VMware Communities : Thread List - Availability: HA & FT

Search the VMware Compatibility Guide

juanin00 — Thu, 19 Nov 2009 11:51:31 GMT

Hello,
i want to buy 2 HP PROLIANT DL580 G5 6-CORE INTEL XEON E7450 PROCESSOR 2.40 GHZ. That processor is compatible following the vmware kb 1008027 document. But if we look inside the "Search the VMware Compatibility Guide" and we put that server it says that is not compatible with FT.

therefore , my future server will be compatible or not?

it seems there is a contradictory on it.
Thanks a lot.

vSphere Fault tolerant ? is it possible to have the primary and secondary virtual server on different physical storage ?

chrisAMS — Wed, 18 Nov 2009 12:03:51 GMT

my idea is to get 2 SAN and make the primary on the 1st SAN, and the secondary on the 2nd SAN. In this case if 1 SAN failed the FT still run.
Is it working ?

Can I enable VMWare-FT on vCenter VM?

habibalby — Sat, 14 Nov 2009 11:25:17 GMT

Hello,

Can I enable MWare-FT on vCenter VM?

If yes, is it supported by VMware?

Is there any special recommendation i have to take aginst vCenter?

Best Regards,

Hussain Al Sayed

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

HA disconnected VM network?

omatsei1 — Mon, 26 Oct 2009 17:48:52 GMT

Over the weekend, we had a serious power-related problem in our data center, which caused a few of our ESX servers to shut down. Fortunately, HA kicked in and brought all those VM's over to the unaffected ESX servers and started them back up. Unfortunately, it disabled the network on those VM's for some reason. Specifically, we noticed that after the VM's were started, our monitoring software was still showing them as offline. Upon investigation, we found that every VM that had been HA'd to a different host had it's network adapter disconnected (in the VM settings, when you select the network adapter, the top box saying "Connected" was unchecked). For a few VM's, that wouldn't be a huge problem, but with the 70-80 VM's that failed over this weekend, it became a huge ordeal to figure out which ones were working and which weren't...

Does anyone have any idea at all about how that checkbox was unchecked, and how to prevent that from happening in the future?

FT prerequisite: two NIC or two portgroup

dingding — Mon, 19 Oct 2009 07:35:49 GMT

FT logging and Vmotion need two separate subnet network, can i use two portgroup in a vswitch with only one pNIC or i must use two vswitch, each for FT logging and Vmotion.

Issues with Upgrading an ESXi 3.5 host to 4.0 in a cluster with other 3.5 hosts

acoustix — Sat, 07 Nov 2009 00:23:35 GMT

I think I'm losing my mind. Here is my scenario: I originally had three ESXi 3.5 U4 hosts in a cluster managed by virtualcenter 2.5. A few weeks ago I did a successful upgrade on VC to version 4.0. All of the configs stayed the same. All of my network adapter/virtual switches were configured the same on all hosts:

vSwitch0 / VMkernel Port / Management Network / vmk0: 192.168.1.x (16 - vmsvr1, 17 - vmsvr2, 18 - vmsvr3)

vSwitch1 / Virtual Machine Port Group / Virtual Machine Network

vSwitch2 / VMkernel Port / iSCSI VLAN 25 / vmk1: 192.168.25.x (13 - vmsvr1, 14 - vmsvr2, 15 - vmsvr3)

vSwitch3 / VMkernel Port / iSCSI VLAN 26 / vmk2: 192.168.26.x (13 - vmsvr1, 14 - vmsvr2, 15 - vmsvr3)

Here is my IP scheme for the physical ports on the servers: Nic1: 192.168.0.x - virtual machines. Nic2: 192.168.1x - Management network, Nic3: 192.168.25.x - iSCSI VLAN, Nic4: 192.168.26.x - iSCSI VLAN. Both iSCSI VLANS have access control lists that do not allow traffic to leave or enter the VLAN. In other words you cannot access VLAN 25 or 26 from any other network like 192.168.0.0 or 192.168.1.0.

Yesterday I evacuated all of my VM's from my first ESXi host: vmsvr1. (my other hosts are vmsvr2 and vmsvr3) I put the host into maintenance mode and moved the host outside of the cluster. I then deleted this host from VC. I did a fresh install of ESXi (embeded) onto my Dell PowerEdge R805 server. After configuring the management interface exactly like the original server I configured the network settings on vmsvr1 exactly the same as above. I did all of the other configs the same as well (scratchconfig, iSCSI SAN, etc). I added the host vmsvr1 to the cluster on my VC. Then got an error when configuring the host for VMware HA:

Host vmsvr1 does not have the following networks used by other hosts from HA communication: 192.168.25.14,192.168.26.14. Consider using HA advanced option das.allowNetwork to control network usage.(FYI: those addresses are my iSCSI adapter addresses for vmsvr2 - are not NOT configured for HA comminication)

Host vmsvr1 has the follow extra networs not used by other hosts for HA comminication: 192.168.1.16. Consider using HA advanced option das.allowNetwork to control network usage. (this IP address is my management network along with 192.168.1.17 for vmsvr2 and 192.168.1.18 for vmsvr3)

Now the REAL kicker is this: When I go and look at my network configs for vmsvr2 I noticed that it changed from the config I posted above. Vmsvr2 now looks like this:

vSwitch0 / Virtual Machine Port Group / Management Network

vSwitch1 / Virtual Machine Port Group / Virtual Machine Network

vSwitch2 / VMkernel Port / iSCSI VLAN 25 / vmk1: 192.168.25.14

vSwitch3 / VMkernel Port / iSCSI VLAN 26 / vmk2: 192.168.26.14

What the heck is going on here? I did not change anything on vmsvr2. Even thought it shows vSwitch0 as a VM port group with no IP address I'm still able to manage the machine as if it were a VMkernel port with an IP address assigned to it. This is why my HA config is bombing on vmsvr1. What is causing vmsvr2 to change on it's own and how can I change vSwitch0 back to a VMkernel port with an IP address without messing everything up?

Thanks in advance,

Nick

HA not function with different generation of processor

siungnc — Wed, 18 Nov 2009 07:45:56 GMT

Hi All,

Current we having two types of processor in a cluster group which are intel xeon E5440 and intel xeon X556. VMotion able to work properly after EVC has been configured. But the problem here is when I tried to shutdown anyone of the ESX host, the VM not able to startup in other ESX host in the same cluster group. Anyone have any idea on what is happening ? Processor not compatible ?

Thanks

RDM support for Fault Tolarent in vsphere 4

Ramji — Sat, 24 Oct 2009 04:59:59 GMT

In vsphere 4 does FT is possible in RDM vm ie A VM has got a rdm as D drive can this be made to Fault Tolerent enabled VM

Regards'

R.Ramji

VMware HA error on vSphere and ESX 4.0

adeelleo — Fri, 05 Jun 2009 08:07:31 GMT

I am having trouble enabling HA on the new ESXi 4 and vSphere 4
environment on my IBM BaldeCenter S environment. Everything was working
perfect untill i activated DPM on aggressive mode. DPM powered down all
hosts except one ruuning all the VM's. Now since then the HA agent is
not configuring properly.

I receive the following errors when i try to enable HA on the cluster.

HA agent has an error : cmd addnode failed for
secondary node: Internal AAM Error - agent could
not start. : Unknown HA error
error
5/27/2009 1:59:51 PM
Administrator

HA agent has an error : Cannot complete the HA
configuration
error
5/27/2009 1:59:51 PM
Administrator

I have tried everything from disabling and re-enabling HA, Creating a
new cluste and DataCenter, entering hosts in maintainance mode and then
exiting, disconecting the hosts and removing then re-adding, installing
frest ESXi and vCenter servers, changing network parameters,
reconfiguring for HA.

I have manually edited the host file and entered the entries for all
esx hosts. All esx hosts and vCenter can ping each other through IP,
FQDN and short names.

Any help would be highly appretiated. I am totally out of options.

Best regards,

Adeel Akram

HA Isolation Response difficulties

Sven Vollmann — Tue, 10 Nov 2009 11:53:10 GMT

Hello everybody,

we got a vSphere Cluster
with 2 Clusternodes.

The problem: We added a new
physical switch to our network, which causes a renewal of the spanning tree
algorithm. The consequence was, that HA on each host meant that the hosts were
isolated, so that all VM´s shut down.

The Question: Are
there any parameters which causes a shut-down delay for the vm´s, so that
theres enough time for HA to get the heartbeat again... which again wouldn´t
cause HA to shutdown the VM´s.

If not, we thought about a
alternative solution:

We thought about a second
Service Console (new vSwitch) on the Hosts, which is attached to a separate
physical switch, which is totally isolated from the main-network (view
attachment). Do you think that will work? Wouldn´t that may be the better solution to our problem?

thanks for answers

sincerely yours

Daniel S.

[
|file:///X:/GAMES/Risen-THEPiRATEGAY/tpg-rsn.r80]

HA Isolation Response difficulties

Sven Vollmann — Tue, 10 Nov 2009 11:44:27 GMT

Hello everybody,

we got a vSphere Cluster
with 2 Clusternodes.

The problem:

We added a new
physical switch to our network, which causes a renewal of the spanning tree
algorithm. The consequence was, that HA on each host meant that the hosts were
isolated, so that all VM´s were shut down.

The Question:

Are there any parameters which causes a shut-down delay for the vm´s, so that
theres enough time for HA to get the heartbeat again... which again wouldn´t
cause HA to shutdown the VM´s.

If not, we thought about an alternative solution:
We thought about a second Service Console (new vSwitch) on the Hosts, which is attached to a separate
physical switch, which is totally isolated from the main-network (view attachment). Do You think that would that work? Wouldn´t that may be the better solution for our problem?

Sincerely yours
D. Siebers

HA Isolation Response difficulties

Sven Vollmann — Tue, 10 Nov 2009 11:41:41 GMT

Hello everybody,

we got a vSphere Cluster with 2 Clusternodes.

The problem: We added a new physical switch to our network, which causes a renewal of the spanning tree algorithm. The consequence was, that HA on each host meant that the hosts were isolated, so that all VM´s shut down.

The Question: Are there any parameters which causes a shut-down delay for the vm´s, so that theres enough time for HA to get the heartbeat again... which again wouldn´t cause HA to shutdown the VM´s.

If not, we thought about a alternative solution:

We thought about a second Service Console (new vSwitch) on the Hosts, which is attached to a separate physical switch, which is totally isolated from the main-network (view attachment). Would that work? Wouldn´t that may be the better solution for our problem?

