I am attempting to deploy a vApp for a customer into their catalog and getting a bit of an odd error.
Has anyone seen this one before?
Operation: Processed descriptor of Virtual Application Template.
Error: Validation failed for the OVF file you provided: Invalid argument '', should be empty.
Invalid argument '', should be empty.
Invalid argument '', should be empty.
Invalid argument '', should be empty.
Invalid argument '', should be empty..
Thanks in advance.
We have just upgraded our vmware vcloud environment: vCloud Director 1.5.1, vCenter 5.1, vShield 5.1 and ESXi 5.1 hosts
The issue happened after restarting the vClould Director Appliance when web UI stopped working.
I went through the vCD log files and found the following error in cloud-container-debug.log:
2013-03-21 11:56:18,920 | ERROR | Start Level Event Dispatcher | JDBCExceptionReporter | The Network Adapter could not establish the connection |
2013-03-21 11:56:19,078 | INFO | Start Level Event Dispatcher | StartupUtils | Successfully bound network port: 80 on host address: 172.16.1.152 |
2013-03-21 11:56:19,079 | INFO | Start Level Event Dispatcher | StartupUtils | Successfully bound network port: 443 on host address: 172.16.1.152 |
VMware vCloud Director 5.1.0.810718, Central European Time, 172.16.1.152 (172.16.1.152)
2013-03-21 11:56:20,244 | DEBUG | Bound HTTP Request Log Files | JettyStarter | Found log file: 2013_03_14.request.log (Size = 2,113,827 bytes} |
2013-03-21 11:56:20,244 | DEBUG | Bound HTTP Request Log Files | JettyStarter | Found log file: 2013_03_17.request.log (Size = 0 bytes} |
2013-03-21 11:56:20,244 | DEBUG | Bound HTTP Request Log Files | JettyStarter | Found log file: 2013_03_20.request.log (Size = 4,943,406 bytes} |
2013-03-21 11:56:20,244 | DEBUG | Bound HTTP Request Log Files | JettyStarter | Found log file: 2013_03_18.request.log (Size = 0 bytes} |
2013-03-21 11:56:20,244 | DEBUG | Bound HTTP Request Log Files | JettyStarter | Found log file: 2013_03_16.request.log (Size = 0 bytes} |
2013-03-21 11:56:20,244 | DEBUG | Bound HTTP Request Log Files | JettyStarter | Found log file: 2013_03_15.request.log (Size = 0 bytes} |
2013-03-21 11:56:20,244 | DEBUG | Bound HTTP Request Log Files | JettyStarter | Found log file: 2013_03_19.request.log (Size = 0 bytes} |
2013-03-21 11:56:20,245 | DEBUG | Bound HTTP Request Log Files | JettyStarter | Request log processing complete. Total log file size: 7,057,233 bytes; Deleted 0 request log files |
2013-03-21 11:56:20,285 | DEBUG | Start Level Event Dispatcher | ConnectionPool | Creating interceptor instance of class:class com.vmware.vcloud.common.jdbcpool.interceptor.ConnectionState |
2013-03-21 11:56:20,286 | DEBUG | Start Level Event Dispatcher | PooledConnection | Unable to connect to database. |
java.sql.SQLException: The Network Adapter could not establish the connection
It seems that vCD is not able to connect to the embedded Oracle database, but I am able to connect to this same db using sqlplus.
-bash-3.2$ sqlplus "/ as sysdba"
SQL*Plus: Release 11.2.0.2.0 Production on Thu Mar 21 13:04:55 2013
Copyright (coffee) 1982, 2011, Oracle. All rights reserved.
Connected to:
Oracle Database 11g Express Edition Release 11.2.0.2.0 - 64bit Production
Also NICs on the vCD VM are up and have the correct IP settings.
I made some reboots but the problem is still present
Please let me know if anyone has faced this issue.
Thanks in advance
Hello all,
I'm a bit of a newbie with vCloud Director so maybe this is dumb question.
The environment is vCloud Director 1.5.
We had 7 VMs were in an Organization vDC called, for arguments sake, vDC_Original which was part of the Provider vDC, PvDC_Original.
We needed to get them onto another Orgainization vDC on another Provider vDC. (i.e. vDC_New and PvDC_New)
As we were unclear on how to migrate them from one to the other, they were powered off and reimported from vCenter into vDC_New. This caused issues because the guest OS computer name changed but that was easily corrected.
The issue now is that we have a vApp in vDC_New with the running VMs and another vApp in the vDC_Original with VMs with statuses like "Inconsistent State" and "Powered off".
I'm wanting to tidy up the environment and remove the vApp on vDC_Original but when I look at the properties of the VMs in the vApp I see that the "Name in vSphere" entry is the same (e.g. MyVM1 (bc7c38b2-9854-4fb2-aec9-6b68e6cb5e51) for both VMs i.e. the VM in the vDC_New and the VM in the vDC_Original.
I'm somewhat concerned that if I delete the VMs from the vApp in vDC_Original it will delete the "live" VMs in vDC_New.
How can I remove the vApp from vDC_Original without affecting the live VMs in vDC_New?
Any guidance will be most welcome!
TIA.
can you run it through OVF tool to validate the structure? alternatively, you can extract the OVF (it's a tar archive of files) and look at the configuration files for some random null entry.
the vCloud Application connects to the Oracle DB as the user vcloud.
So if you login as sysdba, you'd have to connect as vcloud. however, you don't have the password since it's random and encrypted.
What you might be able to do as sysdba is see if vcloud account is locked or if the password expired.
As a side note/reminder, the appliance is only for Proof of concept and not production purposes. there is no migration method to go from the Appliance to a full instance either, so just be aware of that.
