VMware Cloud Community
kasperottesen
Contributor
Contributor

A general system error occurred: internal error: vmodl.fault.HostCommunication

Hi,

Yesterday one of my hosts in a HA Cluster suddenly went into disconnected status. I tried to connect it again but got the above message in a popup. I tried to restart mgmt-vmware and vmware-vpxa agents and then tried to connect it again, same error.

I have tried to manual uninstall and install vpx agent but still no help.

I have checked DNS and network settings, everything looks fine. I cant reboot the host since I have production VMs running on it.

Anyone seen this before and came up with a solution?

Detais about my setup:

vCenter 4.0

ESX 3.5 U2

- Kasper

Reply
0 Kudos
18 Replies
Shakayumi
Contributor
Contributor

Hello,

You can check your /VAR/LOG repository and see what happen ?

Please post results

Thanks

Reply
0 Kudos
kasperottesen
Contributor
Contributor

Yeah sure, here are som copy from hostd.log. I know there are SSL Exception, this one started after I tried to rebuild the certifcates.

Let me know if you want to see more logs or a specific log file. I forgot to say that i tried to copy these to lines into vpxa.cfg, but everytime I reconnect it removes the two lines again.

<serverIp>ip address</serverIp>

<serverPort>902</serverPort>

-


hostd.log------

2010-04-13 14:54:49.327 'Proxysvc' 19573680 warning SSL Handshake on client connection failed: SSL Exception: error:14094412:SSL routines:SSL3_READ_BYTES:sslv3 alert bad certificate

2010-04-13 14:54:55.547 'Proxysvc' 19307440 warning SSL Handshake on client connection failed: SSL Exception: error:14094412:SSL routines:SSL3_READ_BYTES:sslv3 alert bad certificate

2010-04-13 14:54:57.668 'Vimsvc' 69319600 info : User root

2010-04-13 14:54:57.669 'ha-eventmgr' 69319600 info Event 38 : User root@127.0.0.1 logged in

2010-04-13 14:54:57.976 'VmMisc' 41147312 warning Failed to find conversion from winServer2008Datacenter-64 to type: vim.vm.GuestOsDescriptor.GuestOsIdentifier

2010-04-13 14:54:58.101 'ha-eventmgr' 3076440992 info Event 39 : User root logged out

2010-04-13 14:55:01.608 'Vimsvc' 19573680 info : User root

2010-04-13 14:55:01.609 'ha-eventmgr' 19573680 info Event 40 : User root@127.0.0.1 logged in

2010-04-13 14:55:01.942 'TaskManager' 18922416 info Task Created : haTask-ha-folder-root-vim.host.LocalAccountManager.createUser-50

2010-04-13 14:55:01.956 'TaskManager' 18922416 info Task Completed : haTask-ha-folder-root-vim.host.LocalAccountManager.createUser-50

2010-04-13 14:55:01.956 'Vmomi' 18922416 info Activation : Invoke done on vim.host.LocalAccountManager:ha-localacctmgr

2010-04-13 14:55:01.957 'Vmomi' 18922416 info Throw vim.fault.AlreadyExists

2010-04-13 14:55:01.957 'Vmomi' 18922416 info Result:

(vim.fault.AlreadyExists) {

dynamicType = <unset>,

name = "vpxuser",

msg = ""

}

2010-04-13 14:55:01.992 'TaskManager' 19839920 info Task Created : haTask-ha-folder-root-vim.host.LocalAccountManager.updateUser-51

2010-04-13 14:55:04.972 'TaskManager' 19839920 info Task Completed : haTask-ha-folder-root-vim.host.LocalAccountManager.updateUser-51

2010-04-13 14:55:04.976 'TaskManager' 19573680 info Task Created : haTask--vim.AuthorizationManager.setEntityPermissions-52

2010-04-13 14:55:04.977 'TaskManager' 19573680 info Task Completed : haTask--vim.AuthorizationManager.setEntityPermissions-52

2010-04-13 14:55:05.013 'ha-eventmgr' 69319600 info Event 41 : User root logged out