Thanks for answers,

sincerely yours

D. Siebers

one cluster or two?

russ79 — Sat, 07 Nov 2009 02:58:27 GMT

in a nutshell my current enviornment:

1 Intel Cluster of 5 hosts (core 2 and i7) for production, attached to production san
1 Intel Cluster of 2 hosts (core 2) for development, attached to development san
4 Standalone Intel hosts (core 2) some for production, some for development
6 Standalone AMD hosts (older opteron) for development

I am beginning the process of upgrading to vSphere and I've made the suggestion of pooling all of our Intel boxes into 1 cluster on the production san. We have enough capacity to retire our AMD boxes. I could use resource pools and vlans to keep production and development VMs separated. I am looking for any thoughts and comments on this plan. I think this might be a good idea but i'm curious to know if anyone thinks otherwise.

Fault Tolerance - secondary VM consumes large amount of memory?

jameran — Sun, 11 Oct 2009 23:54:37 GMT

Hi,

I'm experiencing a Fault Tolerance problem with a pair of new servers using X5570 processors. The two ESX hosts are 4.0, updated with VUM as of October 11th (today). Whenever I create an FT enabled guest, the secondary guest experiences very high RAM utilization.

As shown on the attached screenshot, the secondary guest shows 88+ % memory utilization, even when the primary VM shows only 27% memory utilization.

On the forums, I've only found one other instance of this error, but no solution or explanation yet. Has anyone determined the cause or fix to this? I'd appreciate your feedback.

Thanks!

Long delay between host not responding and moving VMs

jleyton — Fri, 06 Nov 2009 18:22:28 GMT

We had an issue with one of our esx4 servers last night where it appeared to panic and went unresponsive. According to the cluster even log, it took exactly 30 minutes (to the second infact) from the time that the host was marked as not responding to when HA kicked in and moved the VMs to another host in our cluster. What settings should I look into why it took so long for this to happen? And is there any significance to the round number of minutes? We have our VM monitoring sensitivity set a little low, but that should only account for 2-3 minutes.

HA settings for the cluster as follows: enabled host monitoring, allow VM to power on if it violates availability, high restart priority for all VMs, enabled VM monitoring, VM monitoring sensitivity is low

Thanks for any ideas you might have

backup Vsphere setup

anilchaurasia — Fri, 30 Oct 2009 19:22:04 GMT

HI All,

I have two vSpher server install and two oracle database for each server on two diffrenet Windows 2003 server.

I want to setup a backup vSpher setup using the second server + DB.

IS this possible ?

Is there any way I can imeplemet that. What are all the things I need to take carre of ?

I am using vSphere v 4.0 with Oracle 10g Db on windows 2003 server.

Thanks

Anil

HA Errors, where do I start to trouble shoot?

theblackknight — Sun, 01 Nov 2009 14:17:24 GMT

Im getting the all too familiar HA errors when I try to enable. HA error indicating the HA agent...

I've cheched the following and all is good...

Check to make sure DNS is configured properly Check to see if you can resolve DNS Check DNS records Make sure you are using FQDN's Make sure your using lower case Check Service Consoles have the same names and networks Disable and Re-enable HA Select Reconfigure for HA on the ESX host
What else am I missing?

HA & DRS License ?

barber50701 — Thu, 29 Oct 2009 03:24:27 GMT

I have a cleint that is interested in the HA and DRS options but does not want to spend the extra money for VMotion.

My question is Doesn't HA use VMotion to work? Do they need a VMotion License? I am confused on what to sell them

Nick

Clarificaiton on Fault Tolerance

jeffoutwest — Wed, 28 Oct 2009 02:27:44 GMT

Can someone help clarify the following:

vShpere 4.0, ESXi 4.0.

When a guest OS has been configured for FT running in lock step on two hosts, what happens if that guest OS system suffers from a server panic, or the appication freezes or crashes. Does the guest machine on the secondary host also display the same conditions? Thanks, -Jeff

can HA happen between different version of ESX host

inforhunter — Wed, 28 Oct 2009 15:01:21 GMT

I have two ESX server host,one is ESX 3.5 and the other is ESX 4,both of them are under the management of a VCenter server,

I wonder can I implement HA between the two ESX hosts

Cron job(scheduled task) for Vsphere...

vspherercp — Wed, 28 Oct 2009 15:41:40 GMT

Trying to setup a cron job to run at 12am every morning on each of our esx hosts. All I have to do is run the command "config_mpath --primary --persistent"

Can anyone point me in the right direction...syntax?

Thanks

HA Host Isolation Question

nicholas1982 — Thu, 15 Oct 2009 07:03:52 GMT

Hi Guys,

I have a HA cluster with 3 ESXi 4.0 hosts. Each host has 6 NICS and 3 vDistributed Switches.

vDS 1 has 2 nics in a nic team and has a management port group configured this is the main management used for connecting to VC and can be fully resolved by DNS with FQDN

vDS 2 has 2 nics in a nic team I use this for backups and monitoring, I have also backup management port group configured just in case something goes wrong with the main management port group, I would be able to connect directely to the host via IP.

vDS 3 has 2 nics in a nic team and i use this just for vmotion traffic.

I wanted to test HA, so i disconnected one of the ESX servers uplinks for vDS 1 and sure enough that host disconnected from vCenter, but the vm hosted on that ESx host never shut down to be then powered on in on another host, however when I then disconnected the uplinks for vDS 2 the VM restarted on the other hosts in the cluster. So I guess it heartbeats through the other management port group on vDS 2 as well, but because vDS 1 is the production Network I need the host to isolate only if vDS 1 network is down, I have tried some of the advanced HA configuration without any success, has anyone got any ideas?

Can configure HA but it doesn't move VMs`

bfredette — Mon, 27 Jul 2009 18:14:32 GMT

Greetings,

I have 3 ESX hosts configured with HA in VC. No errors are reported during the configuration. During the tasks, at one point I see an error about not having enough resources in the cluster ... so it would seem that HA is at least monitoring resources. The problem is, when I reboot any server, the VMs do not move. After a minute or so the server shows disconnected with the VMs powered on and unknown state. There is nothing in the events & task panel in VC. No sign the HA agent even went down. I tried looking through the vpx and aam logs but I'm not exactly sure what I'm looking for. There looks like what appears to be errors with ft_gethostbyname but I'm not sure if this is a problem as DNS checks return just fine.

From the vpxa.log:

~~2009-07-27 13:49:51.056 0xf6fa0b90 verbose 'App'~~ VpxaVMAP::InvokeCommand to invoke is /opt/vmware/aam/bin/aamPerl /opt/vmware/aam/ha/aam_config_util.pl -z -shortname=usbvt10pesxhost02 -uname=Linux -cmd=listnodes -domain=vmware
~~2009-07-27 13:49:51.608 0xf6fa0b90 verbose 'App'~~ VpxaVMAP::Invoke task percent done is 100
~~2009-07-27 13:49:51.610 0xf6fa0b90 verbose 'App'~~ VpxaVMAP::Invoke Command output: aamPerl script

Setting environment from /opt/vmware/aam/config/agent_env.Linux

07/27/09 13:49:51 print_args KEY: cmd VAL: listnodes
07/27/09 13:49:51 print_args KEY: -z VAL: 1
07/27/09 13:49:51 print_args KEY: uname VAL: Linux
07/27/09 13:49:51 print_args KEY: shortname VAL: usbvt10pesxhost02
07/27/09 13:49:51 print_args KEY: domain VAL: vmware
07/27/09 13:49:51 issue_cmd CMD: /opt/vmware/aam/bin/ft_gethostbyname usbvt10pesxhost02 |grep FAILED
07/27/09 13:49:51 issue_cmd STATUS: 1
07/27/09 13:49:51 issue_cmd RESULT:
07/27/09 13:49:51 issue_cmd
07/27/09 13:49:51 list_nodes
07/27/09 13:49:51 issue_cli_cmd command is '/opt/vmware/aam/bin/ftcli -domain vmware -port 8042 -timeout 15 -cmd listnodes'
07/27/09 13:49:51 issue_cmd CMD: /opt/vmware/aam/bin/ftcli -domain vmware -port 8042 -timeout 15 -cmd listnodes
07/27/09 13:49:51 issue_cmd STATUS: 0
07/27/09 13:49:51 issue_cmd RESULT:
07/27/09 13:49:51 issue_cmd Node Type State
07/27/09 13:49:51 issue_cmd

07/27/09 13:49:51 issue_cmd usbvt10pesxhost02 Primary Agent Running
07/27/09 13:49:51 issue_cmd usbvt10pesxhost03 Primary Agent Running
07/27/09 13:49:51 issue_cmd usbvt10pesxhost04 Primary Agent Running
07/27/09 13:49:51 issue_cmd

VMwarenodestatus=usbvt10pesxhost02,Primary,AgentRunning,usbvt10pesxhost03,Primary,AgentRunning,usbvt10pesxhost04,Primary,AgentRunning

07/27/09 13:49:51 myexit VMwareresult=success
07/27/09 13:49:51 elapsed_time Total time for script to complete: 0 minute(s) and 0 second(s)
~~2009-07-27 13:49:51.610 0xf6fa0b90 verbose 'App'~~ VpxaVMAP::Invoke Command returned successfully
~~2009-07-27 13:49:55.769 0xf7dbe6c0 warning 'App'~~ VpxaHalStats Unexpected return result. Expect 1 sample, receive 2
~~2009-07-27 13:49:55.770 0xf7dbe6c0 verbose 'App'~~ QuickStats Host CounterId 262165 has no value
~~2009-07-27 13:49:55.770 0xf7dbe6c0 verbose 'App'~~ QuickStats Host CounterId 262165 has no value
~~2009-07-27 13:49:55.770 0xf7dbe6c0 verbose 'App'~~ QuickStats Host CounterId 262165 has no value
~~2009-07-27 13:49:55.770 0xf7dbe6c0 verbose 'App'~~ QuickStats Host CounterId 262168 has no value
~~2009-07-27 13:49:55.770 0xf7dbe6c0 verbose 'App'~~ QuickStats Host CounterId 262168 has no value
~~2009-07-27 13:49:55.772 0xf7dbe6c0 verbose 'App'~~ Set internal stats for VM: 2 (vpxa VM id), 158 (vpxd VM id). Is FT primary? 0
~~2009-07-27 13:49:55.777 0xf7dbe6c0 verbose 'App'~~ Set internal stats for VM: 3 (vpxa VM id), 160 (vpxd VM id). Is FT primary? 0
~~2009-07-27 13:49:55.779 0xf7dbe6c0 verbose 'App'~~ Set internal stats for VM: 4 (vpxa VM id), 162 (vpxd VM id). Is FT primary? 0
~~2009-07-27 13:50:02.956 0xf6fe1b90 verbose 'App'~~ VpxaVMAP Checking Node Resources
~~2009-07-27 13:50:02.961 0xf6fe1b90 verbose 'App'~~ VpxaVMAP CheckThreshold values stored:(20795,29001,20795,29432) - retrieved:(20795,28999,20795,29430)
~~2009-07-27 13:50:02.961 0xf6fe1b90 verbose 'App'~~ VpxaVMAP CheckThreshold percent changes (0,0,0,0) - threshold 5
~~2009-07-27 13:50:13.306 0xf6f5fb90 verbose 'App'~~ VpxaAAM::PollForVMAPCommands Running ftExecute
~~2009-07-27 13:50:15.769 0xf7bcfb90 warning 'App'~~ VpxaHalStats Unexpected return result. Expect 1 sample, receive 2
~~2009-07-27 13:50:15.769 0xf7bcfb90 verbose 'App'~~ QuickStats Host CounterId 262165 has no value
~~2009-07-27 13:50:15.769 0xf7bcfb90 verbose 'App'~~ QuickStats Host CounterId 262165 has no value
~~2009-07-27 13:50:15.769 0xf7bcfb90 verbose 'App'~~ QuickStats Host CounterId 262165 has no value
~~2009-07-27 13:50:15.769 0xf7bcfb90 verbose 'App'~~ QuickStats Host CounterId 262168 has no value
~~2009-07-27 13:50:15.769 0xf7bcfb90 verbose 'App'~~ QuickStats Host CounterId 262168 has no value
~~2009-07-27 13:50:15.771 0xf7bcfb90 verbose 'App'~~ Set internal stats for VM: 2 (vpxa VM id), 158 (vpxd VM id). Is FT primary? 0
~~2009-07-27 13:50:15.774 0xf7bcfb90 verbose 'App'~~ Set internal stats for VM: 3 (vpxa VM id), 160 (vpxd VM id). Is FT primary? 0
~~2009-07-27 13:50:15.777 0xf7bcfb90 verbose 'App'~~ Set internal stats for VM: 4 (vpxa VM id), 162 (vpxd VM id). Is FT primary? 0

