello ,

I am new to Vmware world.

Suppose i have a Esx server . In each of which one 4 VMs are configured.

I want that if i start the VM it should get loaded on the VM Server with minimum load.

How can i create my own application for the same.

What all i have to do for doing it?

If one server is already have load and another server is free Can i migrate the VM from one server to another server.

thank you
Atsushi Kataoka

ESX 2.1.1 & DST. Yet another approach.

markpace — Mon, 26 Feb 2007 20:37:22 GMT

Lets say I do not update ESX server for the DST change. All my W2K servers are patched and ready for the time change. What if on the Friday, March 9th I un-check VMware Tools: Time synchronization. Let W2K change the time automatically, then wait until April 1 and re-check Time synchronization.

Would this approach work? What are the potential problems?

Thanks in advance for any insight.

Mark Pace

Silently upgrade VMware Tools

Tigernoy — Fri, 23 Feb 2007 19:58:09 GMT

Hi All,

I have done a lot of searching and reading a lot of forum threads on how to silently upgrade VMware Tools. They are quite a bit of good stuff people put out there, but does not seem to work correctly for me.

Here is what I have...

VC 1.4
ESX 2.5.4-32233.
Hundreds of Win 2000/2003 VMs members of a domain need tools upgrade. VMs are running different version of VMware Tools.

What I have tried...

Action:
a. msiexec /a "VMware Tools.msi" (build 32233)
b. AD GPO to disable Driver Signing and deploy MSI
c. Apply GPO to test VMs - VM1 is Win2003 and VM2 is Win2000

Result:
1. Both VMs got the GPO applied, MSI was install, rebooted. In Vmware Tools Properties, it has build 32233. But if you go to c:\Program Files\VMware\VMware Tools and check the version of VMwareUser.exe and all the drivers, they still the old version.
2. I removed the MSI package from the GPO and chose "All users to continue using the software, but prevent new installations." radio button and delete the GPO.
3. Re-did "msiexec /a <msi>" and created a new GPO with new name.
4. I added VM3 (Win 2003) to the GPO. This time, the VM uninstalled the tools and rebooted. Of course, once the tools uninstalled, the VM had no network connectivity because there no NIC driver.

Question:
1. Why not all files/drives on the VM1 and VM2 get upgraded?
2. Why VM3 uninstalled the tools instead of upgrade?

Please help!!!
Any help is greatly appreciated.
TN

vmkernel log

fordian — Tue, 30 Jan 2007 13:18:38 GMT

Hello,

Everydays at the same time I have a message coming from the /var/log/vmkernel. And at the same time the windows guest os freeze.

Any idea about what it means ?

/var/log/vmkernel :

cpu2)SCSI: 2791: Reset requested on handle 157345 (0 outstanding commands)
cpu7)elx_sli_issue_iocb_wait_high_priority: reset-wait-time was 1 ms
cpu7)LinSCSI: 3543: done, retval 0x0
cpu7)SCSI: 2840: Completing reset on handle 157342 (0 outstanding commands)
cpu7)SCSI: 2983: Reset on handle 157345 [0/0]
cpu7)SCSI: 2612: handle 157345/157345: 0 outstanding commands
cpu7)LinSCSI: 3487: start
cpu7)<4>lpfc0:0713:FPe:SCSI layer issued Target Reset Data: x1 xlx
cpu7)elx_sli_issue_iocb_wait_high_priority: reset-wait-time was 1 ms
LinSCSI: 3543: done, retval 0x0
cpu7)SCSI: 2840: Completing reset on handle 157345 (0 outstanding commands)
cpu3)SCSI: 827: START_UNIT cmd issued to virt disk on virtual:0:0
cpu2)SCSI: 782: INQUIRY request with EVDP set
cpu4)WARNING: Mem: 2217: mainHeap: vmkernel memory below 8% (2407872 top free, 3837096 free)
Mem: 2221: mainHeap: Mem_Alloc called from 0x40dbf8, 619 times, size 268

Thank you

Dominic

Cannot Connect to ESX management website.

andrewhanna22 — Thu, 22 Feb 2007 13:17:12 GMT

Hello,

I am using ESX 2.5.3 server. I use Internet Explorer to connect to the ESX server to create/modify my virtual servers. This has worked very well in the past but I can no longer access the webpage to work on my servers. I haven't actually used the network in a couple of months, which may be the problem but I need to run some tests and cannot access the virtual servers. I'm getting a "Page Cannot be Displayed" error message when I try to use the IP address for the ESX server. This has worked fine in the past and I get a response when I ping the IP address for the ESX...not the Virtual Machines, but I think they are currently turned off anyway.

Any ideas on why I can't access the website to manage my ESX server or how to get the connection working again?

Thanks,

Andrew

ESX 2.1.1 & DST. A different approach...

markpace — Mon, 26 Feb 2007 14:33:02 GMT

Could I simply copy /etc/localtime from a different linux that has the correct DST values in /etc/localtime?

Implications of changing VM display name

mikeyboy — Fri, 23 Feb 2007 15:43:25 GMT

Hi, I am thinking about changing the display names of all our virtual machines to include the VM role (eg. SERVER1 - Citrix). We use Virtual Center 1.3 16701, ESX 2.5.2 17157, and will use VC to change the display names. Are there any implications (or gotchas) of doing this that might affect the operation or managability of our VMWare environment?

For example, I just discovered something unexpected...if I change the display name using virtual center, then all is OK, nothing else changes. However..if I then migrate that VM to another host (leaving the disk file on the same LUN,) the .vmx filename and path change to reflect the new display name. For example if it was /home/SERVER1/SERVER1.vmx, it changes to /home/SERVER1 - Citrix/SERVER1 - Citrix.vmx. I suspect that if I'd told VC to migrate the disk files on a different LUN, it would have renamed the disk files too. I don't like this, it just feels messy and I have a feeling it will cause confusion in the future.

Thanking you

Michael

Backing Up ESX Server/Service Console

MarkW72 — Tue, 20 Feb 2007 10:55:31 GMT

Hello,

Wondering whether anyone could give me some advice. We are currently using ESX Server 2.5.2 (planning to move to VI3 in the next month or so) and we backup our VM's using esxRanger Pro which seems to work fine.

However, I want to be able to backup the ESX Server OS itself (including the Service Console) in case it ever falls over and we need to rebuild it quickly. We use Backup Exec here and I did purchase the Linux agent last year for this purpose. However, I had a lot of problems getting the agent installed on the ESX server.

So I'm looking for alternative ways to backup this server. I was wondering what other people used to achieve this. Would using Acronis Trueimage be a good option ? Has anyone used other "imaging" based software to "backup" their ESX server ?

Thanks

Mark

Console Operating System Terminology question

gatorback — Sun, 25 Feb 2007 01:31:48 GMT

When I look at my 2003 Server VM's VMware tools, there is an option to sync the VM's time to the 'Console Operating System' (COS). I think COS means the VMware RedHat 7.2 kernal, upon which the VM resides. Is this right?

Spanned LUNS not seen by 2nd Host

devlindj — Mon, 19 Feb 2007 15:35:24 GMT

This is my problem:

I spanned an extent by the size of an additional LUN on my 1st ESX server in a "VMotion" group.
On the 1st ESX server things look good and everything works fine.
VMotioned all guests from second ESX server to the first then re-booted 2nd ESX server.
The 2nd EXS server now sees all LUNS as unformatted individual disks (with exception of 1) and that's the basics of my problem.

I've done this 3 other times previously, to these exact servers, without issue. Not sure what happenned this time?

Also:
Have 2 other ESX servers that share the same LUNs/Spanned extent. These two I have done nothing with, they still see the original configuration of the disks but I don't want to go farther until I figure out why ESX server #2 can see them properly.

Here's how I'm configured:
4 x ESX server versions 2.5.0 build 11548
Each ESX server masked the same 6 x 100GB LUNS.
All 6 LUNS are "spanned" into one VMFS volume.
Masked a 7th LUN Looks like this:

Server One
1. VMLUN01 (VMFS-2.11 Volume) | Edit...
Total Capacity 707.12 G
Maximum File Size 456.00 G
Access Mode Public
Spanned Extents vmhba0:0:46:1 101.02 G
vmhba0:0:47:1 101.02 G
vmhba0:0:48:1 101.02 G
vmhba0:0:49:1 101.02 G
vmhba0:0:44:1 101.02 G
vmhba0:0:50:1 101.02 G
vmhba0:0:45:1 101.02 G

Server 2 looks like this:
Disk vmhba0:0:44: (0% of 101.02 G Available)
1. Unformatted VMFS Volume | Edit...
Capacity 101.02 G

Disk vmhba0:0:45: (0% of 101.02 G Available)
1. VMFS-2.11 Volume | Edit...
Capacity 101.02 G
Maximum File Size 456.00 G
Access Mode Public

Disk vmhba0:0:46: (0% of 101.02 G Available)
1. Unformatted VMFS Volume | Edit...
Capacity 101.02 G

Disk vmhba0:0:47: (0% of 101.02 G Available)
1. Unformatted VMFS Volume | Edit...
Capacity 101.02 G

Disk vmhba0:0:48: (0% of 101.02 G Available)
1. Unformatted VMFS Volume | Edit...
Capacity 101.02 G

Disk vmhba0:0:49: (0% of 101.02 G Available)
1. Unformatted VMFS Volume | Edit...
Capacity 101.02 G

Disk vmhba0:0:50: (0% of 101.02 G Available)
1. Unformatted VMFS Volume | Edit...
Capacity 101.02 G

Any thoughts?
and thanks in advance...
David

VM Rebooted Unexpectedly

santans — Fri, 16 Feb 2007 10:55:48 GMT

Hi
I have a ESX running 2 guests,win2k3 & Novell Netware 6.5.
The VM hosting Novell was restarted suddenly & generated the below Logs