2010-04-13 14:55:05.058 'Vimsvc' 19573680 info : User vpxuser

2010-04-13 14:55:05.059 'ha-eventmgr' 19573680 info Event 42 : User vpxuser@127.0.0.1 logged in

2010-04-13 14:55:05.253 'ha-eventmgr' 19839920 info Event 43 : User vpxuser logged out

2010-04-13 14:55:05.370 'Vimsvc' 18656176 info : User vpxuser

2010-04-13 14:55:05.371 'ha-eventmgr' 18656176 info Event 44 : User vpxuser@127.0.0.1 logged in

2010-04-13 14:55:05.376 'ha-eventmgr' 18922416 info Event 45 : User vpxuser logged out

2010-04-13 14:55:05.394 'Vimsvc' 18922416 info : User vpxuser

2010-04-13 14:55:05.394 'ha-eventmgr' 18922416 info Event 46 : User vpxuser@127.0.0.1 logged in

2010-04-13 14:55:05.401 'ha-eventmgr' 18922416 info Event 47 : User vpxuser logged out

2010-04-13 14:55:05.411 'Vimsvc' 19573680 info : User vpxuser

2010-04-13 14:55:05.411 'ha-eventmgr' 19573680 info Event 48 : User vpxuser@127.0.0.1 logged in

2010-04-13 14:55:05.430 'Vmomi' 41147312 info Activation : Invoke done on vmodl.query.PropertyCollector:ha-property-collector

2010-04-13 14:55:05.430 'Vmomi' 41147312 info Throw vmodl.query.InvalidProperty

2010-04-13 14:55:05.430 'Vmomi' 41147312 info Result:

(vmodl.query.InvalidProperty) {

dynamicType = <unset>,

name = "licenses",

msg = ""

}

2010-04-13 14:55:07.578 'VmMisc' 19839920 warning Failed to find conversion from winServer2008Datacenter-64 to type: vim.vm.GuestOsDescriptor.GuestOsIdentifier

2010-04-13 14:55:08.672 'App' 19839920 error Failed to send response to the client: Broken pipe

Reply
0 Kudos
geddam
Expert
Expert

Please restart the vcenter server service.....

Thanks,,

Ramesh. Geddam,

VCP 3&4, MCTS(Hyper-V), SNIA SCP.

Please award points, if helpful

Thanks,, Ramesh. Geddam,
Reply
0 Kudos
kasperottesen
Contributor
Contributor

geddam I'v already tried that, it did not help.

Reply
0 Kudos
Dagnabbit
Contributor
Contributor

Kasper- did you get any resolution on this? I just had the same thing happen to me... running vCenter 4, and a standalone ESX host running ESX 3.5. Host is running fine, restarted mgmt-vmware and vmware-vpxa services. DNS name resolution is fine (DNS server is fine, host file as a backup is fine, checked DNS PTR records as well). About to put int a trouble ticket weith VMware... just not looking forweard to the wait, and then having to go through all the steps I just went through trroubleshooting on my own, and then never getting an answer back from tech support.

Reply
0 Kudos
GarethWilson
Contributor
Contributor

did you get a fix for this from vmware after you logged a support ticket ?

Reply
0 Kudos
janardhanr
Enthusiast
Enthusiast

Hi,

You also give a try by removing the host from VC and adding it back again, this will not trouble your VM's, You can also have a look at the following link which will help you to reinstall the vcenter agent.

http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=100371...

Regards,

Janardhan.

NOTE: If your problem or questions has been resolved / answered, please mark this thread as answered and award points accordingly.

Regards, Jana. NOTE: If your problem or questions has been resolved / answered, please mark this thread as answered and award points accordingly.
Reply
0 Kudos
Dagnabbit
Contributor
Contributor

I ended up shutting down all of the VM's, and upgrading the server to ESX 4.0 U2. I wasn't sure if the update would work- I have two other standalone ESX servers that I cannot upgrade. Hoping that this issue doesn't hit those other servers. Pulling the server out of the VC and putting it back in did not work, or removing and reinstalling the HA agent didn't work (in fact, I couldn't reinstall the HA manually or by adding back to VC). DNS is working fine, name resolution is working fine. The only thing I didn't try was making sure I was patched to 3.5 U5. I tried the update to 4.0, and it worked, so no need to try the patch to 3.5 U5.

