Assignment of IP with vCAC external network profile

Similar Questions

• Problem with external network profile and DHCP

  Hello

  Is it possible to create a network without IP range profile, in order to use the external DHCP protocol.

  If I create a network without ip range profile I got this error when I submit an application VM:

  [catalog] Com.vmware.vcac.platform.rest.client.error.ResponseErrorHandler.handleError:55 - [error rest] ERROR: {status code: 400}, {error code: 42300}, {error Source: null}, {Msg of error: a server error occurred.} Error the applicant unit. {The list of unallocated IP addresses for the network profile (mynetworkprofile) has been exhausted.}, {System Msg: error of Infrastructure service provider}

  I'm sure we can do it, but I don't know how

  Max

  Simply create the profile of any network. Just assign the reservation with the network that you want to use, and the virtual machine will get his IP from the DHCP server.

  vCAC will not be able to manage these addresses because DHCP manages them--if that's what you want to accomplish...

• Problem of external network (Client)

  Hi guys,.

  After the transfer of my "production" environment, I am facing a problem with the external network (not API).

  On my lab environment it works fine, but on the production environment something strange happens: the machine can be put into service, but sometimes he gets 'customization' Openstack, sometimes not.

  Also, when it "works", I can reach a floating IP to one machine, and it works for something like 5 seconds, then this IP stops responding.

  I noticed that my switch TRANSPORT DVS Checkup complains to the VLAN ID 0 (which is not effective in these physical network cards), I use the VLAN ID 102 as the Vlan of Transport VXLAN. Is there something else I'm missing?

  My setup is composed with:

  -1 bunch of Mgmt

  -1 bunch of nova

  -1 bunch of edge.

  They are all on the same Transport area of the NSX, and connectivity to the Openstack API and the Mgmt API now works well I put almost present on 'production' (in fact it is an internal test with some users), but this external network problem is a show-stopper for me

  Thanks for the advice, sorry to bother you with so many questions

  Ah, an offtopic question: when you release the 2.0 version? Is it going to be compatible with the 1.0 version, or I have to remove and reinstall my whole environment? (VMworld, probably?;) (P)

  Thanks again!

  Reading your post, I understand that you have 2 numbers:

  Problem with the metadata service (give the host name):

  Communication to the metadata service works like this:

  

  Can you do some tests for me.

  1 test VXLAN connectivity between the ESXi-Compute (hosting your VM of Instance OpenStack) and the edge of ESXi (hosting your logical shared edge router)

  Ping ++ netstack = vxlan x.x.x.x (to test the connectivity of your IP)

  Ping ++ netstack = vxlan d s 1570 x.x.x.x (to test the MTU on the physical fabric is 1600 or greater)

  Note: x.x.x.x is the remote IP of VXLAN ESXi @.

  If this test does not work, then your VM can speak to its default gateway correctly.

  If this test does NOT work, then it is probably a problem with the physical fabric (VLAN misconfigured or MTU misconfigured).

  2 validate connectivity VLAN between the edge of ESXi (hosting your edge of metadata) and ESXi-Mgt (hosting your LoadBalancer VIO)

  . SSH to the VIO - WHO

  . then make a SSH for balancing load-VIO ssh loadbalancer01

  . can do a ping to Metadata Edge1 ping x.x.x.x (ping should fail, BUT you should see an ARP entry for that IP @: arp - na)

  Note: You can find the Edge Mgt IP metadata @ under edition of the edge metadata_proxy_router_xxx NSX and its uplink interface checking.

  If this test does not work, your edge of metadata can talk to VIO load balancing.

  If this test does NOT work, then it's probably a problem with the Mgt ESXi and ESXi points not to the same port of Mgt group.

  Problem with the external network:

  You validate people can talk to the VM floating IP @?

  If this isn't the case, it is probably because the edge of ESXi (hosting your logical shared edge router) is not on the right VLAN.

  Dimitri

• Several external networks on a single vswitch

  I am trying to understand (among others) if I need more a vmkernel on the same vswitch. It is related to a question I posted last week, but I understood some things since. Here is the configuration (slightly reduced for reasons of this discussion).

