- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hey,
I'm having an issue at the moment adding ESXI hosts Via their Natted IP address to vCenter VCSA 6.5.
Previously I have successfully used this KB successfully Using NAT between the vCenter Server system and ESXi/ESX hosts (1010652) | VMware KB for all previous version of vCenter.
The Server IP address in the vpxa.cfg file is correct and port 902 is open and telnet works
I have increased the config.vpxd.heartbeat.notRespondingTimeout via this KB
Here is a tail of the vxpa agent log
2017-07-10T19:00:41.341Z verbose vpxa[8ED0B70] [Originator@6876 sub=VpxaHalCnxHostagent opID=WFU-376885b1] Received WaitForUpdatesDone callback
2017-07-10T19:00:41.341Z verbose vpxa[8ED0B70] [Originator@6876 sub=VpxaHalCnxHostagent opID=WFU-376885b1] Applying updates from 479 to 480 (at 479)
2017-07-10T19:00:41.341Z verbose vpxa[8ED0B70] [Originator@6876 sub=VpxaHalCnxHostagent opID=WFU-376885b1] Starting next WaitForUpdates() call to hostd
2017-07-10T19:00:41.341Z verbose vpxa[8ED0B70] [Originator@6876 sub=VpxaHalCnxHostagent opID=WFU-376885b1] Completed WaitForUpdatesDone callback
2017-07-10T19:00:41.422Z verbose vpxa[8F54B70] [Originator@6876 sub=VpxaHalCnxHostagent opID=WFU-2f2143e2] Received WaitForUpdatesDone callback
2017-07-10T19:00:41.422Z verbose vpxa[8F54B70] [Originator@6876 sub=VpxaHalCnxHostagent opID=WFU-2f2143e2] Applying updates from 480 to 481 (at 480)
2017-07-10T19:00:41.422Z verbose vpxa[8F54B70] [Originator@6876 sub=VpxaHalCnxHostagent opID=WFU-2f2143e2] Starting next WaitForUpdates() call to hostd
2017-07-10T19:00:41.422Z verbose vpxa[8F54B70] [Originator@6876 sub=VpxaHalCnxHostagent opID=WFU-2f2143e2] Completed WaitForUpdatesDone callback
2017-07-10T19:00:42.268Z warning vpxa[8E8EB70] [Originator@6876 sub=Heartbeat opID=SWI-1c06dac8] Failed to bind heartbeat socket for host address 10.12.X.X: Cannot assign requested address
2017-07-10T19:00:45.946Z verbose vpxa[8F33B70] [Originator@6876 sub=VpxaHalCnxHostagent opID=WFU-30efc29a] Received WaitForUpdatesDone callback
2017-07-10T19:00:45.947Z verbose vpxa[8F33B70] [Originator@6876 sub=VpxaHalCnxHostagent opID=WFU-30efc29a] Applying updates from 481 to 482 (at 481)
2017-07-10T19:00:45.947Z verbose vpxa[8F33B70] [Originator@6876 sub=halservices opID=WFU-30efc29a] [VpxaHalServices] HostChanged Event Fired, properties changed [runtime.healthSystemRuntime]
2017-07-10T19:00:45.947Z verbose vpxa[8F33B70] [Originator@6876 sub=vpxaInvtHost opID=WFU-30efc29a] [HostChanged] Found update for tracked MoRef vim.HostSystem:ha-host
2017-07-10T19:00:45.947Z verbose vpxa[8F33B70] [Originator@6876 sub=VpxaHalCnxHostagent opID=WFU-30efc29a] Starting next WaitForUpdates() call to hostd
2017-07-10T19:00:45.947Z verbose vpxa[8F33B70] [Originator@6876 sub=VpxaHalCnxHostagent opID=WFU-30efc29a] Completed WaitForUpdatesDone callback
2017-07-10T19:00:46.270Z warning vpxa[8E8EB70] [Originator@6876 sub=Heartbeat opID=SWI-1c06dac8] Failed to bind heartbeat socket for host address 10.12.X.X: Cannot assign requested address
10.12.X.X is the host natted ip address.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
VPXA.log is not sufficient. Can we also take a look at /var/log/vmkernel.log, var/log/hostd.log, /var/log/vmkwarning.log on the ESXi host that we are failing to add to vCenter? Also as per KB Using NAT between the vCenter Server system and ESXi/ESX hosts (1010652) , using NAT between vCenter and ESXi host is unsupported.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Aishr
thank you for replying - I'm aware it's not supported it a temporary connection in order to use VC to run some PowerCli over the temporarily connected hosts. I Managed to fix the issue afterward.
I had to create a new VMK with the natted IPaddress as the interface IP ie 10.12..X. Xand stability improved.
"Failed to bind heartbeat socket for host address 10.12.X.X: Cannot assign requested address" error were eliminated
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi cliffcahill,
We are facing the same issue, Can you please share the steps you had done to resolve this issue ?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Fix was to create a VMK on vSwitch. Use the ESXi host natted ip address as the VMK interface address. No vlan is required.
Hope this help.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
cliffcahill
Yes.It worked.
In addition,we added this line as well.
1) We added this line in vpxa.cfg "<preserveServerIp>true</preserveServerIp>".After configuring this line,VC Public IP address will not be changed.
2) We added the NAT public IP address in vpxa.cfg "<serverIp>212.*.*.*</serverIp>"
3) And we created a new vmkernel port group(with new vSwitch and no VLAN,no nic binding needed) with ESXi NAT public IP address.
it worked well and thanks much for the inputs.
Thanks,
Manivel RR