Reply
0 Kudos
Noyzyboy
Contributor
Contributor

I had this same/similar problem on ESXi 4.1. VMware tech support fixed it with the following internal only procedure. Hope this helps someone.

1. Select the ESX in vCenter

2. Click on the "Configuration" tab and select "System Resource Allocation"

3. Under "System Resource Pools", select "host", then "vim", then "Edit Settings"

4. When the "Edit vim Resource Allocaation" window opens, set the "Memory Resources" Limit to "Unlimited"

Reply
0 Kudos
billywhizuk
Contributor
Contributor

Had the same error today on an ESX 3.5.0 build 82663 server, it disconnected from the Cluster it was in and wouldn't connect in VCentre

An internal error vmodl.fault.HostConnection.  It then went through the wizard of adding a host but failed at 9%

Rebooted ESX host and same fault when trying to connect in  VCentre.   I could connect directly to the ESX host itself and see the hosts (all down)

I had to reboot our server hosting VCentre and then connect worked straight away on this ESX host

VCentre was the issue for my fault

Reply
0 Kudos
MarianoMileti
Contributor
Contributor

I had the same issues, the solution? I Log on to the console (in my case throgh HP ilo - remote console) and them restart the management agents (in thoubleshooting options).

Any other option causes a down of the service.

4 years later, but, now you have the real answer Smiley Happy

Regards,

Reply
0 Kudos
EdmundTan
Contributor
Contributor

Bloody hell this actually works! Might be 4 years late for OP but certainly helped the others in future! Thanks alot! BTW im on Vcenter 5.5 with esxi host 5.5

Reply
0 Kudos
esxi1979
Expert
Expert

i have esxi 5.5 & VC 5.5 too, all of sudden all 5 nodes had same issue.

dns is fine

also tried the mgmt network restart from console, no luck

Reply
0 Kudos
rcmessner
Contributor
Contributor

I'm using esxi5.1.x and found the following solution to work after a couple previously mentioned solutions failed my case:

Log in directly to each host via vSphere despite them being technically managed by VC server. You'll get the popup that says it's managed (even though it's not showing up in vSphere via VC anymore...) so just click through that nonsense.

On the Summary tab, there's a section called Host Management in the bottom right. If your hosts are still thinking they're associated with the VC server, it should say "This host is currently managed by vCenter Server a.b.c.d and you'll have the Disassociate host from vCenter Server... option.

Disassociate each host one by one and when you go back in through vSphere via the main VC login, you should be able to successfully re-add each host without issue.

Hope this helps @esxi1979 and anyone else dealing with inexplicable vmware BS

Reply
0 Kudos
Filonowst
Contributor
Contributor

orked for me on ESXi 5.5. Thanks!

I restarted the mgmt agents using these two commands from https://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=10034...

/etc/init.d/hostd restart
/etc/init.d/vpxa restart

Reply
0 Kudos
tfourk
Contributor
Contributor

On mine I tried the restarting of the mgmt agents to no go. I noticed then on connecting it was trying to create "vpxuser" and the user already existed. So it gave an error and knocked me off. So I went to the host and deleted the user on the host. Reconnected the host to Vcenter and it worked like magic. ESXI 6u2 here

Hope this helps someone other there.

Reply
0 Kudos
Dw2rf
Contributor
Contributor

/etc/init.d/hostd restart

/etc/init.d/vpxa restart


Restart the agents from the cli not helped, even restart the management network from DCUI not helped.

Only help its from DCUI -> Troubleshooting Options -> Restart Management Agents

ESXi 5.5

Reply
0 Kudos
sojab0on
Contributor
Contributor

Same issue here ESXI 6.0 and ESXi 5.5 host

When i try to Convert an w7 x64 pro mashine physical i get error

FAILED: A general system error occurred:

InternalError at 98% and

FAILED: A general system error occurred:

InternalError when i try te reconfigure with the reconigure option.

can some one point out why.has 3,5gb ram installed and onley  1,5gb used at point wehre convert is running.

The mashine i try ti configure

kind regrads michel

Reply
0 Kudos