I did the numerous DNS checks to make sure DNS was configured properly. It would seem that it is. I am able to ping/vmkping all hosts with no problems. Also did the full remove HA, delete cluster, remove HA agents manually from each host and reconfigured to no avail. One thing that does come to mind is that the FQDN for these servers is >30 char. I read that this was a problem at one time but should have been resolved.

When I configured HA I selected all the defaults:

enabled host monitoring
prevent VMs from being powered on if they violate availability constraints
host failure cluster tolerates ... 1
3 guests on host1, 2 guests on host 2, 1 guest on host 3
admission control enabled
current failover capacity 2 hosts
configured failover capacity 1 host

I'm not sure where else to look. Anybody have any other suggestions?

Thanks,
Brian

FT VM Backup

Anton V Zhbankov — Thu, 14 May 2009 18:41:41 GMT

Is there a way to backup powered on FT VM since snapshots are not supported?

---
VMware vExpert '2009
http://blog.vadmin.ru

Virtual machine high availability error and VM resets

dkieschnick — Mon, 26 Oct 2009 21:08:24 GMT

We're seeing a couple of VM's restart spontaneously. I'm suspecting HA is triggering this based on some error condition, but I can't figure out why. Other VM's on those hosts don't restart.

We're running vSphere, fully patched, with HA, DRS, vMotion.

Logs include:

This virtual machine reset due to a guest OS
error. Screenshot is saved at
/vmfs/volumes/fa74840f-cecf5f9d/ITP-VM-TS02/-
ITP-VM-TS02-screenshot-0.png
info
10/25/2009 5:17:27 AM

Alarm 'Virtual machine high availability error'
on entity ITP-VM-TS02 send SNMP trap
info
10/26/2009 5:18:40 AM

Alarm 'Virtual machine high availability error' on
ITP-VM-TS02 triggered an action
info
10/26/2009 5:18:40 AM

Alarm 'Virtual machine high availability error'
on ITP-VM-TS02 changed from Gray to Gray
info
10/26/2009 5:18:40 AM

ITP-VM-TS02 reset due to a guest OS error.
Screenshot is saved at
/vmfs/volumes/fa74840f-cecf5f9d/ITP-VM-TS02/-
ITP-VM-TS02-screenshot-1.png
info
10/26/2009 5:18:40 AM screenshot shows the Windows screensaver, not really very useful

On the host I see:

~~2009-10-26 05:16:35.501 F65FC6D0 verbose 'vm:/vmfs/volumes/fa74840f-cecf5f9d/ITP-VM-TS02/ITP-VM-TS02.vmx'~~ Updating current heartbeatStatus: green

~~2009-10-26 05:16:55.502 F63CFB90 verbose 'vm:/vmfs/volumes/fa74840f-cecf5f9d/ITP-VM-TS02/ITP-VM-TS02.vmx'~~ Updating current heartbeatStatus: red

~~2009-10-26 05:16:59.157 F634DB90 verbose 'vm:/vmfs/volumes/fa74840f-cecf5f9d/ITP-VM-TS02/ITP-VM-TS02.vmx'~~ Running status of tools changed to: notRunning

~~2009-10-26 05:17:14.923 F630CB90 verbose 'Cimsvc'~~ Ticket issued for CIMOM version 1.0, user root

~~2009-10-26 05:17:15.502 F62CBB90 info 'vm:/vmfs/volumes/fa74840f-cecf5f9d/ITP-VM-TS02/ITP-VM-TS02.vmx'~~ Turning off heartbeat checker

~~2009-10-26 05:17:16.302 F638EB90 verbose 'DvsManager'~~ PersistAllDvsInfo called

~~2009-10-26 05:17:21.594 F65FC6D0 verbose 'vm:/vmfs/volumes/fa74840f-cecf5f9d/ITP-VM-TS02/ITP-VM-TS02.vmx'~~ Running status of tools changed to: running

~~2009-10-26 05:17:27.197 F65FC6D0 verbose 'vm:/vmfs/volumes/fa74840f-cecf5f9d/ITP-VM-TS02/ITP-VM-TS02.vmx'~~ Actual VM overhead: 231948288 bytes

~~2009-10-26 05:17:27.197 F65FC6D0 verbose 'Vmsvc'~~ RefreshVms updated overhead for 1 VM

~~2009-10-26 05:18:01.594 F630CB90 verbose 'vm:/vmfs/volumes/fa74840f-cecf5f9d/ITP-VM-TS02/ITP-VM-TS02.vmx'~~ Updating current heartbeatStatus: green

~~2009-10-26 05:18:01.595 F630CB90 verbose 'PropertyJournal'~~ ERProviderImpl<BaseT>::_GetChanges: Aggregate version Overflow 16 resourceConfig

~~2009-10-26 05:18:09.577 F634DB90 info 'TaskManager'~~ Task Created : haTask-16-vim.VirtualMachine.createScreenshot-68861

~~2009-10-26 05:18:09.577 F634DB90 info 'vm:/vmfs/volumes/fa74840f-cecf5f9d/ITP-VM-TS02/ITP-VM-TS02.vmx'~~ State Transition (VM_STATE_ON -> VM_STATE_CREATE_SCREENSHOT)

~~2009-10-26 05:18:09.577 F634DB90 verbose 'vm:/vmfs/volumes/fa74840f-cecf5f9d/ITP-VM-TS02/ITP-VM-TS02.vmx'~~ Createsnapshot request queued

~~2009-10-26 05:18:11.099 F62CBB90 info 'vm:/vmfs/volumes/fa74840f-cecf5f9d/ITP-VM-TS02/ITP-VM-TS02.vmx'~~ State Transition (VM_STATE_CREATE_SCREENSHOT -> VM_STATE_ON)

~~2009-10-26 05:18:11.099 F638EB90 verbose 'HostsvcPlugin'~~ Received state change for VM '16'

~~2009-10-26 05:18:11.100 F62CBB90 info 'TaskManager'~~ Task Completed : haTask-16-vim.VirtualMachine.createScreenshot-68861 Status success

~~2009-10-26 05:18:11.100 F638EB90 verbose 'Statssvc'~~ Adding vm 16 to poweredOnVms list

~~2009-10-26 05:18:11.120 F63CFB90 info 'TaskManager'~~ Task Created : haTask-16-vim.VirtualMachine.reset-68862

~~2009-10-26 05:18:11.120 F63CFB90 verbose 'vm:/vmfs/volumes/fa74840f-cecf5f9d/ITP-VM-TS02/ITP-VM-TS02.vmx'~~ Reset request recieved

~~2009-10-26 05:18:11.120 F63CFB90 info 'ha-eventmgr'~~ Event 1858 : ITP-VM-TS02 on vmhost03.cloudcomputing.com in ha-datacenter is reset

So it appears that heartbeat is lost. But I can't figure out why. The problem does seem to follow the VM, as I've moved it to various hosts and still see the same problem.

Any suggestions?

HA with SQL 2008

macka001 — Mon, 26 Oct 2009 16:22:39 GMT

Hi, we running under vsphere4 we run a single SQL server with about 5-8 client databases. One of our questions is how can we protect this VM SQL Server under VMWarre. the types of possible unplanned outages that im thinking of are OS corruption, and planned maintenace on the SQL server itself.

We have been advised that if the VM Host (physical Server) should fail, then another would bring up the running VM's in seconds, however what about if windows dies and i get Blue screen, any ideas/ best practice most welcome.

Our SQL runs with 2vCPU & 4GB Ram. Data storage is vai a SAN (iscsi) running windows Datacentre 2008

vSphere HA problem

jbarber — Thu, 22 Oct 2009 11:26:58 GMT

Hi there !

we are working with vSphere + HA (no DRS nor Vmotion). We have create one cluster with two hosts (x2 HP Proliant BL465c G5) and we are trying to probe the HA service.

The storage of the VM's are LUNs in a Netapp filer exported with iSCSI.

In each host are a few VM. All is working fine. When we stop one of the hosts, all the VM's reboots automatically in the other host. The problem is when all the VM are in one host (the other host is empty, obviously) and we stop it. Only in this case the "Reconfiguring HA" task fail in the destination host with the message:

"Configuring HA
Cannot complete the configuration of the HA agent on the host. See the task details for additional information."

I the task detail no much more details are shown. Only "cmd startagent failed".

