VMware Cloud Community
BICh2
Enthusiast
Enthusiast
Jump to solution

Host unreachable from VMs

We have 2 ESXi Host 5.5, and vCenter is a VM on Host2. The following are the versions of our VMWare setup:

ESXi 5.5.0 - 1623387 (Both the Hosts)
vCenter 5.5.0 - 9911210

When we start Veeam Backup it fails to connect to Host2 from cluster, we cannot ping Host2 from all VMs but can ping Host1 and all the VMs on both Hosts. All are using VLAN1 (192.168.1.0). All LAN workstations can ping Host2 (my PC IP 192.168.3.11 can ping Host2). 

Also found that sometimes VMs on Host1 can ping Host2, but VMs on Host2 cannot ping Host2 itself. So testing ping between both the Hosts using command vmkping.

We tried the following without success:

  1. Restart Management Agent on Host2
  2. Restarted All Services
  3. Restarted Host2
  4. Changed MTU to 9000

Removed the Network cable from vmnic0, and it started pinging. It worked well until we had to restart the Hosts due to the power interruption. We removed the cable from vmnic0 again, but this time it's not pinging at all.

Is it replaced with vSwitch or Cisco switch? cannot understand why we can ping from outside the vSwtich, but not within the vSwitch.

 

Now totally lost on which part needs to be fixed.


Licenses are not renewed, so we cannot upgrade or contact vmware support.

Labels (2)
Reply
0 Kudos
2 Solutions

Accepted Solutions
CallistoJag
Hot Shot
Hot Shot
Jump to solution

Correct, this is also the recommended setup (should be able to work on same vlan, but I would suggest you try this).

View solution in original post

BICh2
Enthusiast
Enthusiast
Jump to solution

I tried changing the IP, but i didn't restart the management network.

So if failed, wasn't pinging the new IP address.

Reverted back the original IP address, but this time Restarted the management network.

And it worked.

View solution in original post

Reply
0 Kudos
17 Replies
CallistoJag
Hot Shot
Hot Shot
Jump to solution

Can you share the 2 vswitch configurations? Ip conflict?
Reply
0 Kudos
BICh2
Enthusiast
Enthusiast
Jump to solution

Not very good at ESXi cli, can provide the commands to get this information

Reply
0 Kudos
CallistoJag
Hot Shot
Hot Shot
Jump to solution

Retrieving Information About Virtual Switches with ESXCLI
You can retrieve information about virtual switches by using esxcli network vswitch commands.

Specify one of the options listed in Connection Options for ESXCLI Host Management Commands in place of <conn_options>.

List all virtual switches and associated port groups.
esxcli <conn_options> network vswitch standard list
The command prints information about the virtual switch, which might include its name, number of ports, MTU, port groups, and other information. The output includes information about CDP settings for the virtual switch. The precise information depends on the target system. The default port groups are Management Network and VM Network.
List the network policy settings, such as security policy, traffic shaping policy, and failover policy, for the virtual switch. The following commands are supported.
esxcli <conn_options> network vswitch standard policy failover get
esxcli <conn_options> network vswitch standard policy security get
esxcli <conn_options> network vswitch standard policy shaping get
Reply
0 Kudos
BICh2
Enthusiast
Enthusiast
Jump to solution

I guess this is the Host2 vswitch configuration:


~ # esxcfg-vswitch -l
Switch Name Num Ports Used Ports Configured Ports MTU Uplinks
vSwitch0 3312 12 128 1500 vmnic0,vmnic1

PortGroup Name VLAN ID Used Ports Uplinks
VM Network 10 6 vmnic0,vmnic1
Management Network 10 1 vmnic0,vmnic1

Switch Name Num Ports Used Ports Configured Ports MTU Uplinks
vSwitch1 3312 5 128 1500 vmnic2

PortGroup Name VLAN ID Used Ports Uplinks
EXCHANGEREP 0 1 vmnic2
VMOTION 0 1 vmnic2

~ #

Reply
0 Kudos
BICh2
Enthusiast
Enthusiast
Jump to solution

the outputs:


~ # esxcli network vswitch standard list
vSwitch0
Name: vSwitch0
Class: etherswitch
Num Ports: 3312
Used Ports: 12
Configured Ports: 128
MTU: 1500
CDP Status: listen
Beacon Enabled: false
Beacon Interval: 1
Beacon Threshold: 3
Beacon Required By:
Uplinks: vmnic1, vmnic0
Portgroups: VM Network, Management Network

vSwitch1
Name: vSwitch1
Class: etherswitch
Num Ports: 3312
Used Ports: 5
Configured Ports: 128
MTU: 1500
CDP Status: listen
Beacon Enabled: false
Beacon Interval: 1
Beacon Threshold: 3
Beacon Required By:
Uplinks: vmnic2
Portgroups: EXCHANGEREP, VMOTION
~ #
~ #
~ #
~ # esxcli network vswitch standard policy failover get -v vSwitch0
Load Balancing: srcport
Network Failure Detection: link
Notify Switches: true
Failback: true
Active Adapters: vmnic0, vmnic1
Standby Adapters:
Unused Adapters:
~ #
~ #
~ #
~ # esxcli network vswitch standard policy failover get -v vSwitch1
Load Balancing: srcport
Network Failure Detection: link
Notify Switches: true
Failback: true
Active Adapters: vmnic2
Standby Adapters:
Unused Adapters:
~ #
~ #
~ #
~ # esxcli network vswitch standard policy security get -v vSwitch0
Allow Promiscuous: false
Allow MAC Address Change: true
Allow Forged Transmits: true
~ #
~ #
~ #
~ # esxcli network vswitch standard policy security get -v vSwitch1
Allow Promiscuous: false
Allow MAC Address Change: true
Allow Forged Transmits: true
~ #
~ #
~ #
~ # esxcli network vswitch standard policy shaping get -v vSwitch0
Enabled: false
Average Bandwidth: -1 Kbps
Peak Bandwidth: -1 Kbps
Burst Size: -1 Kib
~ #
~ #
~ #
~ # esxcli network vswitch standard policy shaping get -v vSwitch1
Enabled: false
Average Bandwidth: -1 Kbps
Peak Bandwidth: -1 Kbps
Burst Size: -1 Kib
~ #
~ #
~ #
~ #

Reply
0 Kudos
CallistoJag
Hot Shot
Hot Shot
Jump to solution

Firstly the MTU here is 1500 not the 9000 you mentioned. Also you seem to
be using the VLAN10 not VLAN1 for the management and VM networks. Can you send the same info for the first host?
Reply
0 Kudos
CallistoJag
Hot Shot
Hot Shot
Jump to solution

Can you try putting the uplinks into active standby for VM and Management, test and then try swapping the active uplink? See if there is an issue with one of the uplink cards?
Reply
0 Kudos
BICh2
Enthusiast
Enthusiast
Jump to solution

The MTU was returned back to 1500, and VLAN1 - might have missed the "0".

These are the Host1 output

~ #
~ #
~ # esxcfg-vswitch -l
Switch Name Num Ports Used Ports Configured Ports MTU Uplinks
vSwitch0 3312 13 128 1500 vmnic0,vmnic1

PortGroup Name VLAN ID Used Ports Uplinks
VM Network 10 7 vmnic0,vmnic1
Management Network 10 1 vmnic0,vmnic1

Switch Name Num Ports Used Ports Configured Ports MTU Uplinks
vSwitch1 3312 5 128 1500 vmnic2

PortGroup Name VLAN ID Used Ports Uplinks
EXCHANGEREP 0 1 vmnic2
VMOTION 0 1 vmnic2

~ #
~ #
~ # esxcli network vswitch standard list
vSwitch0
Name: vSwitch0
Class: etherswitch
Num Ports: 3312
Used Ports: 13
Configured Ports: 128
MTU: 1500
CDP Status: listen
Beacon Enabled: false
Beacon Interval: 1
Beacon Threshold: 3
Beacon Required By:
Uplinks: vmnic1, vmnic0
Portgroups: VM Network, Management Network