  2 vSphere 5.5 hosts, each with:

  1 vmnic connected to external switch capable of trunk ports (vSwitch0); It is currently the management network

  1 vmnic connected to the switch for vmotion (vSwitch1)

  1 vmnic connected via switch private iscsi array; the private switch VLANs separated for iscsi (vSwitch2)

  VMotion and iscsi works very well, so I am concerned mainly with the external networking. Currently, all virtual machines are on VLAN not signposted. We will change to 2 VLAN tagged on different subnets - 10.1.10.x and 10.2.20.x. The VMs will have to talk to each other so that on different hosts. So, does that mean that I need 2 vmkernels on vSwitch0 - one for the 10.1 subnet and one for subnet 10.2? Then the Group at a port by vmkernel and matched vlan id for group of ports and vmkernel?

  OR

  I really only need a single vmkernel on vSwitch0 with 2 groups of different ports for the different VLANS? IE, VMs with different networks than network mgmt will be able to communicate through the external switch to virtual machines on other host?

  Also, I expect that external ports must be set to allow 1 marked and tagged 10 labeled 20 - is that correct?

  So, does that mean that I need 2 vmkernels on vSwitch0 - one for the 10.1 subnet and one for subnet 10.2? Then the Group at a port by vmkernel and matched vlan id for group of ports and vmkernel?

  No.... you need not multiple VMkernel can use different virtual machine networks in your ESXi host.

  I really only need a single vmkernel on vSwitch0 with 2 groups of different ports for the different VLANS? IE, VMs with different networks than network mgmt will be able to communicate through the external switch to virtual machines on other host?

  Yes, just use the existing VMkernel management interface and create two groups of ports, one for each VIRTUAL local area network. Yes, VMs of different VLAN in network management will be able to communicate, BUT your physical switch must be configured to allow traffic from multiple VLANs.

  Also, I expect that external ports must be set to allow 1 marked and tagged 10 labeled 20 - is that correct?

  It should work.

• vCAC 6 network IP range profile exhausted problem

  Hello

  can someone confirm this please:

  create a profile to external network... to create a range of pool of 5 IP addresses for example.

  create a plan that uses the reserve network,

  ensure that the reserve of network has 1 only free to use address (deploy 4 virtual machines to consume IP addresses)

  now deploy 3 vms from the action plan.

  you will get this error:

  Failure of the application after the filing of machines 1. Error the applicant unit. The list of the unassigned IP addresses for the network net-Lab-961 profile has been exhausted.

  Well, here I found something strange on my environment:

  (1) the application process is in the failed State

  (2) no VM is present in the title of the points

  It's ok, but

  (3) the virtual 1 st(of 3) machine is present in vCenter inventory

  (4) the virtual 1 st(of 3) machine has the ip address allocated within the ranges of IP addresses of the network profile

  There is a way to delete the entry on the IP ranges from network associated with the 1st VM profile?

  a new data collection does not solve the problem

  Thank you

  Alan

  If you go to Infrastructure > Machines > managed Machines that see you the VM there?

  Grant

• Unable to assign the IP of network profiles.

  Hello

  I want to activate the user select network based on the network available, chosen in PG when you set a reservation. I use XML to list outside these exchanges but facing problems with it. I use vCAC 5.2 with vSphere 5.5

  The model is configured with a network adapter, keeping in mind that what ever network selected by the user would add to the provisioned VM.

  1. I am able to get the Port Group reflected in vcenter under the network section, but the setting of the virtual machine does not added adapter list. I could see Mac address associated to the Portgroup and allocated IP network profile, but I do not see these values in the VM.

  2 - I custom configured VM from vCAC and added additional NIC with different portgroup who also has associated the different subnet network profile. This time I could see port group reflected in vCenter under network and also the adapter is also added to the VM, but still could not see IP address other APIPA address then. However a single IP address again responsibility network profiles that does not matter where in the VM.

  have I missed something? is there an agent that must be installed on a virtual computer to enable this feature (just a wild guess! )

  Help, please!

  BR,

  MG

  Yes this actually solved the problem, but it's still strange that we need to care about customization in vCenter and reference should be placed in the strategic plan, how construction profile Important in this scenario?

  However, custom, property of "CloneSpec" is not required to mention under properties if you use the field 'customization Specs"of Blueprint.

  This requires no installation of the Agent. It is agentless.