In the vpxa.log:

~~2009-10-22 12:40:05.021 0xf7aa5b90 error 'App'~~ VpxaVMAP::EnableDAS cmd startagent failed:
~~2009-10-22 12:40:05.021 0xf7aa5b90 verbose 'App'~~ VpxaVMAP::SetDasState dasstate changing from configuring to error
~~2009-10-22 12:40:05.021 0xf7aa5b90 verbose 'App'~~ VpxaInvtHost Increment master gen. no to (87): Das:VpxaVMAP::SetDasState
~~2009-10-22 12:40:05.021 0xf7aa5b90 verbose 'App'~~ VpxaVMAP Error reason changed to configFailed
~~2009-10-22 12:40:05.030 0xf7aa5b90 verbose 'PropertyProvider'~~ RecordOp ASSIGN: info.state, task-48
~~2009-10-22 12:40:05.031 0xf7aa5b90 trivia 'PropertyCollector'~~ TriggerProcessGUReqs: Session 525a7e9c-16cc-3d72-7f80-da206c3efb46
~~2009-10-22 12:40:05.031 0xf7001b90 trivia 'PropertyCollector'~~ ProcessGUReqs Start: Session 525a7e9c-16cc-3d72-7f80-da206c3efb46
~~2009-10-22 12:40:05.031 0xf7aa5b90 verbose 'PropertyProvider'~~ RecordOp ASSIGN: info.cancelable, task-48
~~2009-10-22 12:40:05.031 0xf7aa5b90 verbose 'PropertyProvider'~~ RecordOp ASSIGN: info.error, task-48
~~2009-10-22 12:40:05.031 0xf7aa5b90 info 'App'~~ VpxLRO -- FINISH task-48 -- -- vpxapi.VpxaService.enableDAS -- 525a7e9c-16cc-3d72-7f80-da206c3efb46
~~2009-10-22 12:40:05.031 0xf7aa5b90 info 'App'~~ VpxLRO -- ERROR task-48 -- -- vpxapi.VpxaService.enableDAS: vpxapi.DasConfigIntFault:
(vpxapi.DasConfigIntFault) {
dynamicType = <unset>,
faultCause = (vmodl.MethodFault) null,
reason = <unset>,
output = <unset>,
errorcat = "",
errortext = "cmd startagent failed: ",
msg = "",
}

We have check the DNS's and all this things (/etc/opt/vmware/aam/FT_HOSTS, /etc/opt/vmware/aam/vmware-sites,...). Everything seems ok.

In the cluster are enabled the "Host Monitoring", "VM monitoring" and "Allow VMs to be powered on even if they violate availability constraints" options.

We have apply all the last patches and updates.

Rarely fail only when the "destination" host is empty !!

¿Anyone have seems this case?

Thanks in advance.

Javier Barber.-

FT for a web application server

DLevine — Sat, 24 Oct 2009 17:00:43 GMT

I am looking at upgrading our current VI to vSphere and take advantage of FT for our web application server... I am curious though... as the machines run in lockstep, would the secondary VM be performing the same I/O on the backend database? What would happen there... ?

HA and Templates

wonkert — Fri, 23 Oct 2009 10:57:23 GMT

Hopefully somebody can help clarify something.

We are currently using templates on a ESX 3.5 / VC 2.5 / HA environment. Should a server fail, then all VMs are restarted elsehwere in the cluster, but not templates it would appear. The result is templates are no longer usable until the host is reconnected or the template is re-registered with another name. Interstingly, if a host is put in maintenance mode, the VMs are evacuated, but again, not the templates.

The issue with this is around the use of provisioning tools and their dependency on the templates. When the host is in main mode, or has failed, the template becomes unusable and there's no obvious way to move it to another host and maintain the original name (as it fails to add a second time with a duplicate name conflict)

How are others managing templates in an HA environment?

HA is not working vSphere

winsrock — Thu, 22 Oct 2009 07:47:00 GMT

Hi All,

Continously we are getting HA agent error message after migrating to ESX4. Please help me to resolve this issue.

Thanks

Jaweed

das.allowNetwork doesn't work with Distributed Virtual Switch

nicholas1982 — Wed, 21 Oct 2009 05:19:59 GMT

Does anyone know if the advanced HA option "das.allowNetwork" is supported on a DVS because I can only get it to work with a Standard Switch?

Enabling FT on a VM - Getting "Insufficient Memory Resources"

Tigerstolly — Thu, 15 Oct 2009 13:56:03 GMT

Hi,

I am trying to enable FT for a test VM. When i select "Turn on Fault Tolerance" it tries to enable it, but pretty quickly comes up with a pop-up saying "Insufficient Memory Resources"

The host in question has 128gb of RAM, and this is the only VM on it so physical host memory shouldn't be the problem....what else can we check ?

Thanks !

ESX 4 HA spontaneous host disconnects

thomn8r — Thu, 15 Oct 2009 16:03:47 GMT

I have a 16-way cluster of ESX 4 hosts. Yesterday, I thought I'd try playing with the fault tolerance features, so I took steps to set it up. It got ugly when I turned on "VCenter requires SSL certificates" - even though I confirmed the certs for all the hosts, VCenter still threw errors so I turned the feature back off.

Today, individual hosts are being turned off at random. When I say "turned off" I mean the networking on the host is disabled - running ifconfig and netstat shows that the network services are running, but you can't ping the host, and you can't even ping the network gateway from the host. I have to reboot the host from its console, and on some of them, I'll see perl errors for esxcfg-reboot. On top of that, when a host is disconnected, vmotion is unable to migrate the VMs off.

I've tried putting a host into maint mode and removing/re-adding to the cluster - doesn't make a difference.

I'm re-installing one of the hosts right now, don't know yet what if any help that will be.

I desperately need to get this stable again, lest I be drawn and quartered for lunchtime entertainment...

HA/FT Solution for patching?

jcorbin121 — Mon, 19 Oct 2009 02:40:58 GMT

Is there a managable solution for applying monthly Microsoft securtiy bulletins within HA/FT features? Can I somehow (quickly) create a temporary VM clone while the original VM gets patched and restarts? If I were to set the VM as FT does a reboot activate the secondary FT VM? MSCS clusters are painful solution, we have implemented on DB and Web servers already, would like to move these servers to VM and looking for a no-downtime patching solution.

John

enter standby mode disabled

dingding — Tue, 13 Oct 2009 05:28:23 GMT

i have two ESX 4 each with 2 NICs in a cluster, in the configuration|network adapters tab, i can see both are wake on LAN suppported. i want to test the WOL, but in one ESX, the menu item "enter standby mode" is gray out. why? thanks in advance

Snapshots for Fault Tolerant VMs

Bramha — Tue, 06 Oct 2009 13:21:56 GMT

The VSphere availability guide says that you cannot Snapshot fault tolerant VMs . How else would one backup such VMs ?

How to determine which VMs failed over in an HA event

rreynol — Sat, 17 Oct 2009 16:10:47 GMT

We had a COS panic under ESX 4 (vSphere) and HA performed as expected. We would like to let the systems administrators of the VMs know if their VM was involved but so far no luck in determining that. VMware only answers production down issues over the weekend in the live queue so as we wait for their email can anyone point us to a log that would list which VMs were restarted? I have looked through vCenter logs and all the logs under /var/log/vmware/aam but so far no luck. vmkernel logs for all the ESX servers in the cluster would likely be a source of last resort but I was hoping there would be some type of HA log that would list them.

Cluster w/ Non FT Capable Machines?

obpa — Thu, 15 Oct 2009 13:41:38 GMT

I have 6 ESXi 4 servers - (3) w/ AMD 2222 procs and (3) w/ AMD 2380 procs. The 2222 procs are not FT capable, at least I don't think so... As of now, all (6) are running in a single HA/DRS enabled cluster (EVC Mode - AMD Opteron Generation 2 even though the 2380s support Gen 3). The (3) hosts w/ the 2380 procs are FT enabled as of now.

I don't really have any reason to split this setup into 2 clusters unless it is a bad idea to run a mix of FT-capable and NON FT-capable servers in the same cluster. Thoughts on this?

HA with physical devices attached to a VM

ScottBentley — Tue, 13 Oct 2009 10:11:14 GMT

Quick question about HA

We are considering Essentials Plus to enable HA over a couple of host servers for a small environment.
One of the guests will host a backup solution and be attached to a physical tape device using SCSI passthrough.

I know if the server were to be restarted on the other host the physical device would obviously not be available, my question is would HA actually be able to relocate the machine to the second host in the event of a failure, or would it fail due to it having a physically attached device that is unavailable on the second host?
I know this causes VMotion to fail but right at this present time I don't have access to a test environment to check, done a bit of digging here and couldnt find anything definitive.

Many Thanks

Scott

Datto's Inexpensive VMware FT Capable Lab Hardware

Datto — Fri, 26 Jun 2009 19:01:33 GMT

For those of you interested in creating a low-cost
(relatively speaking) lab for VSphere 4.0 that will also do VMware FT, here's
what I've used for my three white box VMware FT setup.

White Box 1 -- AMD AM2 Opteron 1354 (quad-core single socket
CPU to save money - CPU cost is $85.00 each delivered from Newegg), Asus M2N-LR
motherboard (about $100 delivered Open Box from Newegg, much less on eBay when
available) and you'll need to get the latest M2N-LR BIOS installed but you may
not be able to flash the board using the Opteron 1354 since the board may
require a lesser processor to flash the BIOS -- I had a low-end AM2 3600 CPU
sitting around and used that for flashing, then put in the AMD Opteron 1354),
two PCI-X dual port Intel gigabit cards (I got mine for $16.50 delivered for
each dual port card on eBay -- this combination of add-in NICs and the two
gigabit NICs on the motherboard that work properly with ESX 4.0 give me six
physical gigabit NICs in the white box), 8GB of non-ECC memory (4x 2GB GSkill
PC2-6400 DDR 800 -- the board will take ECC memory if that's your preference
but I had this GSkill memory already), 20G Maxtor IDE drive for $15.00 each as
the ESX 4.0 boot drive. For CPU fan/.heatsync I'm using a Zerotherm NV120
($50.00 from Newegg or about 60% of that used on ebay when available). I don't
put a permanent CD/DVD drive into any ESX host in my lab and just put the
CD/DVD drive temporarily in when I need it. I also never use a permanent floppy
drive and just plug in a USB floppy drive if I need it for say BIOS flashing.

White Box 2 -- identical to White Box 1 above

White Box 3 -- AMD AM2 Opteron 1356 (also a quad-core single
socket CPU to save money and it's within the 400MHz speed difference limit required
by VMware FT between CPU speed in White Box 1 & 2 above), Asus M2N-L
motherboard, combination of PCI-e dual port Intel gigabit NICs in the two PCI-e
16X slots and straight PCI 1000MT gigabit NICs to get to a total of 6 physical
gigabit NICs in the white box, 8GB of ECC memory (this CPU, motherboard and
memory came as a bundle I'd bought on eBay). Note I couldn't get the two
existing Marvell physical gigabit NICs on the M2N-L motherboard to work with
ESX 4.0 and haven't had time to chase down a NIC driver shoehorn process for
the Marvell NICs so I just disabled the Marvell NICs in the system BIOS for now
and use the other PCI-e and PCI gigabit NICs for my FT purposes. The CPU
Heatsync is also a Zerotherm NV120. Boot drive is also a Maxtor 20GB IDE drive.