I've seen asymmetric routing cause a problem once and a while. something similar to this: https://access.redhat.com/knowledge/solutions/53031/.
Also, note that the appliance is only for Proof of Concept. If you intend to roll into production, this means you need to install a full version on a complete OS with an External Database.
Vcloud director 5.1 with win2k3 sysprep correctly installed. When I deploy a new vapp with a 2k3 VM, If I ask to customize the guest OS it works fine: it change the admin password, the computer name etc. But if I ask to change SID, when the VM is powerd on I get the error: The Windows Sysprep deployment package is missing files for certain versions of Windows. Contact your cloud administrator. The following VMs cannot be started: "2k3'".
vmb01 wrote:
Vcloud director 5.1 with win2k3 sysprep correctly installed. When I deploy a new vapp with a 2k3 VM, If I ask to customize the guest OS it works fine: it change the admin password, the computer name etc. But if I ask to change SID, when the VM is powerd on I get the error: The Windows Sysprep deployment package is missing files for certain versions of Windows. Contact your cloud administrator. The following VMs cannot be started: "2k3'".
With versions of Windows older than 2008/Vista, you have to manually copy the sysprep binaries (from deploy.cab found on the Windows Server 2003 CD) to the vCenter server. The location to which you need to copy the files is as follows:
2003 64 bit - c:\ProgramData\VMware\VMware VirtualCenter\sysprep\svr2003-64
2003 32 bit - c:\ProgramData\VMware\VMware VirtualCenter\sysprep\svr2003
Once you do that, you should be able to sysprep during customization of Windows 2003 deployments from template.
Hi There,
I have an Organization Admin account to my organization virtual data center. I want to to change the DHCP configuration for the network. When I log in to vCD, select my organization network, right click to configure the service, then I find that Configure Service option is disabled?!
Also, if I try to create a new netwrok, the wizard allows me to create an internal network only, but the option to create a routed network is disabled.
Can you please tell me what is wrong? what is missing?
Thanks much!
you need to copy the files into one cell, then run the command to compile them into a sysprep package.
All of the steps you need are in the vCloud Director guides, and you can download sysprep as required from Microsoft.
1. On what type of network, you are trying to configure DHCP. For Fenced network - DHCP will be diabled.
2. Ho are you trying to create routed network in wizard? If you have one org vdc network and it is already added to vApp, then option for org vdc network will be disabled. you can create vApp Network and then in connection drop down you can select org vdc net to make routed vapp network.
HTH
When i try to upload to catalog, director re-directs me to Oracle for java. I've downloaded version 6 and version 7 of JRE and setup environmnet variables on my windows 7 box.
What gives, what have i missed?
Attached are screenshots to see what i'm talking about.
Hi,
I am using Vcloud Director 5.1 appliance. When i create an Org VDC Network(isolated), a vshield edge vm gets deployed and later disabling the DHCP services on this Org VDC Networks, the vshield edge vm is deleted. Is there any way to disable DHCP services when Org VDC Network is created, instead of waiting till the edge vm is deployed and later disable it.
Thanks & Regards,
Divya
Thanks, I did the sysprep installation as explained kb.vmware.com/kb/1026313 but at the end I did not restarted the vCloud Director cell
I've correctly installed the VRMC plugin on my I.E. 8 browser. When I try to open the console of a VM from the vcloud director 5.1 web gui, the browser opens a new windows , I see the VRMC console in the CONNECTING state but after a while it goes in DISCONNECTED state and I do not see the VM console.
it's always one little thing that causes a hiccup. glad I could help.
vmb01,
So we are able to produce the console screen, which means the VMRC application is starting in the background.
Then we basically try to connect to the Console Proxy address as specified in the Public Addresses section of vCloud Director (System > Administration).
If we cannot resolve the hostname or communicate with the IP Address it would resolve to, then that's just a general failure to communicate.
If you are using 1 cell, try putting the IP address in the public addresses section instead of a hostname and ensure your desktop can access the IP on port 443. if IP works, then it's name resolution from your system.
If you are using more than one cell and or a load balancer, can you bypass the load balancer and point to the console proxy IP of only one cell? if we bypass and it works, it's either the load balancer causing problems or how the certificates were setup.
I don't believe there is a way to have a default set of services enabled/diabled on an Org vDC network. Ideally, as long as your network isolation is functioning, there isn't much harm in this. If the feature is annoying, there is a way to script around it.
How many networks are you making where you need DHCP disabled?
Greg,
Glad to see some Lab Manager people around here 
The workflow with Lab Manager is surprisingly similar, but the details were hidden.
In Lab Manager, you would 'unpublish' a template. This process created a new child link in the chain to store changes. This is eqivilent to copying a vApp from Catalog to 'My Cloud' for editing purposes.
Once you did the update in Lab Manager you could then publish. The difference is with vCloud, that you have to make 1 more additional link in the chain to put it back into the Catalog.
However, the performance degredation as chain length increases has significantly reduced. If there is VAAI support for linked clone chains on your storage, e.g. Native Linked Clones, then it's even better from what I hear.
If you run this process and delete the original from Catalog, nothing will be damaged. The pre-existing VMs based on the "previous" linked clone chain will still exist. vCenter manages these linked clone chains now, as Shadow VMs.
Lab Manager just did all the micro-management in its Database and Datastores, as undeployed objects were not listed in vCenter invetory.
As always, if you want to consolidate VMs in the Catalog of vCloud ... that'll reset chain length and start a new Linked Clone tree for any VMs created from that vApp/VM afterwards.
I hope that helps.
Jon Hemming
In vCloud Director 5.1.x you need to make an Edge Gateway first. This is your "router" from the isolated Org Network to the External. If a Gateway does not exist, then it's not possible to make a Routed Organization network.
In vCloud 1.5.1 this was all done in one step.