Feb 15 13:17:08: vcpu-0| SVGA: Unregistering IOSpace at 0x1400 (0x1400)
Feb 15 13:17:08: vcpu-0| SVGA: Unregistering MemSpace at 0xf6000000(0xf6000000) and 0xf5000000(0xf5000000)
Feb 15 13:17:08: vcpu-0| SVGA: Registering IOSpace at 0x1400 (0x0000)
Feb 15 13:17:08: vcpu-0| SVGA: Registering MemSpace at 0xf6000000(0xf6000000) and 0xf5000000(0xf5000000)
Feb 15 13:17:16: vcpu-0| Flushing LA 0xf88cf3e1 in [0x4c7..0xf8952] at 0x4029:0x218b1a
Feb 15 13:17:17: vcpu-0| LSILogic: Adapter 0 Port 0 Operational.
Feb 15 13:17:24: vcpu-0| Flushing LA 0xf8804244 in [0xf8804..0xf8804] at 0x4029:0x1f930e
Feb 15 13:17:29: vcpu-0| Flushing LA 0x31cd828c in [0x31cd8..0x31cd8] at 0x4029:0x1a8fca
Feb 15 13:17:35: mks| (MKS) DeclareThread mks module=8 source=0x100000
Feb 15 13:17:35: mks| Accepted new connection at 19 for thread mks (0x81a5f14)
Feb 15 13:17:35: mks| MKS server remote MKS connected.
Feb 15 13:17:35: mks| MKS server new remote connection.
Feb 15 13:17:35: mks| MKS server powering on remote display.
Feb 15 13:17:35: mks| VMxfer: Waiting for version negotiation...
Feb 15 13:17:36: vcpu-0| Flushing LA 0x62771e38 in [0x62771..0x62771] at 0x4029:0x1a9f4b
Feb 15 13:17:36: mks| IPC version negotiation version: MKS returning 2.1 to mks
Feb 15 13:17:36: mks| IPC remoteMKS-temp version: MKS returning 10.34 to mks that tried 11.4
Feb 15 13:17:37: mks| IPC remoteMKS-temp version: MKS returning 10.34 to mks that tried 10.34
Feb 15 13:17:37: mks| VMxfer: done.
Feb 15 13:17:40: vcpu-0| Flushing LA 0x62771e38 in [0x62771..0x62771] at 0x4029:0x230481
Feb 15 13:17:45: mks| MKS IPC closed the connection with thread mks (0x81a5f14)
Feb 15 13:17:45: mks| MKS server disconnecting thread.
Feb 15 13:17:49: vcpu-0| Flushing LA 0x6605c442 in [0x6605c..0x6605d] at 0x4029:0x377c56
Feb 15 13:17:50: vcpu-0| Flushing LA 0x655f64c0 in [0x655f6..0xf8801] at 0x4029:0x27dabe
Feb 15 13:17:51: vcpu-0| Flushing LA 0xf883f0c4 in [0x64c76..0xf883f] at 0x4029:0x1a6b1b
Feb 15 13:17:53: vcpu-0| Flushing LA 0xf693d274 in [0x6471c..0xf693d] at 0x4029:0x1b5c75
Feb 15 13:17:56: vcpu-0| Flushing LA 0xfba20ee8 in [0xaf7..0xfbbfc] at 0x4029:0x1aac52
Feb 15 13:18:01: vcpu-0| GuestRpc: Channel 0, registration number 1, guest application toolbox.
Feb 15 13:18:01: vcpu-0| Flushing LA 0x669a1780 in [0x34fb1..0x669a1] at 0x4029:0x290ff4
Feb 15 13:18:01: vcpu-0| Guest: toolbox: Version: build-16390
Feb 15 13:18:01: vcpu-0| TOOLS setting the tools version to '6309'
Feb 15 13:18:01: vcpu-0| TOOLS unified loop capability requested by 'toolbox'; now sending options via TCLO
Feb 15 13:21:50: vcpu-0| Flushing LA 0xb6e46c88 in [0xb6e46..0xb6e46] at 0x4029:0x19941e
Feb 15 13:21:57: vmx| (VMX) DeclareThread ui module=6 source=0x40000
Feb 15 13:21:57: vmx| Accepted new connection at 31 for thread ui (0x8273130)
Feb 15 13:21:57: vmx| VUI: A new gui connected (user = aeadmin) (ip = 10.71.3.41).
Feb 15 13:21:57: vmx| IPC version negotiation version: VMX returning 2.1 to ui
Feb 15 13:21:57: vmx| IPC remoteUI-temp version: VMX returning 10.34 to ui that tried 10.34
Feb 15 13:21:57: vmx| VUIUpdateProgress not implemented
Feb 15 13:21:57: vmx| vuiupdatehelp not implemented
Feb 15 13:21:57: vmx| VMX: changing state 2 from 1878 to 1878
Feb 15 13:21:57: mks| (MKS) DeclareThread mks module=8 source=0x100000
Feb 15 13:21:57: mks| Accepted new connection at 19 for thread mks (0x81a5f14)
Feb 15 13:21:57: mks| MKS server remote MKS connected.
Feb 15 13:21:57: mks| MKS server new remote connection.
Feb 15 13:21:57: mks| MKS server powering on remote display.
Feb 15 13:21:57: mks| VMxfer: Waiting for version negotiation...
Feb 15 13:21:58: mks| IPC version negotiation version: MKS returning 2.1 to mks
Feb 15 13:21:58: mks| IPC remoteMKS-temp version: MKS returning 10.34 to mks that tried 10.34
Feb 15 13:21:58: mks| VMxfer: done.
Feb 15 13:23:24: vmx| VMX IPC closed the connection with thread ui (0x8273130)
Feb 15 13:23:24: vmx| VMX: ui disconnected.
Feb 15 13:23:24: mks| MKS IPC closed the connection with thread mks (0x81a5f14)
Feb 15 13:23:24: mks| MKS server disconnecting thread.
Feb 15 13:25:36: vcpu-0| Flushing LA 0x6b072b4c in [0x19..0x6b072] at 0x4029:0x1a41a8
Feb 15 13:31:58: vcpu-0| Flushing LA 0x67cab72c in [0x67cab..0x67cab] at 0x4029:0x1ac7ca
Feb 15 13:38:20: vcpu-0| Flushing LA 0x6bc55920 in [0x6bc55..0x7112e] at 0x4029:0x197148
Feb 15 13:44:30: vcpu-0| Flushing LA 0xf8a91b80 in [0x6ae84..0xf8a91] at 0x4029:0x198fdc
Feb 15 13:50:21: vcpu-0| Flushing LA 0x627c3024 in [0x627c3..0x627c3] at 0x4029:0x197434
Feb 15 13:55:51: vcpu-0| Flushing LA 0x68d45620 in [0x68d45..0x6b9da] at 0x4029:0x1970cf
Feb 15 14:01:32: vcpu-0| Flushing LA 0x68e3ff08 in [0x68e3f..0x6e0ec] at 0x4029:0x1ad63f
Feb 15 14:07:26: vcpu-0| Flushing LA 0xf8801a44 in [0x69f19..0xf8801] at 0x4029:0x199cf8
Feb 15 14:13:16: vcpu-0| Flushing LA 0xb7d0cc40 in [0xb6dce..0xb7d0c] at 0x4029:0x1ab153
Feb 15 14:32:34: vmx| Stopping VCPU threads...
Feb 15 14:32:34: vmx| VTHREAD destroy thread 4 "vcpu-0"
Feb 15 14:32:34: vmx| Poll: double lock disable
Feb 15 14:32:34: vmx| LSILogic: LSILogicPowerOff: Adapter-ID:0 (0x831eee8)
Feb 15 14:32:34: vmx| AIOPowerOff ide0:0
Feb 15 14:32:34: ide0:0| AIOSlave: ide0:0 exiting cleanly.
Feb 15 14:32:34: ide0:0| VTHREAD thread 6 start exiting
Feb 15 14:32:34: ide0:0| VTHREAD counting thread 0
Feb 15 14:32:34: ide0:0| VTHREAD counting thread 1
Feb 15 14:32:34: ide0:0| VTHREAD thread 6 exiting, 2 left
Feb 15 14:32:34: vmx| VMX IPC closed the connection with thread ide0:0 (0x82730b4)
Feb 15 14:32:34: vmx| AIOPowerOff Floppy
Feb 15 14:32:34: Floppy| AIOSlave: Floppy exiting cleanly.
Feb 15 14:32:34: Floppy| VTHREAD thread 5 start exiting
Feb 15 14:32:34: Floppy| VTHREAD counting thread 0
Feb 15 14:32:34: Floppy| VTHREAD counting thread 1
Feb 15 14:32:34: Floppy| VTHREAD thread 5 exiting, 2 left
Feb 15 14:32:34: vmx| VMX IPC closed the connection with thread Floppy (0x8273038)
Feb 15 14:32:34: vmx| VMX: changing state 1 from 1873 to 1873
Feb 15 14:32:35: vmx| VMX: changing state 0 from 1872 to 1870
Feb 15 14:32:35: vmx| VMX idle exit
Feb 15 14:32:35: vmx| VMX IPC closed the connection with thread servercontrol (0x8272fbc)
Feb 15 14:32:35: vmx| VMX: Remote VMControl client servercontrol disconnected.
Feb 15 14:32:35: vmx| IPC_exit: disconnecting all threads
Feb 15 14:32:35: vmx| VMX IPC closed the connection with thread MKS (0x8272f40)
Feb 15 14:32:35: vmx| VMX IPC The "MKS" thread exited cleanly.
Feb 15 14:32:35: vmx| VMX exit.
Feb 15 14:32:35: mks| MKS IPC closed the connection with thread VMX (0x81a5e98)
Feb 15 14:32:35: mks| MKS: Thread VMX exited as expected.
Feb 15 14:32:35: mks| Local MKS exit.
Feb 15 14:32:35: mks| VTHREAD thread 1 start exiting
Feb 15 14:32:35: mks| VTHREAD counting thread 0
Feb 15 14:32:35: mks| VTHREAD thread 1 exiting, 1 left

Can any one let me find out the cause for the reboot.

Thanks

Problems loading ide-scsi.o for ESX 2.5.4, Build 38650(kernel 2.4.9-vmnix2)

BStrauss — Fri, 23 Feb 2007 21:11:21 GMT

Re: ESX 2.5.4 Build 38650

In order to attempt to use raw CD-ROM access I need to utilize the ide-scsi.o module. I have absolutely no problem loading the module in either the "linux" or "linux-up" boot configs, but under the normal "esx" config, the ide-scsi module will not load. A manual attempt to load it via "insmod ide-scsi" results in the following message:

/lib/modules/2.4.9-vmnix2/kernel/drivers/scsi/ide-scsi.o: unresolved symbol scsi_register_Rvmnix_4001ce65

Apparently a symbol version suffix mismatch is occurring here.

Where can I obtain a copy of ide-scsi.o appropriate for this build?
-OR-
What is the appropriate procedure for compiling a suitable copy?

