Converter standalone stalls at 1% due to username/password

[whositwhatnow]

Hello,

I'm attempting to do a hot P2V linux.  I get coonected, machine creates new vm, boots but when asking for helper password it fails attempting to login with helper login, and password saying not found.   Then get message vmacore application loop.

Can anyone assist?

Where do i enter helper login/password

I'm using standalone app on windows.

[ivivanov]

You should not need to enter you password in the helper. Converter Server should connect automatically to it. Do you have a valid IP address of the Converter Helper VM? You should be able to see it in the vCenter client when selecting the VM in the inventory tree on the left. Is there a DHCP server running in the network where the target VM is created?

[whositwhatnow]

No dhcp manual IP free IP set in helper config.  NO DHCP used in this

Error is waiting for config variable helper password, then after several attempts in contnues on and gives me Entering vmcore application loop.

[ivivanov]

Nothing wrong so far. So Converter server should connect to the helper and instruct it to start partitioning the target disks and transfer data. Do you actually get an error in converter GUI? Can you export a log bundle (switch to Task View, right click on the failed task and select "Export logs" command)?

[whositwhatnow]

O yeah i didnt mention im using the local standalone clinet on my windows 7 machine v5

[whositwhatnow]

Task editor shows:

Connecting to the Converter helper server on the destination virtual machine.

Waiting for the destination virtual machine to boot up as the Converter helper server.

Connecting the Converter helper ISO image to the destination virtual machine.

Powering on the destination virtual machine to run as the Converter helper server.

Its been running for over an hour stuck at this task, in the console in just show Entering vmacore application loop....

cannot get log unless i cancel task.

[vinodvasu]

Hi

We had the same issue and our Physical server was 2.5 TB in size with LVMs.we solved the issue using the latest VMware convertor 5.0 and rebooting the physical machine before the migration.

vino

[parixit]

Were you able to solve this?  We are stuck at the same spot.

[whositwhatnow]

Yes i finally got it to work, stumped even the VMware support guys but we figured it out. 

This worked for us

It stalled because it did not have all completed info from the vmware helper IP.  We do no use DHCP on our server subnet, and manually assign IP's.

So in the tool we gave it IP, subnet DNS etc. 

We had to set the table routing by doing these commands after logging in to the helper (using root password for current LInux box we are PtVing):

route add GATEWAY dev eth0

route add default gw GATEWAY

After i did that poof it proceeded and completly cloned machine.

[parixit]

Thanks for the reply.  Are you logging in via ssh while its still stuck in that vmacore loop?

[bdaeumling1]

can you please tell me the username which you are using to login ?

if i try "root" it says Login incorrect

[ERImaging]

You should not need a user and password.  I found that my problem was the helper VM did not have the correct VLAN in the properties.  Once I changed the Network connection for the Network adapter, it instantly started to convert the linux machine.

[Khawaja_Jawad]

@whositwhatnow    

Can you please further detail, where you added below routes (on source or on destination) , if destination how you added these routes...

route add GATEWAY dev eth0

route add default gw GATEWAY

[jfrigola]

I get the same problema. Please answer someone: where are those routes inserted?

[patanassov]

The route is in the helper VM, i.e. the destination. See this KB in order to log in: http://kb.vmware.com/selfservice/microsites/search.do?cmd=displayKC&docType=kc&docTypeID=DT_KB_1_1&e...

However the helper's start-up invokes "route add default gw GATEWAY". Since it has only one interface, the other command should not be necessary.

Please provide some feedback in case this workaround succeeds (and the usual way doesn't); I'd like to diagnose the issue and hopefully find a solution.

[Vishalthom]

Thanks Plamen,

This worked for me . I have followed steps given in

http://kb.vmware.com/selfservice/microsites/search.do?cmd=displayKC&docType=kc&docTypeID=DT_KB_1_1&e...

[waiganjo]

the IP helper should be an IP that's on the same subnet as your vcenter but not in use (not pingable)

[RussellArdo]

I think I have found the cause of this issue.

I have an ESXi server that has multiple vswitches so that I can dual home guests or provision which VLAN the guests land on.

When trying to P2V a linux server onto this host, I encounter this issue.

The P2V process from Linux includes creating a temp guest that boots with the live cloning tools to begin the conversion.

After which, the temp becomes the newly converted virtual guest.

What I've discovered is that at this stage, the temporary cloning OS ends up with the default vSwitch that may not be able to communicate back to the physical server.

You can see this by editing the guest whilst stuck at logon, and look at the network that gets allocated.

My issue now is that there is no 'Converter' option to specify the vswitch network that the cloning tool should use.

There is an option to specify the network for the new guest... once converted but this is not applied at this stage.

Essentially... there is a bug with the conversion process as it should if you an IP Helper config to select a vSwitch/Network.

So... I need to think of how to change the default vswitch.

I'm sure if this helps anyone or me right now but thought I'd share.

[RussellArdo]

OK... perhaps there are a number of things going on here.

I changed the adapter NIC for the new host to be E1000 and did not connect the second NIC on power up.

By checking the guest propertise, I compare the MAC address to these DHCP scope leases and could test by ping the IP that was release to it.

When it doesn't ping... it's not gonna clone.

The Physical linux server also has 2 NIS and originally I left them both in.

It also appeared to work after having two nics but leaving the second one set to Not connect on boot.

With only one... it works so I'm sure it's getting confused at this stage..

One thing to note... when it does work and connect to the network correctly... and the converter begins to progress... the guest console will continue to displace the prompt to login.

So make sure you check the progress of the converter when testing.

Hope this helps.

[anvointsev]

RussellArdo  thanks , your comments helped me to find solution

I also have two standart vSwitchs , one for all vmk inderfaces (10.10.10.0/24 for example) and second for VM network (10.10.11.0/24)

Host for migration and Converter were placed in 10.10.10.0/24  , but  helper vm can get interface only in VM network -  10.10.11.0/24

These networks are not crossed in my environment  and the converter was failed by timeout on the stage

Connecting to the Converter helper server on the destination virtual machine.

I temporary added VM network to vSwitch0 (10.10.10.0/24)  and the converting process finished successfully.

The document is described how the process  works

http://www.vmware.com/pdf/convsa_50_guide.pdf