vSwitch1
Name: vSwitch1
Class: etherswitch
Num Ports: 3312
Used Ports: 5
Configured Ports: 128
MTU: 1500
CDP Status: listen
Beacon Enabled: false
Beacon Interval: 1
Beacon Threshold: 3
Beacon Required By:
Uplinks: vmnic2
Portgroups: EXCHANGEREP, VMOTION
~ #
~ #
~ # esxcli network vswitch standard policy failover get -v vSwitch0
Load Balancing: srcport
Network Failure Detection: link
Notify Switches: true
Failback: true
Active Adapters: vmnic0, vmnic1
Standby Adapters:
Unused Adapters:
~ #
~ #
~ # esxcli network vswitch standard policy failover get -v vSwitch1
Load Balancing: srcport
Network Failure Detection: link
Notify Switches: true
Failback: true
Active Adapters: vmnic2
Standby Adapters:
Unused Adapters:
~ #
~ #
~ # esxcli network vswitch standard policy security get -v vSwitch0
Allow Promiscuous: false
Allow MAC Address Change: true
Allow Forged Transmits: true
~ #
~ #
~ # esxcli network vswitch standard policy security get -v vSwitch1
Allow Promiscuous: false
Allow MAC Address Change: true
Allow Forged Transmits: true
~ #
~ #
~ # esxcli network vswitch standard policy shaping get -v vSwitch0
Enabled: false
Average Bandwidth: -1 Kbps
Peak Bandwidth: -1 Kbps
Burst Size: -1 Kib
~ #
~ #
~ # esxcli network vswitch standard policy shaping get -v vSwitch1
Enabled: false
Average Bandwidth: -1 Kbps
Peak Bandwidth: -1 Kbps
Burst Size: -1 Kib
~ #

Reply
0 Kudos
BICh2
Enthusiast
Enthusiast
Jump to solution

Tried moving both the NICs to standby, it worked for around 45mins then it stopped, and also swapping the Nics didn't change the state.

Do the vSwitches configuration show any abnormalities? 

I have tried using an app called arp-ping.exe that shows the mac-address with the ping result, when I ping using this app, I can get the reply, but the mac-address is of the Cisco switch port that is connecting to the firewall. But the regular Windows ping is timed out.

Reply
0 Kudos
RajeevVCP4
Expert
Expert
Jump to solution

Is host pingable from where is the VEAM installed 

Did you try to check compatibility of vSphere with veam 

Rajeev Chauhan
VCIX-DCV6.5/VSAN/VXRAIL
Please mark help full or correct if my answer is use full for you
BICh2
Enthusiast
Enthusiast
Jump to solution

When it stops pinging, it stops from veeam too.

Veeam is compatible as we cannot install the newer version of Veeam 12 which is not compatible with ESXi 5.5. Veeam is Ver11 on Windows Server 2012.

Reply
0 Kudos
CallistoJag
Hot Shot
Hot Shot
Jump to solution

Can I suggest that you put the management and VM traffic on different vlans? Trunk the network on the physical and tag at the vswitch. Let me know if this helps?
Reply
0 Kudos
BICh2
Enthusiast
Enthusiast
Jump to solution

So you are suggesting to try putting Management Network to another VLAN (say VLAN100 -subnet 192.168.100.0) and VM Network to be in VLAN10

Let me be clear that you are suggesting to try putting Management Network to another VLAN and VM Network on a different VLAN.

So the Management Network should be on VLAN100 (say VLAN100 -subnet 192.168.100.0) which has the Host IP addresses.
This will be configured on both Hosts.

Reply
0 Kudos
CallistoJag
Hot Shot
Hot Shot
Jump to solution

Correct, this is also the recommended setup (should be able to work on same vlan, but I would suggest you try this).
BICh2
Enthusiast
Enthusiast
Jump to solution

If I change both hosts' IP addresses, will the VMs be able to communicate on the LAN? Do I need to restart the Host after changing the IP addresses?

Reply
0 Kudos
BICh2
Enthusiast
Enthusiast
Jump to solution

Changed the Host2 Management VLAN to All (4095), and we could ping from any system.

REverted back to VLAN 10.

Reply
0 Kudos
BICh2
Enthusiast
Enthusiast
Jump to solution

I tried changing the IP, but i didn't restart the management network.

So if failed, wasn't pinging the new IP address.

Reverted back the original IP address, but this time Restarted the management network.

And it worked.

Reply
0 Kudos