Regular VMs (non-COS VMs) sit on an Openfiler 2.3 box
providing shared iSCSI storage for the cluster of white boxes. One gigabit NIC
on each box is dedicated to FT Logging-- the likely max number of FT Primary
and Secondary VMs on any single box is likely three to six of these FT Primary
or Secondary VMs before FT logging might get swamped. So far I'm running a
total of five FT protected light-duty VMs in the cluster and they seem to have
no problems and FT does correctly transfer over to the Secondary if the Primary
FT protected VM fails. I'm also running non-FT VMs in the cluster and there are
also two other ESX 4.0 boxes in the same cluster that are not FT capable (they
have AM2 Kuma processors which are VMotionable between the Opteron 1354/1356 CPUs
but ESX 4.0 needs a keystroke hit on the bootup on those Kuma boxes to keep the
bootup from stopping -- something having to do with the Kuma / M2N-E
motherboard combination used in those non-FT capable white boxes and ESX 4.0 --
if I put a standard AM2 Brisbane CPU in those M2N-E boxes they boot normally
for ESX 4.0 (but won't VMotion straight-away with the Opteron 1354 / 1356 CPUs)
so the problem is with using Kuma processors and ESX 4.0). I don't down ESX
boxes much so this isn't a problem for me at the moment.

Note that the M2N-LR boxes will not go into Standby Mode for
unknown reasons (regardless of which NIC is the Standby-capable VMkernel NIC)
but the M2N-L box will sometimes go into Standby Mode so it's likely a BIOS
issue with these Asus motherboards I suspect. The Asus M2N-E / Kuma CPU
motherbaord/CPU combination have no problem going into Standby mode but with
the Kuma processors in the M2N-E motherboards the boot back up process needs a
keystroke hit so that's a problem.

Also note the huge Zerotherm NV120 heasyncs cover the
closest (the first) motherboard slot in every motherboard I use. Some people
have just trimmed the fins of f the NV120 to make room as necessary.

VCenter runs as a VM on another cluster.

These are not a supported configuration so don't go using
this in your company production setup. Hope this helps folks wanting to build a
relatively economical VMware FT lab setup.

Datto

Service that Continue to be Available if vCenter is lost

herevirtually — Tue, 13 Oct 2009 11:26:41 GMT

Hi,

This isn't strictly a HA or FT question but it's an availability question and this is the best forum I could find

On vSphere 4, suppose my VC server is a physical box without heartbeat for vCenter set up. It's not a VM so it's not protected by HA / FT and no heartbeat so when it's gone, it's gone. Assume I have the top vSphere license (Enterprise Plus). Is there a definitive document available from VMware that states what continues to be available if the vCenter server goes down. I'm thinking things like HA, DRS, licensing etc. Obviously the configuration that I describe is not ideal but I just wanted to describe that artificial scenario to illustrate my requirement for this definitive list. I've googled this and found some information but would be interested to get the most definitive information I could, hopefully a VMware doc.

Thanks,

herevirtually.

FT for AD domain controller

steve31783 — Thu, 08 Oct 2009 18:53:46 GMT

Does anyone have any experience using FT on a windows domain controller?

I received an inquiry today asking me if this is possible, which I know it is... im just not entirely sure how a DC would react to this technology, or if there is any benefit of using FT rather than just building out 2 DCs.

Any thoughts?

Split Brain in HA

omera — Thu, 08 Oct 2009 05:54:33 GMT

Hi ,

In Vmware HA, if one of the ESX hosts lost its' Service Console connection the hearthbeat was lost. Then ESX host goes into Split Brain condition. If you use only two ESX hosts and one of them has Service Console connection problem which one goes to Split Brain Condition ? Both of them ?

Please help me about the question...

How To Disable Fault Tolerance (FT) without vCenter?

kegwell — Wed, 07 Oct 2009 19:21:40 GMT

I have a bit of a strange situation here. I setup a lab environment to test FT on 2 blade servers. One of the virtual machines I created and enabled FT on was vCenter. Since the creation of my lab environment, one of the ESX blades was removed from the cluster and repurposed for another non-VMware use before I was able to disable FT on the VMs I was testing. Since then, a new test blade was put in place. So, I'm back to having two ESX blades. However, when I try and add the blade now hosting vCenter w/ FT it says:

Add standalone host
***
Cannot add a host with virtual
machines that have Fault
Tolerance turned On as a
stand-alone host.
Administrator
VCENTER
10/7/2009 2:20:05 PM
10/7/2009 2:20:05 PM
10/7/2009 2:20:06 PM

My question to you is, how can I disable FT without vCenter? I cant get the enable/disable/off FT without logging into vCenter, but I cant add the host with vCenter to get to that screen. Is there something in a configuration file that I can change that will at least mask the presence of FT so I can get past this?

Scrubbing

tsjellie — Wed, 07 Oct 2009 20:47:11 GMT

How long does the scrubbing take?

Auto start vm's

jsteppe — Wed, 07 Oct 2009 15:43:02 GMT

I have a vm that is dedicated to 1 host because this host has a serial connection to a modem for it. Now if I need to restart the host for any reason and I have shutdown the vm since I can not migrate it to another host due to the serial connection. Once this host is back online it will not power on the vm automatically.. With DRS and FT enabled do I still need to use the Auto Start options specific to this host? Wasn't sure if that was necessary with it being in a cluster... If that is the case then I will enable the autostart functions on this host for this 1 vm..

Just wanted to make sure I wasn't missing anything.

HA Error. Message: Misconfiguration in the host network setup.

csabahitter — Sat, 03 Oct 2009 13:28:27 GMT

Hi,

I have a configured cluster with 2 hosts. (ESXi4 and VC 4.0) I installed virtual machines, it's working, and i can ping management network for all host, but when i configure HA, i get the following message: Cannot complete the configuration of the HA agent on the host. See the task details for additional information. Misconfiguration in the host network setup.

Have ideea what is wrong?

HA & FT on Dell PowerEdge R300

GHLSYSTEMS — Thu, 01 Oct 2009 11:38:22 GMT

Hi,

I'm currently evaluating VMWare Fault Tolerance on 2 identical Dell PowerEdge R300 Servers.

A simple setup with both Servers running on Dell custom ESXi 4.0, both runs on 8GB Memory & Quad Core CPU's

1. 1 Cluster created via VMWare Vcenter by the name of Cluster01

2. Both Servers added into the Cluster01

3. Fault Tolerance Logging & VMotion on Network Interface enabled on both servers

4. Both Servers are on the same network segment.

However, I have not been successcul in activating the HA option on the Cluster.

These are the error messages that I get...

On VCenter

1. Insufficient Resource to satisfy HA Failover on level Cluster01

2. Unable to contact primary HA Agent in cluster Cluster01

3. cmd addnote failed for secondary node: Internal AMM error - Agent could not start: Unkown HA error

Question:

Did I miss out any steps?

FT pNIC work around to remove logging traffic from NIC?

Sketchie — Thu, 01 Oct 2009 12:35:00 GMT

We are currently setting up a lab environment (that looks nothing like our production). Blade servers all around, but the problem is that they only have 4 pNICs, with no available slots (so no chance to increase this number).

So, the problem is how to use FT without using a dedicated NIC. I just finished reading KB 1011965, which gives somewhat of a work around, however this does not mention anything regarding the heartbeat connection. I guess theoretically I could move the heartbeat over the the SC.

Can anyone verify that the logging can be completely removed from the FT NIC according to KB 1011965 and that all I would need at that point would be the heartbeat connection?

Thanks!

HA Alogrithm used to restart VM's

campbellabbey — Thu, 01 Oct 2009 05:43:12 GMT

I have a query about HA regarding the alogirith (be it ‘parallelism'/'serialism') used to DRS a number of VMs per host on a cluster where the VM's were prviously housed on a Host that has gone down

For example If a host suffers a failure and that Host has 20 VMs on it, how does HA determine;

i) Which host(s) to restart those recognised failed VMs
ii) Does HA restart VMs sequentially or in parallel and are there any ‘number rules' - for example it can only start 5 VMs on the first host it selects for recovery, then another 5 on another host etc etc. How does the algorithm essentially work?

Thanks

Cam

Option "Support clustering features such as Fault Tolerance" dimmed (unavailable)

AHatting — Wed, 30 Sep 2009 15:02:54 GMT

Hi,

For a customer we are trying to simulate a SQL failover using MSCS.
I found a lot of information about how to configure this, using only virtual machines.

However, for some reason, we can not select the option "support clustering features such as Fault Tolerance" , when trying to create the quorum disk in Vsphere. (See image)
We deployed 1 DC (W2K3 -x86) and 2 SQL 2005 servers on a W2k3 x64 environment.

We use ESX 3.5 and Vsphere 4.0

I'm fairly new to this configuration and i could not find any appropiate document, pointing in the right direction.

Thanks in advance.

Armand

HA Help

prutter — Tue, 29 Sep 2009 22:22:12 GMT

i have a question concernign HA. i just put in a third host and it's a vShpere host, and i can seem to get HA configures on the new host. Are there compatibility issue between 3.5U4 and 4.0??

Configuring HA
192.168.1.27
Cannot complete the configuration of the HA agent
on the host. See the task details for additional
information. Other HA configuration error.
System

pNIC configuration with FT / vMotion / Management

pigbloke — Tue, 15 Sep 2009 08:48:02 GMT

In our planned vSphere implementation we have 3 hosts each with 8 pNICs. 2 pNICs will be used for VMs and 2 will be used for iSCSI.

My question is, what would be the preferred configuration for the remaining 4 pNICs?

I was thinking 2 for Management and then the other 2 configured for FT and vMotion with explicit failover.

Does this sound reasonable or is there any reason FT should be kept on its own dedicated pNICs?

We may not actually be using FT initially, but plan to in the future and want to ensure that the configuration is in place right from the start.

Thanks in advance for any advice.

Nick

Mise en oeuvre de FT

collivier79 — Fri, 25 Sep 2009 08:31:35 GMT

Bonjour,

J'ai testé l'ensemble des nouvelles fonctionnalités offertes par ESX 4.0. Toutefois je reste dans l'impossibilité de faire fontionner FT. J'ai bien pris en compte la longue liste de limitations qui accompagne cette fonctionnalité. Mes processeurs sont compatibles, l'outil de compatibilité fournit par VMware me garantit que tout est opérationnel. Pourtant lors de la mise oeuvre de la Fault Tolerance d'une machine en Windows 2003, un message d'erreur me signale que la FT a échoué car la seconde machine n'a pu démarrer (Host incompatible).

