vRealize Automation DEM worker cannot connect to Orchestrator

In vRA 6.2, using vRO 6.0, you may find that the data collection and other vRO workflows fail with the error “You must have at least one properly configured vCenter Orchestrator endpoint that is reachable”.  The IaaS/Monitoring/Log will show which DEM worker threw the error.  When you check the DEM worker logs for that instance, if you find the message “Could not create SSL/TLS secure channel. —> System.Net.WebException: The request was aborted: Could not create SSL/TLS secure channel“, you have probably been affected by VMKB 2123455 and MS KB 3061588.

Although both articles seem to suggest that removing the offending patch will solve the problem, I think figuring out exactly which patch is rather awkward.  The easier fix is to apply a quick registry hack to your DEM workers (and wherever the vRA Designer runs).

  1. Logon as an account with admin rights (suggest the account your IaaS services run under)
  2. locate or add the key

    HKLM\SYSTEM\CurrentControlSet\Control\SecurityProviders\SCHANNEL\
    KeyExchangeAlgorithms\Diffie-Hellman

  3. Add/update the DWORD value ClientMinKeyBitLength and set the value to 512 decimal (200 hex)
  4. Restart the DEM worker service

 

Notes:
The Microsoft patch sets the default minimum group size to 1024.  It appears that the vRO 6.0.x appliances use something less than that.  This registry hack indicates that SCHANNEL should accept keys as small as 512 bits.  I suggest only applying this to the necessary and affected machines since it does lower the bar for the DHE security requirements.

Thanks to Zach Milleson for reminding me that this workaround may not resolve everyone’s issue, depending on which MS patches are installed.  If this workaround doesn’t work for you, you may have to locate and remove the offending patch.  YMMV.

About Cross-vCenter NSX

I’ve spent the past several weeks testing, trying to understand Cross-vCenter NSX and make it work in a useful way.  Here’s what I’ve learned so far.

Environment and Set up:

I knew I needed two “sites” with discrete storage, but had a few physical limitations.  I.e; I only own one managed router (Cisco C2821) and one managed switch (Force 10 S50).  I trunked the management and vtep network to all the hosts, but configured a discrete transit network for each “site”.  I hope to work with a much larger lab environment and do a thorough review of the setup and configuration.  In the meantime, here’s the basics.

  • Two-host cluster, NSX manager and vCenter Server for primary site
  • 1 host cluster, NSX manager and vCenter Server for recovery site
  • vSphere 6.0 U1b
  • NSX for vSphere 6.2.1
  • Universal Transport Zone includes both clusters, configured for Local Egress.
  • VTEPs in same L3 network (for simplicity’s sake)
  • Edge per site, discrete transit networks
  • Universal DLR for tenant networks, uplinked to both site ESGs
  • OSPF area for each ESG->UDLR
Partial Lab Network
Partial Lab Network

Objectives

  1. Eliminate need for manual synchronization between sites/NSX instances
    • Success! The distributed firewall and universal objects are respected by the NSX manager at both sites.  Universal Security Groups and Logical Switches are usable at either end.
  2. Span VXLANs between sites
    • Success! This is not really a surprise.  As long as the clusters share a Transport Zone and teh VTEPs can route to each other, this works great.
  3. Minimize network alterations when used with Site Recovery Manager to failover protected VMs between vCenter Servers.
    • Partial success.  A placeholder/recovered VM will be in the same Universal Security Groups as the protected VM and the distributed firewall rules will apply as expected without any changes needed.  However, even with Local Egress enabled, you’ll have to apply some sort of route updates so that traffic destined for the recovered VMs can get there.  It looks like the route redistribution for egress us handled automatically though.  This configuration pushed the limits of what I can do with my lab.
  4. Make use of site-to-site microsegmentation (my definition:  application of security policy regardless of L3 network scope)
    • Partial Success.  I was really disappointed here. When the document states that you can add IP Sets, MAC Sets and other Universal Security Groups to a Universal Security Group, it means that’s ALL you can add.  I’ve blogged a few times about adding VMs to a Security Group – even doing it as a Resource Action in vRA.  That won’t work with Universal Security Groups!  The only workaround I see is to add the desired VM’s IP or MAC to a Set and include the Set in the Security group.  Blech!
  5. Retain ability to assign security policy via security group membership through vRA
    • FAILURE.  As noted above, we cannot assign Universal security group membership to a VM using any of the existing custom properties.

Notes and Observations

I guess I get it.  The vm id between vCenter Servers will differ, so saying that “vm-123 is a member of securitygroup-456” is not valid on a different vCenter Server.  But a table of IPs and MACs would be universally valid.  I’m hoping that in a near-future version, the universal security group membership capabilities can be extended; perhaps with a shared or replicated vCenter Inventory Service.