Symantec System Center w/ Corporate AV and ESX 2.5

Straterra — Fri, 23 Feb 2007 17:18:01 GMT

I currently have a dying Windows 2000 server I am replacing. This server runs Symantec System Center and runs Symantec Server on it to manage all of my Symantec Corporate Antivirus Clients. I have a fully updated Windows 2003 Server Standard virtual machine created with Symantec Server deployed and System Center installed. When I open System Center to config and move clients from the old server to the new, I get an error message that the new server cannot be contacted and it is listed as "Disabled."

I thought that was strange, so I fired up VMWare Workstation and installed a Windows 2003 Server Standard VM and installed the exact same software on it. Everything worked beautifully. So...I am wondering...Is this a problem with ESX or is there a patch I need?

URL's in the Forum...

Mike_Laverick — Fri, 19 Aug 2005 09:45:52 GMT

Hi,

Can anyone remind me of the syntax of how to post long URL's in the forum so they just appear as a hyperlink?

Also, is possible to embedd a graphic into a forum post...?

So the image would be on my webserver - but display in a post not as a URL but as an image?

Regards
Mike

ESX 2.5.2 Failed Vmotion

sheetsb — Thu, 01 Feb 2007 16:46:10 GMT

I am trying to use Vmotion to move virtual machines between different ESX 2.5.x hosts so that I can upgrade them to ESX 2.5.4 for the upcoming daylight savings time changes. The hosts were all running ESX 2.5.2 build 18819. I have been able to use vmotion within VCenter 2.0.1 build 32042 to move the VMs on all the hosts but one which fails with the error "Failed to communicate with remote host, either due to network errors or because the host is not responding."

The error occurs for every VM I try to move. None of the fourteen other ESX hosts had any problem. I suspect an issue with my Vmotion NIC in this host but I don't see any messages that seem to help.

Does anyone have some suggestions short of shutting down the VMs?

Thanks,

Bill S.

Odd behaviour..vmdk name being changed back by VC 1.3 after rename in MUI?

mikeyboy — Fri, 23 Feb 2007 16:07:33 GMT

Hi, please can anyone shed some light on why this odd behaviour occured?

I created a VM with VC. I then logged out of VC, went into the MUI, and renamed and reattached the virtual disks (eg. server1_d1.vmdk and server1_d2.vmdk instead of the VC defaults server1.vmdk and server1_1.vmdk). I then started VC, checked the properties, and the vDisks were named what I'd just changed them too, VM started fine..all OK there as expected. We then cold migrated the VM to another host, and the vmdk names were still the same as I'd renamed them to.

The following day I noticed in VC that the vDisks had gone back to their original names...how? I don't believe anyone renamed them on purpose..so what would have caused them to rename themselves back to their original names? I thought that maybe the cold migrate renamed them back to their originals because VC might have cached the disk names somewhere then reapplied them during the cold migration...but I definitely checked they were still the same when I'd migrated them. What could have happened...and what do I need to do to stop this happening again?

Thankyou
Michael

Adic Acalar24 tape library problem

kristof — Fri, 23 Feb 2007 14:46:12 GMT

Hi All,

I have Adic Scalar24 tape library connected to ESX 2.5.3 server (HP Proliant 385G1 hardware). One of virtual machines is media backup server (NetBackup) which has access to this tape library. I have problems with backup performance (less than 1MB/sec). On this virtual server I have lot of errors connected to tape library.

Inside ESX I can see one vmkernel error which accurs very ofter (sometimes almoust every minute).
Tape library is connected to ESX by dedicated Adaptec 29320A Ultra320 SCSI adapter and it is visible by ESX.

Is that vmkernel error connected to backup performance problems and errors inside virtual machine? How can I solve it?
ESX and tape library were rebooted and it does not solve this problem.

All logs and errors you can find below:

Virtual backup server errors:

Event Type: Error
Event Source: symmpi
Event Category: None
Event ID: 15
Date: 2/23/2007
Time: 2:00:28 PM
User: N/A
Description:
The device, \Device\Scsi\symmpi2, is not ready for access yet.

Event Type: Error
Event Source: NetBackup Tape Manager
Event Category: None
Event ID: 5122
Date: 2/23/2007
Time: 3:27:51 PM
User: N/A
Description:
DeviceIoControl() error on bus 0, target 0, lun 0: The device is not ready.

Vmkernel error inside ESX:

Feb 23 14:27:22 COMPNAME vmkernel: (scsi0:A:4:0): 7:23:50:50.858 cpu0)Abort Message Sent
Feb 23 14:27:22 COMPNAME vmkernel: (scsi0:A:4:0): 7:23:50:50.858 cpu0)SCB 4 - Abort Tag Completed.
Feb 23 14:27:22 COMPNAME vmkernel: 7:23:50:50.858 cpu0)found == 0x1
Feb 23 14:27:23 COMPNAME vmkernel: 7:23:50:51.814 cpu1)SCSILinuxGetDataDirection: unknown opcode: 0x8c
Feb 23 14:27:23 COMPNAME vmkernel: (scsi0:A:4:0): 7:23:50:51.815 cpu0)No or incomplete CDB sent to device.
Feb 23 14:27:23 COMPNAME vmkernel: (scsi0:A:4:0): 7:23:50:51.815 cpu0)Protocol violation in Message-in phase. Attempting t
o abort.

Some additional info about tape library and SCSI adapter:

less /proc/scsi/scsi
Attached devices:
Host: scsi0 Channel: 00 Id: 03 Lun: 00
Vendor: ADIC Model: Scalar 24 Rev: 310A
Type: Medium Changer ANSI SCSI revision: 02
Host: scsi0 Channel: 00 Id: 04 Lun: 00
Vendor: IBM Model: ULTRIUM-TD3 Rev: 64D0
Type: Sequential-Access ANSI SCSI revision: 03
Host: scsi1 Channel: 00 Id: 00 Lun: 00

less /proc/vmware/scsi/vmhba1/3:0

Vendor: ADIC Model: Scalar 24 Rev: 310A
Type: Medium Changer ANSI SCSI revision: 03
Id: 41 44 49 43 5f 31 5f 43 30 30 30 39 36 42 31 35 20 20 53 63 61 6c 61 72
Size: 0 Mbytes
Queue Depth: 2

Partition Info:
Block size: 0
Num Blocks: 0

Partition VM cmds reads KBread writes KBwritten cmdsAbrt busRst paeCmds paeCopies splitCmd
s splitCopies issueAvg totalAvg

VM Shares cmds reads KBread writes KBwritten cmdsAbrt busRst paeCmds paeCopies splitCmds
splitCopies issueAvg totalAvg active queued virtTime
150 1000 843 0 0 0 0 0 0 28 28 0
0 21871 2869155476 0 0 422000000
128 1000 2295 0 0 0 0 0 0 0 0 0
0 10057 114997476 0 0 500000
Total 2000 3138 0 0 0 0 0 0 28 28 0
0 13231 854881222 0 0 422000000

Paths:fixed
vmhba1:3:0 on*#

Active: 0 Queued: 0 Reserved: N Pending Reserves: 0

less /proc/vmware/scsi/vmhba1/4:0

Vendor: IBM Model: ULTRIUM-TD3 Rev: 64D0
Type: Sequential-Access ANSI SCSI revision: 04
Id: 31 32 31 30 31 38 36 33 38 37 55 4c 54 52 49 55
Size: 0 Mbytes
Queue Depth: 2

Partition Info:
Block size: 0
Num Blocks: 0

Partition VM cmds reads KBread writes KBwritten cmdsAbrt busRst paeCmds paeCopies splitCmd
s splitCopies issueAvg totalAvg

VM Shares cmds reads KBread writes KBwritten cmdsAbrt busRst paeCmds paeCopies splitCmds
splitCopies issueAvg totalAvg active queued virtTime
150 1000 483665 11835 1514880 309084 39562752 0 0 60830 60830 0
0 16093 4915433 0 0 2808585000000
128 1000 2292 0 0 0 0 0 0 0 0 0
0 9054 5494172 0 0 500000
Total 2000 485957 11835 1514880 309084 39562752 0 0 60830 60830 0
0 16059 4918162 0 0 2808585000000

Paths:fixed
vmhba1:4:0 on*#

Active: 0 Queued: 0 Reserved: N Pending Reserves: 0

less /proc/scsi/aic79xx/0

Adaptec AIC79xx driver version: 1.3.7
Adaptec 29320A Ultra320 SCSI adapter
aic7901A: Ultra320 Wide Channel A, SCSI Id=7, PCI-X 67-100Mhz, 512 SCBs
Allocated SCBs: 6, SG List Length: 85

Serial EEPROM:
0x17c8 0x17c8 0x17c8 0x17c8 0x17c8 0x17c8 0x17c8 0x17c8
0x17c8 0x17c8 0x17c8 0x17c8 0x17c8 0x17c8 0x17c8 0x17c8
0x9f4 0x152 0x2807 0x10 0xffff 0xffff 0xffff 0xffff
0xffff 0xffff 0xffff 0xffff 0xffff 0xffff 0x430 0xb403

Target 0 Negotiation Settings
User: 320.000MB/s transfers (160.000MHz RDSTRM|DT|IU|RTI|QAS, 16bit)
Target 1 Negotiation Settings
User: 320.000MB/s transfers (160.000MHz RDSTRM|DT|IU|RTI|QAS, 16bit)
Target 2 Negotiation Settings
User: 320.000MB/s transfers (160.000MHz RDSTRM|DT|IU|RTI|QAS, 16bit)
Target 3 Negotiation Settings
User: 320.000MB/s transfers (160.000MHz RDSTRM|DT|IU|RTI|QAS, 16bit)
Goal: 3.300MB/s transfers
Curr: 3.300MB/s transfers
Transmission Errors 0
Channel A Target 3 Lun 0 Settings
Commands Queued 2884
Commands Active 0
Command Openings 1
Max Tagged Openings 0
Device Queue Frozen Count 0
Target 4 Negotiation Settings
User: 320.000MB/s transfers (160.000MHz RDSTRM|DT|IU|RTI|QAS, 16bit)
Goal: 160.000MB/s transfers (80.000MHz DT, 16bit)
Curr: 160.000MB/s transfers (80.000MHz DT, 16bit)
Transmission Errors 0
Channel A Target 4 Lun 0 Settings
Commands Queued 484236
Commands Active 0
Command Openings 1
Max Tagged Openings 0
Device Queue Frozen Count 0
Target 5 Negotiation Settings
User: 320.000MB/s transfers (160.000MHz RDSTRM|DT|IU|RTI|QAS, 16bit)
Target 6 Negotiation Settings
User: 320.000MB/s transfers (160.000MHz RDSTRM|DT|IU|RTI|QAS, 16bit)
Target 7 Negotiation Settings
User: 320.000MB/s transfers (160.000MHz RDSTRM|DT|IU|RTI|QAS, 16bit)
Target 8 Negotiation Settings
User: 320.000MB/s transfers (160.000MHz RDSTRM|DT|IU|RTI|QAS, 16bit)
Target 9 Negotiation Settings
User: 320.000MB/s transfers (160.000MHz RDSTRM|DT|IU|RTI|QAS, 16bit)
Target 10 Negotiation Settings
User: 320.000MB/s transfers (160.000MHz RDSTRM|DT|IU|RTI|QAS, 16bit)
Target 11 Negotiation Settings
User: 320.000MB/s transfers (160.000MHz RDSTRM|DT|IU|RTI|QAS, 16bit)
Target 12 Negotiation Settings
User: 320.000MB/s transfers (160.000MHz RDSTRM|DT|IU|RTI|QAS, 16bit)
Target 13 Negotiation Settings
User: 320.000MB/s transfers (160.000MHz RDSTRM|DT|IU|RTI|QAS, 16bit)
Target 14 Negotiation Settings
User: 320.000MB/s transfers (160.000MHz RDSTRM|DT|IU|RTI|QAS, 16bit)
Target 15 Negotiation Settings
User: 320.000MB/s transfers (160.000MHz RDSTRM|DT|IU|RTI|QAS, 16bit)