  Thank you

  BR,

  MG

• Error with external / network API

  Guys,

  I'm doing some tests on my lab environment, but not over and over again.

  My deployment environment have 2 groups (mgmt and nova). The two have only 1 uplink, and the uplink is attached to a DVS with 2 groups of ports: management and external. Two of these PGs are related to VLAN 10.

  The problem is that, even the deployment is going well, I can not access the API networks (external load balancing, for example).

  I see LB machines have an eth0 (who works on the network in-house / balancing) and the eth1 with access 'external '. I am able to ping, the viomgr IPs on eth0 and the physical IP on eth1, but not the VIP of the eth1 (shown with ip addr command).

  I don't know if this problem is related to something on DVS, NSX or any other thing. I just installed and reinstalled my environment using any means of different configuration and it does not work.

  So I tried a stop of the load balancer and but this IP address on the machine eth0 and the API has worked well.

  No idea why this problem with networking occurs?

  Also, I tryied to set up NSX using frames, as well as MTU 1600.

  TKS in tips.

  Hi Ricardo,

  One of the VIO requirements is that the direction and external port groups should use different network addresses. From what you described, it seems that you are using the same subnet for two groups of ports. If you are not able to transfer using different subnets, use these commands to work around the problem (run on the two load balancers):

  sysctl-w net.ipv4.conf.eth0.rp_filter = 0

  sysctl-w net.ipv4.conf.eth1.rp_filter = 0

  It is not persistent if balancing charge are restarted. Note, that it is not a supported configuration and has specific security implications.

  Hope that helps.

  Best regards

  Karol

• Get the assignment of the IP to an external network

  Hello

  as mentioned above, I want the query wo assignment of IP to a given external network organization.

  As a goal, I want to know, what external IP number (of a certain external network) is attributed to each organization.

  Is it possible to do this via the REST API?

  Thanks in advance!

  It was something, that I recently looked into after that a colleague has asked about it.

  The eaiest IMHO is to use the vCloud API Query Service (http://blogs.vmware.com/vsphere/2012/04/quickly-finding-objects-using-the-vcloud-api-query-service.html)

  Here is a screenshot of an external network called "Ext-network-1" for a VCD 5.1 deployment

  

  Step 1 - make a GET to the following URL: https://[VCD - IP] / api / query? type = externalNetwork

  Step 2 - this should return a list of NetworkRecords that will be your external networks available, here is a screenshot using FF RestClient

  

  Step 3 - Once you have identified the property "name" of your external network, you will need to do a GET on it's HREF

  Step 4 - from here you can see the current range of intellectual property as well as the allocated IP addresses

  

  If you want more details on IP addresses allocated, you will see in the 'link' there is /allocatedAddresses you can do a GET on this URL, which will give you more details

  

  I hope this allows you to get what you are looking for

• Can I set the external network with no gateway

  Hi is possible to create 'External network' without bridge in VCD 1.5?

  No, but you can simply give a false gateway (as some random number)

• There is no network profile from the drop-down list

  Vcac 6.0.1

  Hello.  I created a network profile and assigned an ip address range.  I am now trying to choose the profile to network within the reserve, and he does not appear.  I select the network in the network tab and then click the menu drop down and there is nothing there.  What I'm missing here?

  Thank you

  Jamey

  What type of network profile you created? I think that if it's something other than external, he will not come unless you have configured VXLAN.

• Machines Linux taking two IPs of network profiles, but enrolling in vCENTER.

  Hi all

  I have 2 problems here with vcac network profiles.

  When I use profiles network with reservations, I'm having windows machines configured right but the machines Linux booking 2 IPs in the network profiles, but registering with the profile of a network. I checked to see if the Linux machines have been set up to have two vNIC, but no. However, I tried to assign network profiles at reserves using the custom property 'VirtualMachine.Network0.ProfileName' and this way, I'm having the same problem as described. But this time for windows and LINUX machines.

  Secondly, I have network profiles configured with mask, gateway, DNS server address, and Suffixes. When my machine is using network profiles, inherit from the shape of the network profile DNS suffix or I have to use the 'VirtualMachine.Admin.NameCompletion' property so that my machine to register in the DNS server.

  vCAC version: 5.2

  Custom spec: using

  Static: Yes

  Appreciate the HELP in advance.