 

Fix – Unable to import vCAC/vRA certificates into Orchestrator

Problem:

While in the vRealize Orchestrator Client you find that the Library/Configuration/SSL Trust Manager/”Import a certificate from URL” workflow returns an error reading “InternalError: handshake alert: unrecognized_name” when provided. The URL the resolves to the Load-Balancer VIP for the vCAC/vRA appliances.

 

Background:

Signed SSL certificate installed on vCAC/vRA Appliance, SSL Passthrough on NSX/vCNX Load-Balancer, vCAC/vRA Settings/Hostname set to resolve to VIP, matching SSL cert.

 

Fix:

  1. SSH into the vCAC Appliance as root
  2. Backup /etc/apache2/vhosts.d/vcac.conf to vcac.conf.bak
  3. Use vi to edit /etc/apache2/vhosts.d/vcac.conf
  4. Scroll down to  <virtualHost _default_:443>
  5. Add these lines

    ServerName fqdn.of.appliance.node

    ServerAlias: load.balancer.name

  6. Scroll further to ensure these params aren’t listed elsewhere, remove or revise if so.
  7. save the file and exit vi
  8. restart the vCAC/vRA services

Automating NSX Security Groups with vCAC/vRA – Part 2

***UPDATE***  The download link is currently broken.   I seem to have lost the file, will fix the link as soon as I find or recreate it.  Sorry about that. 😦

In part 1 of this series, we created a list of security groups and displayed that list to users during the request.  In this post, we want to enhance that functionality by adding these features.

  • Creation of Security Groups and inclusion in Dropdown lists
  • Add a VM to a Security Group post-provisioning
  • Import existing Security Groups into vRA inventory
  • Add a Security Group to a Dropdown list

We’re going to do that by importing a vCO package with some new workflows and actions, then link up the workflows to Advanced Services and Resource Actions.  As before, we’ll require the NSX and vCAC/vRA plugins for Orchestrator.

Preparation

  • Complete the creation of the VCNS.SecurityGroup.Names.production property dictionary and valuelist attribute from Part 1.  We’re going to reuse those items. so make a note of the exact name of the property dictionary and the valuelist attribute.  In my case, I’ve named the Property Dictionary VCNS.SecurityGroup.Names.production and also named the valuelist attribute for it VCNS.SecurityGroup.Names.production
  • Make sure vRA Advanced Services Server Configuration is complete and test the connection to the Orchestrator server.  The default, built-in VCO is fine.
  • Login to vCO client as a vCO Admin. Set the mode to “Design” and navigate to the Inventory tab.  Make sure that you have a connection listed under “vCAC Infrastructure Administration” and a connection listed under “NSX”.

 

Confirm that you have the necessary connections on the inventory tab
Confirm that you have the necessary connections on the inventory tab

Get the Package

I’ve put together a handful of workflows and actions that use or expand the NSX plugin to provide information of functionallity back to vRA.  Where possible, I reused existing library workflows, but in some cases, I had to use the API to create a REST call and consume that in an action.

By downloading any code, package or file, you acknowledge that:

There is no explicit or implied warranty or support for the code.  Neither Brian Ragazzi, his employer nor anyone else is responsible for any problems, errors, omissions, unexpected behavior, breakage, trauma, outage, fatigue, lost time, lost work or incontinence that may occur as a result of using the code or package.

Download the zip file.  It contains the package and a couple of images that can be used for the advanced services

Import the Package

  1. Extract the zip file
  2. In the vCO Client, navigate to the packages tab.
  3. Click the “import package” button and select the extracted .package file
  4. On the Package Import Information step, click “Import

    Package Import Information
    Package Import Information
  5. On the Import package… step, check the “Select/Deselect all” box to check all of the items.  Please note the server path, these should not be duplicates of anything else you have in your vCO inventory (unless you’ve already imported this package previously).  Click “Import Selected elements”.

    Select all items
    Select all items
  6. Review the workflows and actions added to your inventory.

Configure Advanced Services – Create Security Group

This service enables the user to create a new NSX Security Group and automatically adds its name to the appropriate dropdown list of security groups.  It can be added once for each different list of security groups.  You’ll need to know the exact name of the Property Dictionary and valuelist attribute you created in Part 1.