Je travaille sur une maquette composée de deux serveurs DL380G5 (intel 54xx) attachés en SAS à une baie MSA2000sa.

Je dispose de deux interfaces réseau dédiées à FT (inteconnectées par un câble croisé). Par commodité, VMOTION et FT s'appuie sur le même réseau en 10.44.10.x/24

Merci de votre aide

HA VM recovery is acting oddly

robkelley — Tue, 22 Sep 2009 13:03:14 GMT

Hi
I am doing some testing/learning on HA using ESX4 vShpere 4.

2 x ESX4 servers, named esx1 & esx2

esx1 has a openfiler VM which is set as the iSCSI shared storage SAN, both esx servers have mappings to this shared storage, also a windows vm running vSphere centre server

esx2 has a Linux VM which is stored in the iSCSI SAN.

When I unplug esx1 from the network after a few seconds the VM appears on esx2 and all is great.

The problem starts when I bring esx1 back onto the network, it seems as if vSphere is trying to move the VM back to esx1, but fails (no error msg's) as it appears on the VM list on esx1 for a second, then disappears again, and then comes back, and disappears again, and so on and on, it has been doing this for 5 days now.

Any ideas why this is happening?

Thanks

FT simply won't work

AlbertFish — Fri, 04 Sep 2009 20:18:56 GMT

I RTFM. I have checked all prereqs. It configures, creates the secondary, they run. It is all happy. I yank the cables. Nothing happens. all 4 die. If I physically turn the primary off - it works. Pulling all of the network cables (so it still gets disk heartbeat but no logging) never fails over.

This is on a pair of dell 905s 128 / 8socket

4.0.0 build 175625

2 service consoles, and a dedicated FT Logging switch. 2003r2 server.

What could I be missing?

Host says FT is not enabled but it is

snootalope — Tue, 04 Aug 2009 20:18:33 GMT

Couple weeks ago I got FT enabled on all the hosts. Each hosts summary showed Fault Tolerance Enabled: yes - So I set a few of my VM's to FT enabled and they did thier thing and became FT enabled just fine. Tested the failover, worked like a champ. Now however, the FT is still enabled and fails over just fine to it's secondary, but viewing the hosts summary, it shows its disabled "Fault Tolereance Enabled: no" - what gives? FT works just fine. I can add new FT enabled VM's and they enable just fine along with failing over, yet the hosts say they're not enabled. bug???

ESX4/vSphere4

Anyone else seeing this?

Switching on FT when the vCenter server is a Virtual Machine.

fatbloke61 — Tue, 15 Sep 2009 18:13:57 GMT

I am completely happy with the process of starting FT on a VM which is turned off and scrubbing the disks etc. However, i have a requirement to run FT on a VM that is itself the vCenter Server, so obviously i cant access it through vCenter to do this because it will not be available. Is there a way to do this other than run another vCenter server to do it from, script perhaps?

Thanks.

Clustering support for Windows Server 2008

rmagoon — Tue, 18 Aug 2009 14:46:07 GMT

In the vSphere support statement for Server 2008 clustering, it mentions that both 2008 nodes must not belong to an HA/DRS cluster. We currently have our cluster designed to accommodate all hosts (HA/DRS enabled) within the datacenter. If we exclude HA/DRS functionality for the individual cluster node VMs, will we be supported?

TIA

-Rick

vswitch name for HA

mure79it — Fri, 11 Sep 2009 19:34:43 GMT

Hi,

i have a doubt. In HA environment must be the same the name of the Vswitch or the name of Virtual Machine Port Group?

Thanks,

Moreno

Hardware Compatibility within HA Clusters

srctech — Thu, 10 Sep 2009 20:27:19 GMT

Does anyone know of a document that shows a cross reference of what hardware is compatible within an HA Cluster. From what I understand, CPU's have to be close in architecture. I would like to be able to move vm's between a new server with a XEON E5430 in a Dell PowerEdge 2900 and an existing server with a XEON Nocona (80546K) in a Dell PowerEdge 2850. Is it possible for HA to work if I were to configure these two machines in a cluster? If there was a table on this information it would greatly help with me with future configurations as well.

Thanks,

SRCTECH

HA Error while running health check

what2do — Tue, 08 Sep 2009 14:29:50 GMT

I have an environment with esx 3.5 servers and vc4. Sporadically I receive errors"

Error <date> <time> HA agent on <host> is cluster <clustername> in <datacenter. has an error: Error while running health check script

Shortly after the ESX host looses connectivity with VC but the vm's still run on the host

The only reference I can find is to windows power shell, but windows power shell is not loaded on the VC box, and task manager shows nothing scheduled.

Is this something new in VC4? if not any tips on troubleshooting. I have reloaded HA and going to monitor to see if it happens again, but I would like to know the script the error is refering to.

Configuration needs for Fault Tolerance(FT)

TMCishikawa — Tue, 08 Sep 2009 02:25:51 GMT

Hello

I'd love to have your advice for FT configuration.

In " vSphere Availability Guide", it says "VMware recomends that you have a minimum of three host in the cluster". It also says its only "Recommendations".

in another day , i heard that VMware says " FT is supported even if its configurations are 2 ESXs in the cluster. ".

But i really can't imagine how works FT with only 2 ESXs.

in FT, primery esx send "logging traffic" to the secondary esx, then secondary esx do the job and send acks to the primery esx, and primery do the job finally

Once primery esx is failed, secondary esx become primery and third esx (spare one) become secondary. that I think how works FT is.

But if FT cluster has only 2 ESXs, after primery ESXs failed, the FT cluster has only one ESX(which is primery?). Primery ESX can do the job only after it receive acks.

but in this situation(FT cluster has only 2ESX and primery is failed), there is no secondary esx so that means primery can't receive acks and can't do the job.

Here is my questions, how can FT works if FT cluster has only 2 ESXs?????

Errors in Configuration HA agent on the host

ITBCM — Sat, 05 Sep 2009 16:48:28 GMT

Hello!

I have a problem with vmware HA.

I have 2 hosts in the cluster (ESX 4.0.0) and one Vcenter server 4.0

I set enable HA in the cluster but my hosts can't configuring HA agent.

It's give such error: Cannot complete the configuration of the HA agent on the host. See the task details for additional information. Other HA configuration error.

In the task details for this task: cmd addnode failed for secondary node: Internal AAM Error - agent cold not start.: Unknown HA error

How can i fix this problem?

Thank you!

Fault tolerance status bouncing from starting to need secondary VM

xavier@lynk.com — Tue, 23 Jun 2009 16:30:18 GMT

When I enable FT on a VM after the task is finished in the recent tasks pane, the status on the the summary tab for this VM keeps going back and forth from starting to need secondary VM, and will always show as not protected. furthermore I cannot turn FT off or do any operations on the VM (power off etc)

I tried again with another vm I created and this time the primary VM was marked as orphaned and could not do anything with it, but interestingly the secondary was showing as ON. I tried to remove the secondary and could not do it so I remove the primary orphaned VM and the secondary disappeared but I still see the files on the data store. I am trying to delete them but they are locked, most likely they are still running on the background but vcenter does not show them anywhere in the inventory .

I have a dedicated 1Gb port for FT logging, 1 Gb port for VMotion and 1Gb for VM traffic. The port group for FT is in a different physical switch. The VMotion and VMs port groups are on other physical switch and I do not see problems with them. The ESX servers are not highly loaded whatsoever.

Any ideas on how to get FT working properly?

Disable HA or use Maintenance Mode when upgrading network switches

vikers — Wed, 02 Sep 2009 10:19:33 GMT

Hi,

We are planning on upgrading our network switches. In case hosts loose network connection i want to avoid having HA kicking in and hosts trying to run already running vms. What is best practise for HA-enabled cluster in case the hosts loose network connection?

Disable HA or putting all hosts in maintenance mode? Or are there any other suggestion?

Vikers

Very Strange HA issue

fusionit — Thu, 03 Sep 2009 19:20:36 GMT

Hello Everyone I have a 3 host vSphere cluster running build 175625. I have enabled HA and DRS on the cluster and it has been running fine for months. I placed one of my hosts in maintenance mode in order to reboot it becasue there was a problem with two VM's on that host in that I could not perform any action on them as I was getting the famous "cannot complete task as antoher task is already in progress" message and a simple restart of the management services on that host did not resolve it. so I rebooted this particular host and after it came back up I took it out of maintenance mode and it failed when trying to configure HA with yet another famous error message of "HA agent has an error : cmd addnode failed for secondary node: Internal AAM Error - agent could not start. : Unknown HA error" so I did the normal troubleshootig (DNS /etc/hosts files etc.) connectivity and name resolution are solid. OK so here's the strange issue. This HA error message was happening on ESX02 host

I disabled HA on my cluster. I then re-enabled it and ESX03 and ESX02 enabled HA just fine and ESX01 now failed with the exact same error message. so I disabled HA on ESX03 (put it in maintenance mode) and reconfigured HA on ESX01 and low and behold HA configured just fine. Took ESX03 out of maintenance mode and wouldn't you know it it failed on HA with the above mentioned error. For some reason it seems like HA does not want to enable on more than 2 hosts in my cluster. Does anyone have any ideas why this may be happening. I knew HA had its issues and I'm comfortable troubleshooting the software (VCP) I just can't figure this one out. Thanks in advance.

HA calculation slot

nics30 — Thu, 03 Sep 2009 05:13:48 GMT

I have two conflicting sources of information for calculating the HA slot - one states on page 76 of the resource managment guide that it is based on the maximum RESERVED memory/CPU for aany virtual machine in a cluster see http://www.vmware.com/pdf/vi3_35/esx_3/r35u2/vi3_35_25_u2_resource_mgmt.pdf

THe other source is a knowedge base article which states the HA slot is calculated based on the maximum CONSUMED memory/CPU of a virtual machine. Anyone know which one it is? *
(This is in 1010594 (knowledge base article))

*

vCenter & HA

curiousagain — Thu, 03 Sep 2009 07:36:43 GMT

Hi,

I have a fundamental question regarding vCenter & HA.

I want to build a HA environment with vCenter & 2 ESXi hosts.

If I put the vCenter as a VM on ESXi host and that host is going down, how the vCenter can start all the vm's on the other host (HA) - as it also will be down?

Thanks.

Insufficient resources to satisfy HA failover

nics30 — Tue, 01 Sep 2009 02:52:57 GMT

I have two ESX Servers in a cluster- each with 32 GB of memory and only 8 virtual machines running on these ESX Servers. Seven of the virtual machines have 2GB memory assigned to them and no memory reservation- the HA slot should be 2GB as the eighth virtual machine is our Virtual Center Server and 2GB is the memory reservation for this virtual machine.

