vCenter 6 fails to restart after IP change

[darkdragon001]

I changed the vCenter 6 IP from within web client.

Since I had problems connecting to the new IP, I restarted the server via vSphere Client directly on the host using VMware Tools.

After the restart, I can't connect to vCenter any more. Trying to connect via https shows me the following error:

503 Service Unavailable (Failed to connect to endpoint: [N7Vmacore4Http20NamedPipeServiceSpecE:0x7f4158068bf0] _serverNamespace = / _isRedirect = false _pipeName =/var/run/vmware/vpxd-webserver-pipe)

Then I directly used the console, enabled ssh and bash and logged in there.

I tried to check the current IP settings via the script used in version 5.5 which still exists (/opt/vmware/share/vami/vami_config_net) but fails with multiple errors.

Then, I tried to reset the certificates via certificate manager (/usr/lib/vmware-vmca/bin/certificate-manager), but I could not login via SSO (seems that SSO service could also not be started).

Then, I had a look in the log files, and noticed a problem with vmware-vpxd:

VC SSL certificate does not exist, it will be generated by vpxd

This did not help either. I tried to manually start the service, with the same result.

How can I fix my vCenter?

[AlexeySav]

try to change ip in DNS and /etc/hosts on vcenter. ping vcenter FQDN from shell must work.

[gjbrown]

This worked for me, thanks for the help.

[HaniIbrahim]

login to appliance management on port 5480 and enable ssh login & bash access. login with ssh and drop into the shell, edit the /etc/hosts file with vi and change the ip address

[Talon256 · Accepted solution]

I got this error after changing the vCenter's IP address

1. I used #6 https://rlevchenko.com/2016/03/24/vcenter-503-service-unavailable/ to enable the shell.
2. ssh into the vcenter server
3. Command> shell
4. vi /etc/hosts changed the IP there
5. service-control --stop --all
6. service-control --start --all

That got me fixed

[CPUNeck]

** make sure the vCenter hostname only has what ever was originally entered during installation **

Late, but hopefully useful for others.  I ran into this today from similar activities (changed IP, name, etc.)  After realizing there's no change flexibility, I reinstalled with certain, permanent values to prevent this from happening again.  This appliance runs alone in a v6 host (HP blade.)  All worked well, even rebooted a couple of times to ensure it'd remain functioning.  Then I made an atypical network change, after-witch my vCenter broke.  The change was at the HP C7000 chassis @ layer 2.

Initially networks were provided to the blades via shared up-link sets (SUS) containing ~ 3 vlans over the first 2 LOM's (on-board NIC's) --> vSwitch0 with vlan support.

Change..., configured same 2 LOM's to "tunnel" the vlans to the host --> vSwitch0.  Both configurations use virtual switch tagging (VST) and should have no effect.  After everything came up and I realized the vCenter was borked, simple troubleshooting proved the network config was solid... I could ping interfaces on all 3 vlans (one which was on the vCenter appliance.)  Pulled hair out, scoured I-net, and searched logs.  Happened to notice the appliance's hostname was populated with its FQDN, generally no big deal - but I'm certain I entered it with only the hostname (just built the dang thing for the ##?? time) So;

I manually edited the hostname from the console back to just the hostname and rebooted... everything came back functioning as designed.  I'm going to keep this thing in dev a bit longer... not sure it's ready for more than playing with at this point, to fragile with little information available for troubleshooting. 

[ausafali88]

dear experts;

i changed nothing...not an IP etc....but it still shows this error....gave a static IP because need ESXi with vCenter for a home lab setup.....

Pls can someone from VMWare post whether this is a known bug or something.....have even heard that it effects ESXi 6U2 also....

Regards

[MBrownWFP]

This process worked for me. Thanks for the tip about the hosts file.

[Portwajn1337]

Thanks, this solved my issue too.

Exact steps for editing hosts file are laid out in Talon256 posts (no.23 in this thread) so kudos go to you both.

[Portwajn1337]

Thanks a lot for a step-by-step guide, much appreciated! This resolved my issue, too.

Please note, that I did not restart all services, just vSphere web client service as in rlevchenko.com guide (cited), and it didn't work apparently, thus I rebooted entire appliance. The vsphere-client didn't work at first, then showed that is 'preparing itself' and only after about 5-7 mins from appliance start the website was available; so be patient.

[ManSolongines]

I had the same problem, saw the same error messages as reported by all. However, in my case, it was a time drift problem on the ESXi host that created the problem. To fix it, I had to:

1) ssh into the ESXi host

2) using the instructions at: How to Change the ESXi System Time & HW Clock on the CLI  I changed the system clock.

3) restarted the VCSA from vSphere client and all is working fine now.

So as FYI for those that may face a timing issue, you may see the same error message.

[ShDog]

I had this exact problem.  I changed the IP addresses on two appliances.  One using the console.  The other using the VAMI. Both had this issue starting vmware-vpxd.  I thought it was certificate related but that was not the case.

When you change the IP, the hosts file is not updated.  From the console or CLI, vi /etc/hosts and update the value for the appliance.  Restart the appliance and it should start up normally.

SD