NSX Management Services
NSX Management Services
NSX Security Groups in vRA Items
NSX Security Groups in vRA Items
  1. While logged into vRA as a service architect, navigate to the Advanced Services tab, click “Custom Resources
  2. We need to make vRA aware of NSX Security Groups.  Click the Add button.
  3. In the Orchestrator Type field, enter “NSX:SecurityGroup“; for the Name, I suggest “NSX Security Group“, click Next

    Add NSX Security Group as Custom Resource
    Add NSX Security Group as Custom Resource
  4. On the details form, we’re not going to make any changes, but if you wanted to hide certain properties, you could here.  Click “Add“.
  5. Click “Service Blueprints”, then the “Add” button.
  6. On the Workflow tab, select the AddNewSecurityGrouptoDropdown workflow, click next.

    Select "AddNewSecurityGrouptoDropdown" workflow
    Select “AddNewSecurityGrouptoDropdown” workflow
  7. On the Details tab, set the name to something like “Create new Production NSX Security Group“, because we’re going to create the security group and add its name to the “production”dropdown list.  Click Next.

    Set the Service Item Name
    Set the Service Item Name
  8. On the Blueprint  Form tab, under the “Step” Form page (default), mouseover the text field labelled “Name of Custom Property Dictionary in vCAC/vRA”.  Click the pencil “edit” icon when it appears.

    Edit the Form Fields
    Edit the Form Fields
  9. Click the Constraints tab of the “Edit Form Field” window.  On the Value field, select “Constant” and enter “VCNS.SecurityGroup.Names.Production” (or whatever suffix you used) for the Property Dictionary.  Set the Visible value to “No” so it doesn’t show up. Click Submit on the Edit Form Field window.

    Set the name of the Property Dictionary to be updated
    Set the name of the Property Dictionary to be updated
  10. Using the same method, set the Name of the Attribute appropriately and its visibility to no
  11. Edit the “Value to be appended to the ValueList attribute” field.  Set the label to “New Security Group Name“.  Do not set a value or make this one invisible, we need the user to enter a value, submit to save.
  12. Edit the vCACIaaSHost field – using the Constraints tab again– when setting the value, choose constant, then click Add by the green plus, to display a treeview, where you can choose your connection to the IaaS Server.  Visible: No, submit to save.

    Select connection to IaaS host
    Select connection to IaaS host
  13. Edit the “NSX endpoint” field in the same way, selecting the NSX connection.
  14. When done, all fields except “New Security Group Name” will have a value.  Click Next.
  15. On the Provisioned Resource tab, select “securityGroup [NSX Security Group]“.  Click Add to save the service blueprint.
  16. Repeat steps 5-15 for any other dropdown lists containing security groups; say “Non-Production” for instance
  17. Highlight the Service Blueprint and click “Publish” to make the blueprint available for entitlements
  18. Navigate to Administration, Services.
  19. Add a new Service named “NSX Management” (for example) – I included a nifty image in the zip file
  20. Under Catalog Items, click the “Create new Production NSX Security Group” item to edit it.
  21. The Catalog item should inherit the Security Group icon from vCO, set its Service to “NSX Management”,click update to save.
  22. Create or Edit an entitlement to include the new Service and/or catalog item.
  23. Try it out, confirm that the Security Group was created in NSX, is visible in vCAC items and it name was added to the Property Dictionary

Configure Advanced Services – Import Security Group

This service allows you to make existing security groups visible as items in the vCAC Items view.  Once this is done, we’ll add actions that allow you to add the security group to a dropdown list.

  1. Click “Service Blueprints”, then the “Add” button.
  2. Select the “GetNSXSecurityGroup” workflow, click Next
  3. On the details tab, set the name to “Import NSX Security Group“, click Next
  4. On the Blueprint form, set the “connection” to the NSX connection in vCO, then hide the field.  Security Group Name will be a dropdown list of existing NSX Security Groups for the user to choose from. Click Next
  5. On the Provisioned Resource tab, select “securityGroup [NSX Security Group]“.  Click Add to save the service blueprint.
  6. Just as before, publish the service blueprint, add it to a service and an entitlement.

 Configure Advanced Services – Add Security Group to Dropdown list

With this service, we’ll let the user add the name of an existing Security Group to a drop down list.  Unlike the first two, this is implemented as a Resource Action, meaning it’ll be executed against an existing item (a Security Group in this case)

  1. Under Resource Actions, click “Add”
  2. For the Workflow, select the “AddExistingSecurityGrouptoDropdown“, click Next
  3. On the “Input Resource” tab, keep NSX Security Group, click Next
  4. On the Details tab, set the Name to “Add Security Group to Production list” or similar, set the description, leave the Type options unchecked.  click Next

    Set Action Name and Description
    Set Action Name and Description
  5. On the Form tab, just like the first service blueprint, set the Property Dictionary and Attribute names as appropriate.  VCNS.SecurityGroup.Names.production in my example, set visible to no on both.
  6. Again, we’ll set the vCACIaaSHost to the connection to the Server and hide the field
  7. Click Add to save the action.
  8. Repeat steps 1-6 for each security group dropdown list (say “non-production” for instance)
  9. Publish the action and add it to an entitlement

    Add Action to Entitlement
    Add Action to Entitlement
  10. Test by navigating toNSX under Items, highlight a Security group and Select “Add Security Group to…” from the Actions menu.