LSILogic Reset

fordian — Thu, 25 Jan 2007 17:20:05 GMT

Hi,

Since 19 jan we have a vm that restart every day at 10:25 pm. The only thing that has been done on the vm is an expansion of volumes with diskpart on Windows 2003.
The volumes that have been expanded are located on adapter 1 and what I see from the vm's log is that adapter 0 is resetting.

I print the vmkernel log and vm log for the time it happened.

Any idea of what happened ???

Sorry it is not very easy to read :

VM's VMware.log file :
Jan 24 22:25:31: vcpu-0| LSILogic: Adapter 0 Port 0 Operational.
Jan 24 22:25:31: vcpu-0| LSILogic: Received SCSITaskMgmt request for adapter 0 TargetID 0 MsgContext 0x0 TaskType 0x4 SEQ: 5 cif 0 mfa (4:0:0)
Jan 24 22:25:31: vcpu-0| LSILogic: Resetting target 0, handle 102398001
Jan 24 22:25:31: vcpu-0| LSILogic: Resetting target 1, handle 102398008
Jan 24 22:25:31: vcpu-0| LSILogic: Resetting target 2, handle 102398011
Jan 24 22:25:31: vcpu-0| LSILogic: LSILogicCmdDoneReset: adapter 0, target 0
Jan 24 22:25:31: vcpu-0| LSILogic: target 0 reset completed (bitmap=0x6)
Jan 24 22:25:31: vcpu-0| LSILogic: LSILogicCmdDoneReset: adapter 0, target 1
Jan 24 22:25:31: vcpu-0| LSILogic: LSILogicCmdDoneReset: adapter 0, target 2
Jan 24 22:25:31: vcpu-0| LSILogic: LSILogicCmdDoneReset: adapter 0, target 0
Jan 24 22:25:31: vcpu-0| LSILogic: LSILogicCmdDoneReset: adapter 0, target 1
Jan 24 22:25:31: vcpu-0| LSILogic: target 1 reset completed (bitmap=0x4)
Jan 24 22:25:31: vcpu-0| LSILogic: LSILogicCmdDoneReset: adapter 0, target 2
Jan 24 22:25:31: vcpu-0| LSILogic: target 2 reset completed (bitmap=0x0)
Jan 24 22:25:31: vcpu-0| LSILogic: SCSI_TASK_MANAGEMENT: adapter 0 taskType 0x4 completed for adapter 0x367c8 ioTerminated 0 Outstanding IOs 0

VMkernel log file :

Jan 24 22:25:31 CPREXPVMWHYP2 vmkernel: 89:11:04:36.077 cpu7)SCSI: 2791: Reset requested on handle 102398001 (0 outstanding commands)
Jan 24 22:25:31 CPREXPVMWHYP2 vmkernel: 89:11:04:36.077 cpu4)SCSI: 2983: Reset on handle 102398001 [0/0]
Jan 24 22:25:31 CPREXPVMWHYP2 vmkernel: 89:11:04:36.077 cpu4)SCSI: 2612: handle 102398001/102398001: 0 outstanding commands
Jan 24 22:25:31 CPREXPVMWHYP2 vmkernel: 89:11:04:36.077 cpu4)SCSI: 2612: handle 79259/102398001: 0 outstanding commands
Jan 24 22:25:31 CPREXPVMWHYP2 vmkernel: 89:11:04:36.077 cpu7)SCSI: 2791: Reset requested on handle 102398008 (0 outstanding commands)
Jan 24 22:25:31 CPREXPVMWHYP2 vmkernel: 89:11:04:36.077 cpu4)SCSI: 2840: Completing reset on handle 102398001 (0 outstanding commands)
Jan 24 22:25:31 CPREXPVMWHYP2 vmkernel: 89:11:04:36.077 cpu4)SCSI: 2983: Reset on handle 102398008 [0/0]
Jan 24 22:25:31 CPREXPVMWHYP2 vmkernel: 89:11:04:36.077 cpu4)SCSI: 2612: handle 102398008/102398008: 0 outstanding commands
Jan 24 22:25:31 CPREXPVMWHYP2 vmkernel: 89:11:04:36.077 cpu4)LinSCSI: 3487: start
Jan 24 22:25:31 CPREXPVMWHYP2 vmkernel: 89:11:04:36.077 cpu7)SCSI: 2791: Reset requested on handle 102398011 (0 outstanding commands)
Jan 24 22:25:31 CPREXPVMWHYP2 vmkernel: 89:11:04:36.077 cpu4)<4>lpfc0:0713:FPe:SCSI layer issued Target Reset Data: x0 xlx
Jan 24 22:25:31 CPREXPVMWHYP2 vmkernel: 89:11:04:36.079 cpu4)elx_sli_issue_iocb_wait_high_priority: reset-wait-time was 1 ms
Jan 24 22:25:31 CPREXPVMWHYP2 vmkernel: 89:11:04:36.079 cpu4)LinSCSI: 3543: done, retval 0x0
Jan 24 22:25:31 CPREXPVMWHYP2 vmkernel: 89:11:04:36.080 cpu4)SCSI: 2840: Completing reset on handle 102398008 (0 outstanding commands)
Jan 24 22:25:31 CPREXPVMWHYP2 vmkernel: 89:11:04:36.080 cpu4)SCSI: 2983: Reset on handle 102398011 [0/0]
Jan 24 22:25:31 CPREXPVMWHYP2 vmkernel: 89:11:04:36.080 cpu4)SCSI: 2612: handle 102398011/102398011: 0 outstanding commands
Jan 24 22:25:31 CPREXPVMWHYP2 vmkernel: 89:11:04:36.080 cpu4)LinSCSI: 3487: start
Jan 24 22:25:31 CPREXPVMWHYP2 vmkernel: 89:11:04:36.080 cpu4)<4>lpfc0:0713:FPe:SCSI layer issued Target Reset Data: x1 xlx
Jan 24 22:25:31 CPREXPVMWHYP2 vmkernel: 89:11:04:36.082 cpu4)elx_sli_issue_iocb_wait_high_priority: reset-wait-time was 1 ms
Jan 24 22:25:31 CPREXPVMWHYP2 vmkernel: 89:11:04:36.082 cpu4)LinSCSI: 3543: done, retval 0x0
Jan 24 22:25:31 CPREXPVMWHYP2 vmkernel: 89:11:04:36.082 cpu4)SCSI: 2840: Completing reset on handle 102398011 (0 outstanding commands)

Thank you

Dominic

Remove NIC from live switch?

encode — Thu, 16 Nov 2006 07:06:56 GMT

Hi,

I'm trying to set up vmotion on an ESX 2.5.2 build 16390 host. Only problem is, the two unused NICs are currently assigned to a virtual switch which is in use.

The server is in production, so I really don't want to reboot it in order to remove a nic from a virtual switch. Is there any way to remove it from a switch dynamically? The nic in question is not physically connected at the moment, so there's no network traffic utilising the nic.

Thanks for your assistance.

Local disk vmfs2 volume showing deprecated after upgrade to 2.1.3

palosbob — Wed, 21 Feb 2007 20:34:58 GMT

We upgraded a test server running ESX 2.1.2 build 10921 to ESX 2.1.3 patch 3 build 33524. On reboot the MUI shows: Warning: No swap space configured or none is active. Selecting the Reconfigure option shows the previously configure swapfile but an activate gets "failed to activate the configured swap file".
The volume is showing the access mode as Private (deprecated mode) with the maximum file size a large negative number. Attempting to Edit and make public gives the error "could not change cluster access for partition "vmhba3:0:0:10".
Removing the volume and attempting to create volume gives errors "Failed to create VMFS on vmhba..."
Failed to rescan for VMFS. Reason: Illegal seek.
Error: Could not create VMFS file system on partition vmhba...Reason: No such file or directory. Make sure the disk is not in use. Also make sure no running virtual machines are using the partition, and that you do not have an active swap file on the partition.
Failed to create new VMFS.
Could not format partition: Resource is temporarily unavailable."

Displaying Disks/Luns again shows the volume as deprecated as before. No VMs are running. The SAN Lun is also showing deprecated but the volume shows all the .dsk/vmdk files via a putty session. A VM will run though.

Not sure what to look at next. Any help would be appreciated.

VMware tools upgrade question

softman — Thu, 22 Feb 2007 15:42:39 GMT

Can anyone tell me how i get the latest version of VMware Tools, or is is already inside the latest update of the ESX VMware server?

Thanks

Failed To Power On VM

MichaelCampbell — Mon, 08 Jan 2007 11:46:28 GMT

Hi

I have a VM on an ESX 2.52 host which will not power on. The VM was not accessible and so we tried to power it off. This hung and resulted in Virtual Center not being able to contact any VM on the ESX host. We resolved this by restarting the vmware-serverd service. This worked OK but the problem VM will still not start up. Can anyone help me out?