  Nits

  Yes, the Linux machine will inherit the config of NP in the CustomSpec. In addition, if you install the GuestAgent (instead of using the CustomSpec) which will also set the relevant parameters. To do this, the action plan requires that the custom property VirtualMachine.Admin.UseGuestAgent be set to True.

• How can I change IP assignemnts in network profiles in the vRA

  A virtual machine has been provided through a VLAN defined in a network profile. An IP address has been allocated on the virtual computer network profile. All good but VM has then been changed manually to get a new IP and VLAN assigned by replacing the original AVP defined IP and VLAN. Anyone know how I can remove the old IP of the vRA network profile?

  From now on, that we develop a workflow vRO to allow users to perform work to change the custom properties of VM to then change the IP VM in the profile.

  Thank you

  Prestonville

  This isn't official supported, but you can do it in the database of the vCAC (here, through vCO):

  var date   = new Date();
  var query  = "UPDATE [dbo].[StaticIPv4Address]";
  query += " SET [Hostname] = ''";
  query += " ,[VirtualMachineID] = NULL";
  query += " ,[StaticIPv4AddressState] = 1";
  query += " ,[LastModifiedDate] = '" + System.formatDate( date,"yyyy-MM-dd HH:mm:ss" ) + "'";
  query += " WHERE [IPv4Address] = '" + IpAddress + "'";
  query += " AND [StaticIPv4NetworkProfileID] LIKE '" + networkProfile + "'";
  var resultRecords = database.executeCustomQuery( query );
  

  Kind regards

  Markus

• Change the subnet on the network profile

  Hello

  Like a fool I created 3 network profiles and paths using 16 subnets when they should have been 24. I wonder now to open firewall rules to route to other VLANs in 24 of the subnets. As its 16 and the two first parts are identical i.e. 10.2.X.X they don't route through the firewall

  The current VMS I can just go to manually change the subnet assigned to the NIC and they work.

  When I look at a network profile to change the subnet - it is greyed out

  Is there a way I can change this manually? so that the new virtual machines created with / 24 subnet and exisiting VMs I can manually change

  Or is the only way to delete the network profile and start over. How this affect the VMs running other than bulk can try and assign an IP address already in use?

  Thank you!

  You can try to update the database

  Find the name of the profile

  SELECT * from StaticIPv4NetworkProfile

  the name of the profile update

  Update dbo. StaticIPv4NetworkProfile

  Set SubnetMaskIPv4 = "255.255.255.0" where StaticIPv4NetworkProfileName = 'YOUR NETWORK PROFILE'

  Open the database in SSMS (SQL Server Management Studio) and run the SQL query when it is connected to your database of vRA. He has not supported but will most likely solve your problem.

  Or if you are not happy to do this directly through the database, you might try using the vRO vRA plugin and call the vCACEntityManager and update the StaticIPv4NetworkProfiles entity. My guess is the update of the database direct would be easier, but please take a backup and only update the field and nothing else.

• Create the first machine with vCac 6.0.1

  Hi guys, I have a lot of confusion with vcac and vco.

  I have installed and configured the identity, vcac e device device a windows server for vcac iaas.

  I have configured the tenant in the vcac reservation... and vsphere vcenter agent after, I create a server model of win2008r2 to clone.

  I create a profile network with the ip address range and a master plan for commissioning the clone of model.

  but the problem is the clone never change the hostname, ip address and never join domain.

  I want to know if all this 'simple' work I do with only vcac or I have to use the vCO

  How can I configure machine clone with a different ip address, hostname and joined in the field by the tenant?

  Thank you

  You will need to build a specification on vCenter and attach it to the EHR Architecture on vCAC. This will customize the OS when commissioning occurs.

  Here's a screenshot for your reference. Locate the field marked in black.

  

• Why must id VLAN unique for each external network?

  Sometimes we have several public networks with the same VLAN, but in this case, we could not add them as external network to vCloud Director, he complains that the VLAN ID has been used, then assign a VLAN ID for each public network. Could someone explain to me why a unique id of VLAN is needed here?

  Thank you!

  If they are the same vLAN on the same Switch, they are actually exactly the same from the point of view of the vCloud Director.

  The alternative is to see if the superposition of external networks allows it works (Admin > General).