I have a policy of strict Admission control on this environment, which should work as 2GB(maximum memory reservation for VMs) X 8 Virtual machines=16GB memory and the "number of host failures the cluster can tolerate" is set to one. I have no idea why but when our ESX Server disconnected tempoarily I got the message "insufficient resources to satisfy HA failover"

There appears to be plenty of resources - both in terms of CPU for a failover to the other ESX host to occur.

Anyone any ideas?

HA / FT - What happens if (one) host looses connection to SAN/LUN

asp24 — Sat, 29 Aug 2009 10:13:15 GMT

I have a few hosts in a cluster running HA/DRS (and I have tested FT a few times).

A few days ago one of the hosts loose connection to the SAN. I tried to do a rescan, but it said another scan was in progress etc.. So I apparently have some vmware/san issues. The problem was that I had to reboot the host to make HA restart the VM's on a different host. Is it possible to protect against these kind of failures? Or do I have to create some sort of cron-script that check connectivity and run a reboot -f or something if storage fails? (remember, the SAN IS UP, but one host looses connectivity.

Errors while configuring FT

ASchinkel — Thu, 27 Aug 2009 16:29:28 GMT

I am working with VMware vSphere and VMware ESX4 on two BL480c blade units.
I have ensured that both have the requisit processors (xeon 45nm 54xx series) and have the virtualization options enabled.

The VMDK file which serves as the virtual HDD for the VM in question is thick formatted and come to about ~57,671,680K to represent a 55GB drive.
The datastore in which the VMDK resides is 78GB, with 22.45GB free.

The virtual settings have been tested with either the vmxnet2 (enhance) or the e1000 network adaptors.
The virtual cd-rom drive has been removed completely.

To enable fault tollerance I have powered down the server, right clicked, and selected 'fault tollerance'.
The highest I've observed the status reach is 33%, usually with a 10minute stall at 28% at the very beginning.
However close to completion is actually gets (which is unknown to me), when I come back to it I am greeted with the error:

Turn On Fault Tolerance
systemThree
A general
system error
occurred:
Connection
timed out
admin
vmlabcenter.tidir.bctest
2009-08-26 4:38:31 PM
2009-08-26 4:38:31 PM
2009-08-26 4:41:03 PM

After which apparently the (i presum) incomplete eager zero formatting has been cancled before complesion, so if I power on the VM and try to enable FT, then it simple returns the error:

Turn On Fault Tolerance
systemThree
Unsupported
virtual
machine
configuration
for Fault
Tolerance.
admin
vmlabcenter.tidir.bctest
2009-08-26 2:08:38 PM
2009-08-26 2:08:38 PM
2009-08-26 2:09:09 PM

This error message also includes the elaboration of:

The unused disk blocks of the virtual machine's disks have not been scrubbed on the file system. This is needed to support features like Fault Tolerance.

So my question is,
Given the above scenario,
What is the likely problem / solution / further troubleshooting steps / alternative method to get the disk properly scrubbed?

Thank you for your time,

-Aaron

HA without shared storage

eccl1213 — Sat, 11 Jul 2009 00:07:46 GMT

Hi everyone,

I have a client that requires an HA solution without shared storage. They are a security monitoring firm and regulations require that he have automatic failover but the machies must not share storage.

I'm having trouble finding a product that both replicates and monitors the ESX server to provide automatic failover. vReplicator doesn't have an auto-failover feature. VCenter HA and FT both require shared storage.

We even looked at two SAN's with replication, but even Site Recovery Manager is a manual fail-over process. Not to mention two sans for 3 virtual machines is a bit over kill. The three VM's would be very low utilization, two are a linux OS and one is Windows.

Any suggestions?

ESX 4 host has gone lazy & HA function was not working!

calvinso — Mon, 22 Jun 2009 02:11:45 GMT

This morning when I back to office, 6 servers were down and I found that the ESX host server that hosts those 6 servers was disconnected from vCenter. It were no response when I do ping test to all ESX hosts and guests. The migration and power up option were grey out (because that not responding ESX server was disconnect from vCenter). HA function is not working, those server fail to restart on other hosts (I got 4 ESX servers and sure there are enough resource to startup). I went to the console of that host server and try to login, it has no response after typing "root". I have no choice and I need to cold boot this host server. At the mean time as I can't restart those guest as the power on option was gery out, I need to remove that crashed ESX host from the cluster and manually add those guest vmx back to clusters inventory in order to power it on.

I have filed a SR 1425498551. I rebooted the ESX and try to create an vm-support file, it was failed.

1. No matter system crash, network disconnect or hardware failure and host server, I expect HA function will restart all those guest system on other machine. (I have enable the HA function and the setting was attached). It is not working.
2. If such case will happens, VM concept are no longer safe for holding important servers. As once problem happens, it was not affected 1 server/service only. It was MANY!

HA with Virtual Center as VM with Distributed Switches

Useless1 — Mon, 24 Aug 2009 10:06:06 GMT

Hi,

I have had an issue with this and was wondering what people would recommend I do..

I have 2 hosts with 6 NIC's

2 NIC's for DMZ in a distributed switch

2 NIC's for Production in a distributed switch

2 NIC's for Service Console and vMotion in a local switch per host (Can these be moved? How? Is it wise?)

Now I have my virtual center VM running inside, connected to the Production switch. I turn off a blade (To simulate disaster), however it does not migrate quoting an error. I have since discovered that whilst HA information is setup with VC but can run without Virtual Center, distributed switches do not even appear when it is off... which is why I think the move failed (Moving it to a portgroup on the local switch and configuring this portgroup across all hosts works though)...

With this in mind, what would people think I should do?

Separate physical box otuside cluster
Separate physical box with vmware standalone with single VC VM installed
Separate physical box with vmware standalone with single VC VM installed with another linked VC in a VM
Just connect the VC to my "service" switch?

Ideas?

ESX 4 ; evaluation license ; HA not working

matoo — Wed, 17 Jun 2009 13:41:35 GMT

Hello VMware community ,

I'm testing new ESX 4 and its HA behavior ( 2 node cluster). I have a hard time with vmware HA - VM doesn't switch (failover) to other ESX server in cluster.
My environment:

hostA, hostB: 2x ProLiant DL580 G5 , each with Intel(R) Xeon(R) CPU X7460 @ 2.66GHz with 32 GB ram
1x shared storage - HP EVA - both servers see this share storage
client: 1x standalone windows machine (vSphere installed on it) (client)
ESX 4 - evaluation copy ( business licenses on the way)

All 3 servers are on the same NWs. I use vSpehre on client to mange both ESX servers. I put them into cluster ( HA has been configured, default values have been assigned). I have two test VMs (windows 2003) inside the cluster.I tested VMotion - worked perfectly - I was able to move VMs from hostA to hostB and vice versa.

Problem is, when i reboot either hostA or hostB ( to test failover) - no failover happens. I see hostA in "not responding" state , VM is in disconnected state. First time i tried this (after cluster configuration) I got no error on console. Second time i tried this i got an error of unsufficient resources to satisfy HA failover. I checked the HA configuration:

Current Failover Capacity: 1
Configured Failover Capacity: 1

VM machine has only 1024MB configured (no limits set) and it's located on shared storage ( 20GB lun, 5GB is allocated for system). I tried several things (reconfigure cluster again, turn off admission control, checked DNS between ESX/vShpere, checked if I have a VMware HA among license features, .. ) - nothing helped.

I'm king of stuck here and would appreciate and hints I've (probably) missed.
Thanks!

FT and Nested VM

76dragon — Thu, 20 Aug 2009 09:26:52 GMT

Ive got 2 x Virtual vSphere servers running on top of vSphere, Ive created a virtual machine (2003 server with Vmware tools installed) and im trying to enable FT but getting the error about how the disk needs to be scrubed.

When I shut down the VM and try to perform the enable FT to scrub the disk it gets to 28% and has a spaz, and I get a "Error occured" and does not complete.

I have "replay.allowBTOnly= True" set

Is there anything else im missing ?

HA or FT across a WAN?

Tizodm80 — Wed, 19 Aug 2009 18:39:01 GMT

Hey guys,

I have a customer that needs a HA or FT type solution with a very small datacenter of 1 ESX Host and 5 Servers. We are looking at VSphere 4 Advanced as it includes FT. He has 2 offices, one in TX and one in MD. They want to use FT in place of say a cluster or doubletake for their exchange VM. My question is can FT or HA operate over a WAN or is it required that they stay local?

Thanks,

Todd

Problem with Hard Shutdown and Guest OS - VMotion issue

edv2009 — Wed, 19 Aug 2009 01:17:15 GMT

Hello All! I am having an issue and hope I can find some help......

Basically we are in the middle of testing HA and VMotion and we have found that when we do a hard shutdown on one of our ESX 4 boxes (simulating loss of an ESX box), the Guest OS goes offline and shuts down on the ESX host we shut off. It does VMotion off to another ESX Host in our environment and then restarts the guest VM back up..... Anyone have any ideas of why it is doing this? It should be keeping it alive, right?

I setup several ESX 3.5 environments and never had this issue and hoping maybe I am missing something in ESX 4.

Thanks!

High availability iSCSI SAN guidance

isz — Tue, 18 Aug 2009 05:33:47 GMT

I am looking for some guidance or a recommendation to eliminating single point of failure down to the SAN which provides storage to vSphere environment.

Previously I looked into OpenFiler because of its iSCSI accommodations, because it is Open Source, and because is comes packaged with DRBD and Heartbeat. While I was able to configure OpenFiler as a HA SAN, it seemed brittle and performing a failback along with other issues proved it would not be a dependable solution.

Is there a recommended alternative in a Windows or Linux environment that would accommodate high availability? What are some options that would allow two identical physical servers with specs mentioned below to provide redundancy if one storage server were to fail? Does Windows Server 2k8 have a block replication feature that I am unaware of?

Shared storage where an iSCSI target server in cluster mode that defers its storage to a separate physical SAN just seems like it is deferring the single point of failure further downstream.

Specs (2x):

2U
3Ware 9550SXU
8gig memory
8 drives (4 raid1 configuration)
Two nic gigabit
AMD Opteron
Asus M2N WS Pro
W2k8 (Storage Server, or whatever)

Thanks for your participation!

Timeout when trasferring a powered on vm's datastore

isz — Tue, 18 Aug 2009 05:17:28 GMT

I have attempted transferring a couple different virtual machines within vSphere vClient while powered on and have noticed that 2 of 3 failed to transfer on repeated attempts. I read that if this occurs it is best to just power down the virtual machine and transfer the datastore to its new home this way. While the method did work for me I was a bit nervous because I did not have adequate insight into the progress of the transfer.