Here are the details of the .vmdk file:

-rw------- 1 root root 17179869696 Jan 5 08:54 CLUKDEVAPP1.vmdk

Here is the vdf -h output ( the vmdk file is in /vmfs/vmhba0:0:0:1):

[root@clukesxprd2 vmhba0:0:0:1]# vdf -h
Filesystem Size Used Avail Use% Mounted on
/dev/sda3 2.0G 738M 1.1G 39% /
/dev/sda1 97M 18M 74M 20% /boot
/dev/sda6 1008M 34M 923M 4% /home
none 392M 0 392M 0% /dev/shm
/dev/sda7 484M 129M 330M 28% /var
/dev/sda2 20G 294M 18G 2% /vmimages
vmhba0:0:0:1 265G 265G 978M 99% /vmfs/vmhba0:0:0:1
vmhba0:0:1:1 265G 228G 37G 85% /vmfs/vmhba0:0:1:1
vmhba0:0:2:1 265G 261G 5.0G 98% /vmfs/vmhba0:0:2:1
vmhba0:0:3:1 265G 250G 15G 93% /vmfs/vmhba0:0:3:1
vmhba0:0:4:1 265G 210G 55G 78% /vmfs/vmhba0:0:4:1
vmhba0:0:5:1 265G 161G 104G 60% /vmfs/vmhba0:0:5:1
vmhba2:0:0:9 42G 16G 26G 37% /vmfs/vmhba2:0:0:9

I have also tried to start the VM through the MUI on the host itself, without any luck.

Any Help Appreciated

Michael

ALERT on my DL360 box

Mikael Tissandier — Thu, 22 Feb 2007 11:33:47 GMT

Hi,

I installed a 2.5.4 release on a DL G1, unfortunately my VMkernel warning look like that :

Feb 21 16:35:58 srvesx02 vmkernel: 0:00:00:00.002 cpu0)ALERT: PCI: 1783: failed for 000:04.0
Feb 21 16:35:58 srvesx02 vmkernel: 0:00:00:00.002 cpu0)WARNING: Host: 3780: irq 255 is not valid

I used the HP firmware maintenance 7.4 to update all the firmware on the box and used smartstart 5.5 to accomodate the bios settings (OS type = Windows 2000)...I can't have the choice to select the full APIC on this box.

The device 0:04:0 is the smart array controleur, the version of the driver is 1.5

I can't figure out this issue!!

Thanks for your help

Mikael

VMWare DST patch

Adi — Thu, 15 Feb 2007 21:16:01 GMT

I have an ESX 2.5.0 server. It's going to be upgraded soon anyway, but I don't know for sure if that will happen before March 11, so I want to update the DST issue in the simplest way possible. I have two questions:

1. Is there any reason why updating the zoneinfo directory and /etc/localtime as on any other Linux server is insufficient? I looked at doc# 8695824, and I see a link to a "workaround" which appears to be a file containing the latest tzdata, with a Perl script to install it, but the document states:

"VMware has made update packages available that will update the time zone files only. These update packages are designed as a temporary workaround to give customers additional time to upgrade their VMware ESX servers to a supported version. It should only be installed by customers who are absolutely not able to install the full patch by March 11."

Is there anything you get from patching the OS (in regard to *this* particular issue) that you don't get from just updating the OS's time zone info?

2. Is there any reason ESX server would need to be restarted after updating the time zone info? All the guest OSes will be patched, but I'm not sure how the virtual BIOS clock is implemented, and whether it depends on the localtime settings that were in effect when vmware-serverd was started.

move converted esx3 disks back to esx 2.5.4

npeterson — Wed, 21 Feb 2007 22:49:33 GMT

Is it possible to move converted disks on an esx3 server back to a esx 2.5.4 box? I do not have virtual center.

Issue w/ Redo Logs Filling up DISK Space

itguy1984 — Wed, 21 Feb 2007 17:14:48 GMT

2950 on ESX 2.5.3?

CoreyIT — Mon, 19 Feb 2007 21:29:15 GMT

The support matrix states it is not supported until 2.5.4 but does 2.5.3 run on a 1950 or 2950 with PERC5/i?

Thanks

SUSE 10.2 problems with ESX 2.5.1 build-14182

McNail — Tue, 20 Feb 2007 13:15:41 GMT

Hi,

I had a very hard way installing SUSE 10.2 on a ESX 2.5.1 server. I've managed it now, but I can only start SUSE 10.2 in "safe mode" or with boot options "maxcpus=0 noapic brokenmodules=BusLogic". Booting without these options doesn't work at all since the BusLogic module hangs really forever. The BusLogic module sends a lot of timeouts messages and hangs. I've alread installed in past some 9.x and 10.0 SUSE workstations and servers in the same ESX server version without any problems, so I suppose that something is broken with SUSE 10.2 since the BusLogic module is always the same. Although I can boot the 10.2 SUSE, the boot process is very slow and when I am logged in the system is also very slow. SUSE 10.0 in comparison with 10.2 is a fast beast. Another strange thing: I tried to use the LSI SCSI controller instead of BusLogic but this finally resultet in a crash of the vmware console. Does anybody has similar experiences with SUSE 10.2 or maybe some hints to fix this?

Thx,
Ralf

VirtualCenter 1.3.1 - Guest Uptime field shows "0 second" for just one host

FredPeterson — Mon, 08 Jan 2007 19:05:57 GMT

As per the topic, we have one host that is showing the running guests as having a "0 second" uptime, yet obv they are all running fine. The Host uptime is accurate, but I'm obv curious as to why this would be happening and would like to fix it.

I'm guessing it is a VMWare Tools thing, but is it on the host or the guests?

Daylight Savings Time patch, 38650, doesn't seem to work on 2.5.4 32233

dante0864 — Mon, 19 Feb 2007 22:53:40 GMT

We have just installed build 38650, patch 4, on top of
a 32233 system, by booting to up, installing the
patch, then coming back up to esx.

however, when we issue the zdump command, we get the output
below - it doesn't seem to have the date correct even with the
patch on.

Anyone seen this before? Appreciate any help we can get.

Vicki
===================================
[root@esxof-001 vmware]# zdump -v /etc/localtime | grep -i 2007
/etc/localtime Sun Apr 1 09:59:59 2007 UTC = Sun Apr 1 01:59:59 2007 PST isdst=0 gmtoff=-28800
/etc/localtime Sun Apr 1 10:00:00 2007 UTC = Sun Apr 1 03:00:00 2007 PDT isdst=1 gmtoff=-25200
/etc/localtime Sun Oct 28 08:59:59 2007 UTC = Sun Oct 28 01:59:59 2007 PDT isdst=1 gmtoff=-25200
/etc/localtime Sun Oct 28 09:00:00 2007 UTC = Sun Oct 28 01:00:00 2007 PST isdst=0 gmtoff=-28800

[root@esxof-001 vmware]# vmware -v
VMware ESX Server 2.5.4 build-38650
[root@esxof-001 vmware]#

Unable to open scsi target, resource busy(2)

rccollins — Thu, 08 Feb 2007 18:11:20 GMT

I had a VM hang, maybe I shouldn't have, but I killed the PID of the vm. Now I can not start the VM because the .dsk is busy. Is there any way I can release the .dsk? I tried vmkfstools -L release however I received permission denied.
The system log is filling up with the following warning:
vmkernel: 371:17:37:21.367 cpu3)WARNING: SCSI: 5633: handle (243) 0x1564ae0 still in use by monitor (179) 2 commands

esxtop also shows:
WID
179 179 vmm 0.00 0.00 nan 57.00

Any help would be much appreciated.

Vmware DST Patch

tcantor — Tue, 20 Feb 2007 18:52:48 GMT

I see all this talk about the patch for ESX Server and the new DST. I guess my question is why this needs done? What happens if it is not done? All the Windows Servers on my ESX Server boxes are patched, why are we concerned about ESX?

VC 1.4 gives "Internal Error" alert box when adding host

glarkin — Tue, 13 Feb 2007 00:08:29 GMT

Hi there,

I'm evaluating VC 1.4, and I have successfully added a WinXP host running VMware Server with a bunch of different VMs. I have another host that's running CentOS 4.4 with a single FC6 guest, and when I try to add it to VC, I get a simple "Internal Error" alert box with an "OK" button. After pressing the button, I get no other information about what the problem is.

On the CentOS host, I see this in /var/log/vmware/vmware-serverd.log:

Feb 12 18:53:32: app| New connection on socket server-vcvmdb from host tweety.entropy.prv (ip address: 192.168.1.9) , user: root
Feb 12 18:53:32: app| SP: New user session for user: root, pos: 2
Feb 12 18:53:32: app| The vm-list file has changed! Reloading the list of registered vms
Feb 12 18:53:32: app| Set my connectionID to 192.168.1.102
Feb 12 18:53:33: app| PAM up and running.
Feb 12 18:53:33: app| SP: VPX user set to : root
Feb 12 18:53:33: app| VPXUSER is set to the user root
Feb 12 18:53:33: app| vmdbPipe_Streams Couldn't read: OVL_STATUS_EOF
Feb 12 18:53:33: app| SP: Deleting user session: 2 username: root

I believe that VMware Server has to connect back to the host that's running VC 1.4 on port 905, so I made sure that Windows Firewall wasn't blocking that port. It's not, but VMware Server appears to fail at reading something.

I've searched the forums and seen references to this OVL_STATUS_EOF error elsewhere, but haven't seen a good solution.

I tried running strace on the vmware-serverd process on the CentOS host to see if I could figure out what it was failing to read, but I didn't see anything obvious.

Is there anywhere else I can look for more detailed information about this "Internal Error" message?

Thank you,
Greg Larkin

Cancel ESX reboot?

leonwarren — Tue, 20 Feb 2007 16:23:39 GMT

Greetings...

Just a quick question hopefully someone can help me with please. If for example you issue the reboot of an ESX host by simply typing "reboot", is there a way to stop it?

It will spend some time attempting to cleanly shutdown the VMs, and during this time I can still puTTY into the box, during this time can I cancel this process or stop the reboot command? Bit of a longshot but just thought I'd ask!

Any help would be great! Thank you very much...

Trouble adding storage to ESX 3.0.1

LFMSupport — Mon, 19 Feb 2007 09:48:24 GMT

Hi guys,

I've recently re-installed my ESX server onto an upgraded HP GL380 G3.

The drive configuration shows as 2x 35Gb drives in RAID 0/1 and 3x 145Gb in RAID 5.

The ESX Server has installed onto the 35Gb drive perfectly fine, however when I try to add the additional drive I get the following error:
'Failed to update disk partition information'.

The only thing I can spot is when I see the drive in the device location screen (ESX Client) its displays a Capacity of 273.45Gb, but an available space of None.

Do I need to format it or something first? I've checked the manuals and searched here but don't seem to be able to find much.

Many thanks,

Lewis

vmAddRedo/vmCommit scripts - W2K3/SQL Server VM

sgunelius — Mon, 19 Feb 2007 21:25:25 GMT

We are preparing to shift from backing up VMs via an agent to using the vmAddRedo & vmCommit perl scripts. I'm not so concerned that I won't be able to perform a file-level restore using this type of backup because I should be able to restore the .vmdk(s) and mount them on a VM if I only need a few files/directories.

The initial testing with our BrightStor ARCserve Backup 11.x environment using the Agent for Linux looks promising, but I was concerned that one of the VMs that supports a SQL Server database might not be a good candidate for this type of backup.

I've read that this type of backup is considered crash-consistent, but it is not application aware and I'd hate to find out the hard way that some transactions didn't survive the commit process.

Can anyone comment on this? Are my fears valid and if so, should I just keep backing up this VM as if it were physical? Thank you in advance for your response.

Scott

Opensuse 10.2 on ESX 2.5.3 - vmware tools problem

sylwus — Thu, 18 Jan 2007 16:28:22 GMT

Hi all
I have problem with opensuse 10.2 - i can't install vmware tools due to differences in kernel- source and linux-kernel-headers (message during running vmware-config-tools...)

Info from my system (rpm -qa | grep kernel) :

kernel-source-2.6.18.2-34
kernel-kdump-2.6.18.2-34
kernel-xen-2.6.18.2-34
linux-kernel-headers-2.6.18.2-3
kernel-syms-2.6.18.2-34

Uname -s shows version 2.6.18.2-34-default.

Should I change opensuse 10.2 to another distro? Or maybe is there any solution for this problem?

Changing SCSI drives??

PWILSON — Mon, 17 Jul 2006 18:26:07 GMT

I'm currently running a DELL Power Edge 6850 server with 3 Maxtor ATLAS 15k 73GB drives. I running out of disk space so I figure why not remove the 3 and install 5 Cheetah 10k.7 Seagate drives?

Initialized a new array all went well. Went to install the a new OS 2.5.3 esx. After clicking next in the Device Allocation screen the following message displays
"The Partition Table on the device SDA was unreadable. To create new partitions it must be initialized, causing the loss of all data on the drive..."

I select YES

"No such file or dir during read on /tmp/sda

"An error was occurred - no valid devices were found on which to create new file systems. please check your hardware for cause of this problem."

Other than correctly setting up RAID 5 on drives in the Megaraid utility, I'm not sure what I missed. Are these drive incompatible with this version of ESX? Has anyone else gotten the same type of error? Thought or ideas greatly appreciated.

HELP???

Thanks!!

Peter

Cannot View Guest Disk Info

CDRippe — Mon, 19 Feb 2007 16:49:14 GMT

For some reason we are not able to view the disk size for any guest residing on one of our hosts within Virtual Center. We can see the disk location, just not the size. VC is reporting it as "Size Unavailable".

At first I thought it was a Virtual Center issue so I bounced xinetd & httpd. That didn't work so I then proceeded to remove and rejoin the host from VC.

No Joy.

When viewed from the MUI all the information is available and from the Service Console I can see all disk and size info. Additionally, the guests don't seem to be complaining.

While this in itself would be mildly troubling, the fact that I cannot VMotion the guests to troubleshoot is of grave concern. VMmotion fails because it cannot retrieve disk info for one or all of the guest disks.

Any ideas?

ESX 2.5.3 Build 24171
x445 (x8)
32GB RAM
NetApp Storage Array

Virtual Center v. 1.3

C. D. Rippe

Message was edited by:
CDRippe

Cannot Communicate with VMs Machine from Production Network?

habibalby — Mon, 19 Feb 2007 07:49:48 GMT

HI Guys,

I'm having the same problem with my ESX 2.5.3. I can Access the Virtual Console, I can access the Service Console. But when I build a new machine and assign this machine the same as the Production Network IP Address, I cannot ping it or access it at all.

ANy help how to get rid of this problem?

Thanks,
Habibalby

Changing .vmdk name

gatorback — Sun, 18 Feb 2007 20:58:43 GMT

I needed to change the name of the .vmdk file and was trying to change it from the MUI, with no luck (timeout error message).

Can I just make the change in the .vmx file?

I could change:
scsi0:0.name = "vmhba0:0:0:6:Visibility_3.vmdk"

to:
scsi0:0.name = "vmhba0:0:0:6:Visibility_7.vmdk"

Do I need to make changes anywhere else?
Will this cause other problems?

ESX 2.5.4 and Intel 1000/PRO PT quad port - supported? need patch?

robert_s — Tue, 30 Jan 2007 09:07:21 GMT

I have a new serverhardware (HP DL380G5 - qualified by VMware) with PCI-express and would like to use the Intel quad port PRO/1000 PT boards. I have found that there is patch for EXS 3.x e1000 driver to enable support of this board.
Is there a similar patch for the ESX 2.5.4?
The board is recognized by the ESX and lspci gives this output:
15:00.0 Ethernet controller: Intel Corporation Gigabit Ethernet Controller (rev 06)
15:00.1 Ethernet controller: Intel Corporation Gigabit Ethernet Controller (rev 06)
16:00.0 Ethernet controller: Intel Corporation Gigabit Ethernet Controller (rev 06)
16:00.1 Ethernet controller: Intel Corporation Gigabit Ethernet Controller (rev 06)

The /etc/hwconfig file:
device.21.0.0.class = "020000"
device.21.0.0.devID = "10a4"
device.21.0.0.name = "Intel Corporation Gigabit Ethernet Controller (rev 06)"
device.21.0.0.subsys_devID = "10a4"
device.21.0.0.subsys_vendor = "8086"
device.21.0.0.vendor = "8086"

We plan to migrate to VM infrastructure 2 an d ESX 3 but first in 3-6 months. In the mean time we need support for the new serverhardware and quad port PCI-xpress borads (Intel PRO/1000 PT)

Robert

Switching .vmdk from persistant to non-persistant

gatorback — Sun, 18 Feb 2007 16:58:00 GMT

When performing this task, is it necessary to shutdown the virtual machine?

Partition Re-Sizing

kerseyr — Fri, 16 Feb 2007 14:15:22 GMT

We currently have the following:

Filesystem Size Used Avail Use% Mounted on
/dev/cciss/c0d0p2 1.7G 794M 889M 48% /
/dev/cciss/c0d0p1 50M 18M 30M 37% /boot
/dev/cciss/c0d0p6 1.7G 33M 1.6G 2% /home
/dev/cciss/c0d0p5 9.6G 33M 9.0G 1% /images
none 93M 0 93M 0% /dev/shm
vmhba0:0:0:8 122G 115G 7.5G 93% /vmfs/vmhba0:0:0:8

We need more space on vmhba0:0:0:8. I would like to be able to shrink the /images partition to 4GB from 9.6GB and extend vmhba0:0:0:8 with what is left over.

Is it possible to reduce the ext3 partition down create a new VMFS partition of 5.6 GB and span vmfs over 2 partitions?

Thus creating 1 vmfs partition of 127GB.

Leaving 12GB available use able space on the vmfs partition.

Would I then be able to create a 10GB vmdk on this available space without any bother (bearing in mind this has been spanned)?

Any tips tricks guides ideas gottchas into doing this would be much appreciated.

Tia.

Rob.

VMkernel modules

fordian — Fri, 16 Feb 2007 20:55:35 GMT

Hello,

Does anyone know what the tcpip and migration modules do into the VMkernel 2.5.3 ? (vmkload-mod -l)

Thank you

Dominic

ESX 2.5.4 and VC1.4x ... is it really needed?

evanstra — Fri, 16 Feb 2007 16:44:59 GMT

Good morning,

I've been holding off on upgrading to the latest 2.5.x version because our migration to 3.x was "just around the corner" ... well it's been "just around the corner" for so long now that I want to get us up to 2.5.4p4 from 2.5.2p2. Knowing that it may be another 6 months before we get to v3.

My dilemna is that the readme for v2.5.4 states that I must be at v1.4x of VC and we are currently at 1.3x.

We have all of the support contracts/entitlements for ESX and VC... but, when I try to download v1.4 of VC the website won't let me.

I've called my sales guy...he put me in touch with the licensing people (have case number) ... they in turn tell me it's a technical problem (have case number) ... who in turn tell me it's a licensing problem.

I'm stuck: can't install 2.5.4 without VC1.4 ... can't download VC1.4 because we have not purchased a "VC entitlement for VM Server" ... don't have any VM Server in my environment (and sure don't want any either!!)

So... who is running 2.5.4 with VC1.3? Is it working? Since I have to provide all the appropriate supporting documentation to the Approval Committee prior to doing my upgrade to the environment... I'm stuck again. Documentation states I have to be at VC1.4 so therefor they won't approve my upgrade of ESX until I get to VC1.4.

AAARRRGGGHHHH!

...now I'm going to flame VMWARE on this! How can they put a restriction on ESX 2.5.4 and then deny me the ability to download the required VC when we pay out the @$$ for Platinum level support. ...plus I've been bounced around VMWare licensing, sales, and tech support for a week on this issue...

Sorry... i'm off my soap box now.

Many thanks!

Eric

Multipath Problem using Lun San disk

dvalenzuela — Fri, 16 Feb 2007 17:50:12 GMT

we have the following problem using direct LUN disk in vmware, the problem is when then SUN make a fail path to other HBA, the perfomance is low and the access is lost in random events. Don't occurs the same problem when the disk is a vmware disk. How can resolv this??

Daniel

Mixed Controllers in ESX Test server

gatorback — Tue, 13 Feb 2007 01:32:17 GMT

I have a server that meets the SCSI requirements for ESX (unsupported Poweredge 1600SC) that also has an IDE controller.

I am wondering if there is any reason why I could not install an IDE drive (200GB drive I have in storage) into the server, format it, and use it to store backups of .vmdk files.

Any insight \ caveats into making this happen would be greatly appreciated.

Deploying a Windows Server 2003 from a Template using Sysprep

schmitz — Mon, 20 Feb 2006 14:01:09 GMT

I tryed to deploy a a Windows Server 2003 from a Template using Sysprep.
Everything works fine exept of joing a domain.
I've read in several articles that there is a bug in Sysprep.
How can I fix this Problem, is there (meanwhile) any solution?

How to move a VM using VC and specify a different target LUN for each vmdk

mikeyboy — Thu, 15 Feb 2007 13:39:31 GMT

Hello, I have a VM with 2 virtual disk files, each is on a different LUN. I want to cold-migrate my VM to a different host, and have each disk file reside on different LUNs (there isn't enough room on each LUN for both disks, there is only enough room for only one disk on each LUN).