The biggest concern I had was that I could not see at what point (how far into the transfer) or what file may have caused the timeout. All snapshots were previously deleted. My environment has been built around a Windows 2008 Storage Server iSCSI target. The luns I have exposed are attached to the virtualized hosts, and the best insight into my storage appears to be through selecting Home > Inventory > Datastores and then browsing a selected datastore. But this doesn't give me detailed file information.

Is there a proficient way to access the data within the LUN either directly or through the iSCSI target so that I can monitor the real-time transfer of data from one datastore to another? Perhaps some ssh trick, or shell command?

Thanks for your participation!

FT secondary vm power on failure

acollado175 — Mon, 17 Aug 2009 17:33:07 GMT

I had a vm running with FT in a two node HA cluster. To demonstrate failover, I simulated power failure on the node running the FT vm (by holding the power button), and viola, vm failed over and didn't skip a beat.

Apparently, power failure is a bad option to demonstrate FT since i destroyed my ESXi host and had to repair. Repair went well, rejoined my cluster, reconfigured HA, but FT will not properly configure. I have turned FT off and back on my vm, but when I go to power on, the primary powers up fine, but the secondary vm gives this error:

"Unable to access file <unspecified file> since it is locked."

My Fault Tolerance Status shows "Not protected. Need seconday vm"

I'm running in eval mode on an NFS share.

Has anyone seen this behavior? Why can't vcc spit out what file is locked?

FT Scrubbing Error

scottiewin — Fri, 14 Aug 2009 08:20:12 GMT

Hi All,

We have an issues regarding Scrubbed Disks and FT

When attempting to FT a machine we get the following message (attached).

Has anyone seen this before, can anyone point us in the right direction?

Many Thanks

Scott

FT Scrubbing Error

scottiewin — Fri, 14 Aug 2009 08:10:34 GMT

Hi All,

We got an odd message today (attached) when we attempted to FT a machine.

"Unsupported Virtual Machine Configuration for Fault Tolerance. The unused disk blocks of the virtual machine's disks have not been scrubbed on the filesystem. This is needed to support features like Fault Tolerance.

Anyone seen this? Can any point me in the reight direction to relove it?

Many Thanks

Scott

BFS and HA

Gordon Freeman — Fri, 07 Aug 2009 16:15:14 GMT

Our current vSphere hosts are configured as BFS. We recently had an issue where one of the fiber cables on a host was jostled and appeared to be intermittently connecting. The host showed disconnected in vCenter. I completely shut down the host but none of the guests were HA'ed off this host. All guests showed as disconnected in vCenter. I eventually had to remove the host from inventory and manually import each guest back into vCenter. My question is: What is the expected behavior of HA with BFS hosts? My understanding is that if the BFS host loses both paths, this should initiate HA for the guests on the host, as this would be the same as a service console failure. Because the host in this example was intermittently connecting, perhaps that might not have been the case initially. But when I completely shut down the host, this should have initiated HA of the guests, no?

They're waiting for you Gordon... in the test chamber.

VMware HA with 2 different nodes...

Sergio E. — Tue, 11 Aug 2009 23:32:31 GMT

Hi Folks,

This question was made by one of my co-workers, and both nodes share the same overall configuration (# of cores, NICs and RAM) but different processors (old Xeon Vs. Nehalem). I think it may be possible since the VMs being carried over to the surviving server will be re-booted, evading potencial CPU specific (Vmotion) issues. The nodes will be running Essentials Plus licensing, storage is a small SAN.

Thanks in advance for any insight!

FT giving mac address conflicts on secondary

rockowwc — Sun, 09 Aug 2009 11:27:06 GMT

I've had a look through these forums, and currently running vmxnet2 (enhanced) NIC as suggested.

However it still errors out when trying to start the secondary VM with a mac address conflict. I have the mac on the original set to automatic..

What are the requirements\limitation for configuring Fault Tolerance?

theblackknight — Mon, 10 Aug 2009 21:06:32 GMT

What do I need to get it configured correctly?

How does restart vm in different esx host in HA work?

vhii81 — Mon, 10 Aug 2009 04:22:48 GMT

I understand that VMOTION = live migration of vms across esx hosts. So, if HA doesnt use VMOTION, how does HA restart vm on a another esx host when the original one is down?

I suppose it will have to create another vm on the available host using the same .vmx file on the san?

IF we were to do this manually, it will just be an import virtual machine to another host right?

Correct me if im wrong?

Just a bit confused on how HA works behind the scene

Cheers!

Nics needed for FT

tjg091 — Sat, 08 Aug 2009 01:04:17 GMT

I am trying to understand how the additional NIC's are used and configured for a typical vSphere FT installation. I understand that at least 3 NIC's are needed per server but please someone explain how they connect. ESX Management/VMotion, virtual machine traffic, FT logging are described as the services that need the Nics. Are multiple vlans needed on a switch that these services connect to? All we are trying to do is to have 2 servers connected to an EMC Clarion SAN.

Thanks in advance

Does anyone tried Zenoss to monitor vSphire 4 Environment

sjadapa — Wed, 05 Aug 2009 11:49:31 GMT

Hi VM Gurus,

I am desperately wiating to hear from some one on the implementation of Zenoss Monitoring tool to mornitor vSphere environment ( vmotion, DR etc )

Any inputs and suggestions will be highly appreciated.

****If you find this or any other answer useful please consider awarding points by marking the answer CORRECT or HELPFUL ****

Jadapa RHCE, MCSA

VMWARE HA and shared storage

george78 — Wed, 05 Aug 2009 18:40:52 GMT

Dear Friends

I have two questions regarding vmware HA.

1) Can vmware HA be used with NFS (e.g a linux pc with nfs shares)

2) If i understood well vmware HA provides recovery among different ESX hosts sharing the same vmdk files over a shared storage like iSCSI or SAN or NAS. What if the shared storage fails? Can somebody tell me how we can provide HA for the shared storage?

Thanks

vCenter Server running as a VM and being FT

billm1974 — Tue, 04 Aug 2009 04:56:37 GMT

Hi Guys, i just wanted to know, is it possible to have the vCenter Server running as a virtual machine, and have fault tolerance enabled for that virtual machine?

Are there failover testing & benchmark tools?

nicholas1982 — Wed, 05 Aug 2009 11:13:05 GMT

I currently have a cluster of three ESXi hosts, running on IBM HS21XM, 4Gb FC HBAs There are 4 sets of Cisco MDS9000 FC switches, two in the BladeCenter chassis basically used as pass through connected into 2 core external MDS9000 FC switches, zoned to the best of our knowledge.We are using dual controller IBM DS3400 Active / Passive SAN each SP with two ports.

I would like to conduct some failover testing while having some VMs running.

The tests I plan to run will consist of disabling HBA ports to simulate HBA failover, I plan to power cycle FC switches and power cycle SAN storage processors.

I have already conducted some of theses tests however there was only with one host (so not in a cluster) and a couple of VMs not really doing much, everything seemed to run ok, however I beleive a true test will consist of a cluster of hosts and a number of guest os's performing some real workload, so I would like to know if anyone could suggest any good tools that can be used within a guest os (Windows) to put some stress on the SAN and benchmark and report disk i/o errors.

Our real main concern is data corruption say when running Exchange or SQL, if a DB or mail server lose disk access for a certain periods of time for example HBA or SP failover how does ESX handle this?

Any info or guidence on this is much appreciated.

Regards

Nicholas

Fault Tolerance feature in vSphere

azn2kew — Tue, 21 Apr 2009 17:11:10 GMT

Its a great feature after watching live demo, but how is licensing works in terms of SQL, Exchange or Oracle servers where we have them cloned so are we going to pay 2x SQL, Exchange, Oracle etc...? that's a lot of money! anyone knows?

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

HA and local storage only

Ptitluc — Tue, 04 Aug 2009 09:01:38 GMT

Hi,

I evaluate vSphere Essentiel Plus.

I have 3 ESX 4 hosts Proliant DL380G4

Configured in a cluster with vCenter on a Windows 2003 host.

I would like to know if HA is available without shared storage (SAN)

My ESX hosts only have local storage

How I can replicate VM between Hosts ?

Thanks in advance

Luc

Maximum Availability

flyaway — Fri, 31 Jul 2009 05:08:48 GMT

Sorry if this has been asked before.

Looking to make our VM Machines available 24x7. If host server was to fail, how do you (what product?) makes the VM machine failover to another server?

I have heard the words vRanger, Veeam, HA, and vSphere, just not too sure what I would need without buying the wrong product.

Thanks

Are you using FT or want to use it?

bchanana — Fri, 31 Jul 2009 08:33:32 GMT

I'm looking for data from customers who are currently using FT or thinking about deploying it. Here's what we're looking for:

an export of the CPU utilization of a single VM, for which you want to turn on FT (or have already), during the most busy times for that VM - smaller intervals is better (e.g. 1 hour vs 1 week). For example, an export of CPU utilization from an Exchange VM at 9-10am when user load is highest or the VM is busiest. You can get this from vCenter.

with that export, please let me know what application is running inside the VM (e.g. Exchange 2007, Oracle 10g, etc.)

Please feel free to email me directly with this information (bchanana@vmware.com) and feel free to send samples from multiple VMs of different types (e.g. one from Exchange, one from SQL, etc). The data can be from any version of vSphere and/or VI and will only be used in aggregate form, you do not have to provide the company or identifying information with it, feel free to keep it anonymous.

First HA Cluster .. Will this setup work for HA?

s1xth — Tue, 28 Jul 2009 18:40:42 GMT

I am in the process of configuring my first HA cluster, I am starting out with a two host cluster and then add another host to make it a three host cluster. If I have the following configuration on my hosts, will HA function and be in admittance? I am trying to decide if I should go with 24Gb or 32Gb or ram and if it would help the HA.

Thanks!

Host 1 Dual Quad Core w/ 24GB Ram (maybe 32Gb?)

1 VM w/ 2vCPU 8Gb Ram(MIGHT be a 4vCPU)

1 VM w/ 1vCPU 4Gb Ram

Host 2 Dual Quad Core w/24GB Ram (maybe 32Gb?)

1 VM w/ 4 vCPU 4Gb Ram

1 VM w 1 vCPU 4Gb Ram

1 VM w/ 1 vCPU 4GB Ram

on-demand FT, can this be automated?

Vliegenmepper — Thu, 30 Jul 2009 09:01:32 GMT

Hi,

VMware mentions that since FT is so easy to enable this can easily be used for on-demand FT.
Question that I have, is there a way to automate the enabling of FT? (it isn't an option in the default VC scheduler)

Maybe this needs to be done via PowerCli?

Thanks!
Kenneth van Ditmarsch