At the moment, the virtual center 1.3 migrate wizard asks me for the target LUN for the whole VM (and therefore all it's virtual disks too) rather than letting me choose where I want to put each disk separately. How do I do this? (I know I can do this manually via the MUI but I'd like my staff to be able to do this via virtual center, and not have to go into the MUI)

Thank you

Michael

Virtual Center Migrate wierdness

Fahlcor — Wed, 14 Feb 2007 21:45:45 GMT

I have been testing the Virtual Center 1.4.x with the trial license to see how easy it is to move VM's from one vmware server to another. The wierd issue i came across is the fact that it break the drive file into 2 GB chunks. Now this might be normal but when I use Virtual Center to create a new drive on a managed server it creates it as a single file. Are there some settings I'm missing? Or can someone clarify which one is done when and why. Is it possible to also set this to one or the other. Otherwise i have to use converter to combine them back together.

Thanks

Roger

Another DST question and a general update question

jbutler — Thu, 15 Feb 2007 21:30:56 GMT

I actually have two questions, both brought about by the upcoming DST issue.

I have ESX Server 2.5.3 build-28065 and understand that I could simply apply the patch here: http://ftpsite.vmware.com/download/tz-update/tz-update.tar.gz
to fix the DST issue however, this ESX installation is behind anyway, what should I apply to take me completely up-to-date (and fix the DST issue) with my current installation of 2.5.3 build-28065?

Thanks

Daylight Savings Time Patch for ESX Server 2.1.0

tcantor — Thu, 15 Feb 2007 19:53:39 GMT

Is there a patch for this version or do I need to upgrade to a newer version to get the patch? I know we originally installed 2.0 and later upgraded to 2.1, just not sure if we can patch what we have to need to upgrade before we can do that? If we need to upgrade, what version do we need to upgrade to?

Thanks

Tom

VMware ESX Server 2.5.3 doesn't support Inter PRO/1000 EB

krasen — Thu, 15 Feb 2007 18:58:32 GMT

Hi. I have a serious problem with installing VMware ESX Server 2.5.3. In the very beginning of the installation it shows a message that my Ethernet controllers are not supported. The motherboard is ASUS DSBV-D and it has two Intel PRO/1000 EB controllers on-board. I will be very thankful for any solution.

Scheduled affinity deployment

foster29 — Thu, 15 Feb 2007 17:08:57 GMT

I have a dell 2850(2 processor) server that runs 2 VMS, one client and one server. The server does all the heavy lifting and has our web servers, relational dbs, etc. The client is simply used to run IE off of.

My question is current my affinity is setup for both vms to run across both processors. To optimize performance should the server be running on Processor 1 and the client run on server 0.

thanks,
pete

ESX server halts once in a week!!

pwcnash — Wed, 05 Jul 2006 14:18:38 GMT

Hi
I am running ichain on one of my ESX server. I see my ESX server crash every week on the weekends. I could only restart the entire machine to overcome the crash. Could any one please help me on this?

I see the following line in the messages file under /var/log

Jul 4 04:02:00 ICHAINESK httpd.vmware: httpd shutdown succeeded
Jul 5 04:02:01 ICHAINESK httpd.vmware: httpd shutdown succeeded

ESX server caused the vmware heartbeat to go from green to yellow to red!! around 11:44 pm.

Is there anything wrong with the processes listed in cron?
Now the status is disconnected. And I am unable to connect as it says: Server agent not responding

please help.

Nash

Message was edited by:
pwcnash

Broadcom failure on x366

Johan Lundqvist — Tue, 26 Sep 2006 07:27:20 GMT

I have problem, again. The threee x366 we have are really keeping me busy .

A little background:

The server has one two-port gigabit adapter onboard (Broadcom 5700). In slot 1 there´s a PCI card (Broadcom 5700). The two ports on the on-board card is dedicated to the service console and vmotion. The two ports on the PCI card is (at least should be) dedicated to the vms.

Problem Description:

After a upgrade to 2.5.3 patch 3 I discovered that my virtual switches where gone (all exept the vmotion switch and the internal switch).

I checked under "Psysical adapter" and found........nothing. The PCI adapter was gone. I rebooted the server and the adapter didn´t even show in the setup. I pulled the card out and then back in, started the server. I could now see the card during the system startup. I ran vmkpcidivy -i and found the adapter again. But it was dedicated to the service console. I changed to "Virtual machines" and rebooted the server.
Then the problem accours again. The adapter disapears.........

If I move the adapter to another slot it works like a charm.....

Loss network connection

mfercor3 — Thu, 15 Feb 2007 08:42:39 GMT

Hello,

I have several clusterÂ´s with windows 2003 sp1 and everydays i have events reporting loss of internal network connection. I have an internal switch for the internal connectionÂ´s (heartbeat/intracluster) between the cluster nodes. Someone have any idea why his happend? Maybe i need one internal switch for each cluster!!!?????

I have ESX 2.5.2

Maybe a patch is necesary...

If someone knows please help me...

Moving VM copy speed...

dimsys — Thu, 15 Feb 2007 03:12:51 GMT

Hello all,

I had to move a VM from one ESX server to another last night. We don't have a SAN, each ESX 2.5.3 server has it's own direct attach storage so my plan was to use ESX Ranger Pro.

We have an internal GigE utility network that a Windows 2003 server (that I access the MUIs from and has Ranger on it) and the service consoles of each ESX machine are on (192.168.0.x).

I shut down the VM to move and performed a backup using Ranger Pro. It worked fine, and created a compressed file of the VM on the Windows server as expected. It wasn't super fast, but not bad-- the VM has a 40GB .vmdk which compressed down to around 6GB if I recall.

OK so far so good. I then kicked off a restore from Ranger to the alternate ESX server. It was taking forever, so I bailed on it.

Next I tried to scp the .vmdk from the one machine to the other. It was brutally slow, like 3MB/s. As it copied, I was scouring these forums and read someone saying not to copy a .vmdk with scp and to use the proper import process, etc. so since it was painfully slow I bailed on the copy and went back to the Ranger restore method.

3.5 hours later Ranger completed the restore and did a nice job of also copying the .vmx, registering it, and luckily I set it to bring the VM back online because by now it was almost 5:30AM and I had passed out waiting.

All the NICs are GigE, the switch is GigE, I checked and all the NICs are reporting themselves to the switch at Gig speeds-- but it's performing worse than 100BaseT.

If anyone has any tips on things I can try to troubleshoot this, I would appreciate it. I'm not sure where to start-- it would be great if there was a way to test the NIC in the ESX box used for the service console to eliminate it as a variable.

Thanks for any tips...

Are upgrade patches cumulative?

JohnInBoston — Sat, 10 Feb 2007 03:36:54 GMT

I'm pretty sure that the answer is no, but:

I untared 2.5.4 upgrade 4 it appears that all of the RPMs from upgrade 3 are included in the "/VMware/RPMS" directory. If the upgrade patches are not cumulative, then why include 38 MB of redundant RPMs?

Note that this is not the case for the "drivers" directories... It seems that the upgrade patches are partially cumulative. Why is that?

esx 2.1.3 won't use my Intel pro/1000 nics

darthghandi — Tue, 13 Feb 2007 21:26:38 GMT

I am installing esx 2.1.3 on a machine with 4 nics. 2 are intel pro/1000. When installing, I am told that these nics are not compatible. According to the minimum server requirements, these nics should work. Anyone have any idea how to force them?
Thanks for your time.

PCI devices

fordian — Wed, 14 Feb 2007 15:33:17 GMT

Hi,

ESX 2.5.3

Is there a means to have more than 6 pci devices into a vm ?

Thank you

Dominic

SAN I-O wierdness

Tom Swigg — Wed, 14 Feb 2007 17:46:29 GMT

BL25p running VMware ESX Server 2.5.3 build-24171
using QLogic Corp QLA231x/2340 (rev 02) HBAs connected to hp EVA4000

This one is really odd. It manifested itself initially with our backup procedure but is now reproducible as shown below.

We were trying to tar up vdisks from a VM when one of them generated an error.

/var/log/vmkwarning.4:Jan 20 21:42:52 br-bl-one-2 vmkernel: 188:01:35:14.417 cpu1)WARNING: FS2: 5452: SCSI read error 0xbad0004

which was seen in the COS as an I-O error.

The file concerned was triton.triton0-s002.vmdk and experiment showed that any read operation at the COS using cp, dd etc would cause an I-O error at the same place in the file.

So I changed the preferred path to use the other vmhba and the read worked.

So I set up another lun to see if this was repeatable and it is.

# vmkmultipath -q vmhba0:0:18
Disk and multipath information follows:

Disk vmhba0:0:18 (102,398 MB) has 4 paths. Policy is fixed.
vmhba0:0:18 on (active, preferred)
vmhba0:1:18 on
vmhba1:0:18 on
vmhba1:1:18 on

when vmhba1 is the active HBA there is no problem.

For vmhba0 as the active controller I tried to copy triton.triton0-s002.vmdk using dd or cp and the destination file was the same size in each case when the I-O error occurred

# ls -l

-rw------- 1 root root 2143354880 Jan 30 17:21 triton.triton0-s002.vmdk
-rw------- 1 root root 1716518912 Feb 14 16:36 tt
-rw------- 1 root root 1716518912 Feb 5 15:56 xx
-rw------- 1 root root 1716518912 Feb 5 16:26 yy
-rw------- 1 root root 1716518912 Feb 5 16:48 zz

# cp -i triton.triton0-s002.vmdk vv
cp: reading `triton.triton0-s002.vmdk': Input/output error

# ls -l vv
-rw------- 1 root root 1716518912 Feb 14 17:54 vv

with loads of

Feb 14 17:53:46 br-bl-one-2 vmkernel: 212:21:42:33.977 cpu1)WARNING: SCSI: 5663: vmhba0:0:18:1 status = 0/2 0x0 0x0 0x0

in /var/log/vmkwarning followed by

Feb 14 17:53:46 br-bl-one-2 vmkernel: 212:21:42:34.028 cpu3)WARNING: FS2: 5452: SCSI read error 0xbad0004

and yet

# ls -l demo.vmdk
-rw------- 1 root root 10737418752 Feb 14 15:36 demo.vmdk
# cp demo.vmdk ww

# ls -l ww
-rw------- 1 root root 10737418752 Feb 14 18:11 ww

works like a dream although there are loads of
Feb 14 18:11:07 br-bl-one-2 vmkernel: 212:21:59:54.792 cpu1)WARNING: SCSI: 5663: vmhba0:0:18:1 status = 0/2 0x0 0x0 0x0

It is almost as though I have copied an I-O error from one lun to another!

Note: the file triton.triton0-s002.vmdk was successfully copied from the original lun on to the test lun(18) by switching the original lun(12) to also use vmhba1

Disk vmhba0:0:12 (819,197 MB) has 4 paths. Policy is fixed.
vmhba0:0:12 on
vmhba0:1:12 on
vmhba1:0:12 on (active, preferred)
vmhba1:1:12 on

This lun is only presented to this blade and yet I am getting HBA busy errors (0/2) on reads. When copying to this lun I do not get (0/2)s

Any ideas?

Adding Processor Licenses to ESX 2.5.4

jckelsey — Tue, 13 Feb 2007 19:57:20 GMT

Single ESX 2.5.4 server with 4 processors and SMP. I've added 4 more physical processors so of course, I need another license before the system will recognize all 8 processors.

I have the .lic file that support says I need, but....no license server. How do I get the 8-proc key? Something support needs to do for me?

Thanks!

using CRTL/ALT/BREAK

foster29 — Wed, 14 Feb 2007 12:38:59 GMT

Does anyone know how to turn the key stroke CRTL/ALT/BREAK into a icon that the user can just click on?

Virtual Switch issue

bradyrandolph — Tue, 16 Jan 2007 10:03:42 GMT

Hi-

I am having a problem with one of my virtual switches. I was trying to vmotion to a server but the network the guest was on wasn't configured on the destination server. When I went to add the network via MUI, it says that it already exists but it doesn't show this in the MUI. This server is 2.5.1.

Has anyone run into this same issue?

Thanks in advance,

Brady

Unable to start Virtual center Server-Failed to connect to external DB

pwcnash — Tue, 13 Feb 2007 21:28:43 GMT

Hi all,
I am using virtual center 1.3.1 on win 2003
I am unable to start the VMware server all of a sudden. We use MSDE table as teh database to store information and use ODBC to connect to the database. I remember changing teh password for the sa account in MSDE table. Now I am not able to connect to the server as it cannot be started.
Please help!! This is very urgent!!

Nash

Message was edited by:
pwcnash

Port Group Issue

bigusdadius — Tue, 13 Feb 2007 13:46:36 GMT

I have customer who has and existing ESX 2.5.2 server that we have setup a new switch with VLANs presented to it and we cannot get any VM to talk to the VLANs. I sent the network engineer the VMware VLAN WP and he states the VLANs were setup accordingly.

There was an existing virtual switch connected to their local network which the VMs are currently using. We have added an additional NIC, created another vswitch and added the necessary port groups and when I switch the VM to one of those port groups I can send packets, but not receive any...

Just to be sure, I setup a ESX 2.5.4 server in the same manner, tested it and receive the same results. I believe its the network, but I'm having a difficult time nailing down the issue...

Any feedback will be greatly appreciated. Thanks!

VC 1.3.1 and Thin Provisioning

hortencio — Tue, 13 Feb 2007 13:46:39 GMT

Hi,

I have a VC 1.3.1 and ESX 2.5.4 running and attached to a storage IBM N-Series. We are attempt to use Thin Provisioning for storage overallocation.
Thin Provisioning only works when we manipulate the VMDK files using the Service Console. But Thin Provisioning is not working when we perform any operation by using VC. e.g.: If we deploy a VM from a Template using VC, the storage allocates the role space of the VMDK file while it should be allocating only the used space (that's what Thin Provisioning does). So as a workaround we are copying (using service console) a standard VMDK file to the Storage and running Sysprep (by this way Thin Provisioning works), but this is many time consumimg and not friendly.

Someone have any idea of how to make the Thin Provisioning to work with VC ? Any idea if by upgrading the VC to version 2.x it will resolve the problem ?

ESX 2.5.3 how to connect

clx-it — Tue, 13 Feb 2007 14:25:44 GMT

hi
we have an old esx 2.5.3 server for testing. now i have the problem that i can not connect with the new Virual Infrastructure 2.0.1 Client.
is this possible, or have i to install an older one ?

thx 4 u help

Need to install compat-libstdc++

morrisew — Fri, 15 Dec 2006 22:45:18 GMT

I have a PowerEdge 2950 and would like to install the diagnostic utilities on the host O/S. I am running ESX 2.5.4. The utility install is complaining that libstdc++.so.5 does not exist. I believe this is in the compat-libstdc++ package. Is there a compat-libstdc++ RPM package available for this os?

Netware 4.11 on ESX 2.5.2

simon.l — Mon, 12 Feb 2007 14:54:59 GMT

Hi,

Can anyone tell me what type of adapter the buslogic and lsi logic controllers in 2.5.2 are based on? I've been trying to find a drive for Netware 4.11 and I'm slowly losing the will to live. HEELLLPPP!!

The reason for the need to run 4.11 on Netware is our currently aging Novell server is having a few problems and we need a standby. We cannot upgrade to Netware 6 either.

Kind regards

Si

I Can't start one VM

Galew — Fri, 17 Feb 2006 07:40:53 GMT

Hi,

here is what happened:

User did something wrong on VM. I take remote console session to look what was wrong and the session crashed. I was kicked out of graphical interface (restart xinetd or httpd.vmware didn't help) so I was forced to kill -9 vmware-serverd. Then the graphical interface worked and all other server were up and running, only the one was down. I tried to start it but it won't start. When I tried to start it via ssh at the service console there was an error message:

[root@mycomp]# vmware-cmd /vmdata/mycomp/mycomp.vmx start
/usr/bin/vmware-cmd: Could not connect to VM /vmdata/mycomp/mycomp.vmx
(VMControl error -14: Unexpected response from vmware-authd: Error connecting to /usr/lib/vmware/bin/vmware-vmx process.)

How can I start the machine? System is esx 2.1

Removing RAM from VM

gatorback — Tue, 13 Feb 2007 01:46:52 GMT

I had a webserver that was slow, and thought that it did not have enough RAM. I added RAM from 512 to 1280 MB, however this was not the reason it was slow. The root cause was corrected and now I am interested in backing down the amount of RAM.

I will be using the MUI to see if there is room to backing out the amount of RAM. If there is room and I back it down, what are the pitfalls and how can they be mitigated?

lost VC connection

UFS — Mon, 12 Feb 2007 20:51:23 GMT

Lost the VC connection to one of my hosts (2.5.3). I can't get a MUI connection. All guest are still running and I am able to get to the console.

The vmware-serverd service does not show as running, but I am unable to start or restart it.

Solution: Corrupted REDO files after power failure

ramschel — Mon, 12 Feb 2007 21:03:18 GMT

All,

After having a power outage and having several VM's that were in undoable mode report corrupted REDO files, I searched these forums for a possible fix before accepting that all of the changes in the REDO files would be lost. I didn't find an solution that worked for me already listed so I tried something on my own that did turn out to work. Apologies if this was already listed by someone else in another thread. I view VMTN as a valuable resource and this is my attempt to 'give back' by helping someone else that might have a similar problem.

To recover an ESX 2.5.3 VM in Virtual Center 1.3.1 with a corrupted REDO file:
1. Cancel the dialog if still up and DO NOT DISCARD the redo.
2. Optionally copy the main disk file(s) to another location for backup in case the next steps corrupt the main disk files.
3. With VM powered off, go to the drive properties and change to persistent.
4. Power on VM and answer question to commit.
5. This will take a long time, longer if the REDO log is very large. It may take so long that Virtual Center times out and says that the âpower operation filedâ but just say OK and the ESX Host will keep working on it.
6. If the VM had more than one disk in undoable mode, you will then be prompted to commit each one with a long delay between.
7. Be patient and wait it out and you should have a booted VM with the committed changes.

Hope it works for you as it did for me.

Mirror in DR configuration

LucaR — Mon, 12 Feb 2007 17:17:21 GMT

Hello,
I have a poor man's DR configuration as following:
Guest OS with 2 disks presented to it on 2 different LUNs, mirrored.
DR instance on a different ESX host, with one of the 2 disks from Guest OS
presented to it.
When Guest OS is down, the DR instance will not boot from the mirror side presented to it====>Failed to Power on VM: operation failed to change the VM to the expected power state.

Any idea will be aprreciated.
Tahnk you.

ESX 2.5 server inaccessible after reboot

morrisew — Mon, 12 Feb 2007 17:35:45 GMT

Hello,
I had to reboot my ESX server and am now having trouble accessing it remotely using SSH. I cannot even ping it at this point. I have had this same issue with two other ESX servers in the past after reboot. The only way I can access a server remotely after a reboot is if I connect to it from another server on the same subnet or if I log in at the local console. Once I do either of these, I can then access the server(s) from other subnets. My problem with this server I just rebooted is that I don't have access to another server on the same subnet and it's not easy to get access to the local console due to physical security constraints. I have asked other VMWare guru's that I work with and they have never had this issue. Has anyone ever seen this?
Thanks

Failed for vmhba1:0:16 status = 195887114

rodria — Fri, 09 Feb 2007 11:44:23 GMT

I'm getting this error in /var/log/vmkwarning every second:

Feb 9 12:22:09 esxserver6 vmkernel: 86:00:26:23.390 cpu0)WARNING: SCSI: 6710: Failed for vmhba1:0:16 status = 195887114
Feb 9 12:22:09 esxserver6 vmkernel: 86:00:26:23.391 cpu1)WARNING: SCSI: 5657: vmhba1:2:16:0 status = 2/0 0x5 0x25 0x0
Feb 9 12:22:09 esxserver6 vmkernel: 86:00:26:23.391 cpu0)WARNING: SCSI: 8754: returns error: "I/O error". Code: 0xbad000a.

Do you know why?

My /var filesystem is full:
[root@esxserver6 root]# df -k
Filesystem 1k-blocks Used Available Use% Mounted on
/dev/sda2 6048352 1876356 3864756 33% /
/dev/sda1 101089 17953 77917 19% /boot
/dev/sda8 1510032 34144 1399180 3% /home
none 192280 0 192280 0% /dev/shm
/dev/sda6 4538124 32836 4274760 1% /tmp
/dev/sda9 1510032 550288 883036 39% /usr
/dev/sda7 1510032 1435052 0 100% /var

VMkernel swap space query

lesw — Mon, 12 Feb 2007 12:06:12 GMT

Hi

Just looking at one of our hosts, One of the VMs is showing in the MUI as having swapped out to the VMkernel swap space.

However, the host is showing 7.1GB of free physical RAM. As VMkernel swap is supposed to be last resort, why should this be. There is no ballooning shown either, which I though should be used first.

Any thought?