    Yay! A Resource Action
    Yay! A Resource Action

 Configure Advanced Services – Add VM to a Security Group

This service lets you add a provisioned VM to additional Security Groups.  So, at provisioning-time, the VM is added to the Security Group selected by the user, but we may need to refine the security by adding that VM to additional Security Groups.

  1. Under Resource Actions, click “Add
  2. For the Workflow, select the “AddVMtoSecurityGroup“, click Next
  3. On the “Input Resource” tab, keep IaaS VC VirtualMachine, click Next
  4. On the Details tab, set the name to “Add VM to a Security Group“, click Next
  5. On the Form tab, set the connection Value to the NSX connection.
  6. Leave the NSX Security Group field visible, click Add to save the action
  7. Publish the action and add it to an entitlement
  8. Test by selecting a machine under Items and “Add VM to a Security Group” from the Actions menu

    VM Resource Action for Security Groups
    VM Resource Action for Security Groups
  9. You’ll be presented with the list of allNSX Security Groups to which you can add the selected VM

    Select Security Group
    Select Security Group

Conclusion

This part of the series should help streamline the management of VMs and their membership in Security Groups.  Obviously, items like removing a VM from a Security Group or even removing a Security Group are not included here.  The NSX plugin is missing quite a bit of functionality available in the API, so those additional functions require significantly more configuration.

Thanks to John Dias for his information and examples posted here.

 

Automating NSX Security Groups with vCAC/vRA – Part 1

In this series, I’ll document how to automate the creation and (some of) the management of NSX security groups within NSX.

First, what’s the use case?  Why is this interesting?  Let’s assume that you’ve decided to use large “flat” networks instead of many small networks.  One reason you may make that decision is because of the challenges with either having many blueprints (one per network!) or making changes to the workflows to reliably set the appropriate properties.

Background

In this solution, we’ll have to have vCAC 6.1 or vRealize Automation 6.2, NSX 6.x and vCenter/vRealize Orchestrator with the vCAC and NSX plugins installed and configured. We have two Logical Switches, one for Production and one for Non-Production.  In addition, there’s a corresponding network profile and the business groups have reservations.  Now, we have to ensure that there  are security boundaries within the flat networks.  We’ll accomplish this through Security Groups.

Caveats

We’ll create security groups and nod in the direction of security profiles, but will not be automating the creation of security profiles nor their assignment to the Security Group(s).  That can be done by the security admins through the NSX interface or maybe later we’ll add that capability too. 😉

 Procedure

  1. Create Security Groups.
    • Open vSphere Web Client and navigate to Networking and Security, then Service Composer.
    • Click the “New Security Group” icon
    • Enter a Name and Description for your new Security Group and click Next
    • If you want to create rules for dynamic membership or include/exclude existing VMs, you can do so in the subsequent steps.  Finish the wizard.
    • Repeat to create all of your security groups
    • Create Security Groups in NSX
      Create Security Groups in NSX
  2. Create Property Dictionaries invCAC/vRA.
    • Log into vCAC as an Infrastructure Admin and navigate to Infrastructure|Blueprints|Property Dictionary
    • Click “New Property Definition”, for the name enter “VCNS.SecurityGroup.Names.Production“.  You can replace “Production” with a name of your choosing, so you can have multiple lists.
    • Select “DropDownList” as the control type and check to make it required, click the green check to save.

      Create Property Dictionary
      Create Property Dictionary
    • Click the “Edit” link in the Property Attributes column
    • Click “New Property Attribute”, select “ValueList” as the attribute type
    • Set the name to something appropriate, such as the same name as the Property Definition or “ValueList” or “SecurityGroups”
    • In the Value field, enter the names of the security groups you want included.  Separate the group names by commas (no spaces).  If you have groups whose names include spaces or commas, put them in quotes.  Click the green check to save.
    • Repeat to create another property dictionary and attribute for the Non-Production list
  3. Update Blueprints.
    • Edit your “production” blueprints by adding the “VCNS.SecurityGroup.Names.Production” custom property. Set the value to your default security group or leave it blank to require a selection. Be sure to check the “Prompt User” box. Click the green check to save.

      Add Custom Property to Blueprint
      Add Custom Property to Blueprint
  4. Test
    • Submit a request for the affected blueprint and verify that the dropdown list of security groups looks like you expect it to. Remember, that unlike many other custom properties in vCAC (eg: Network Profiles), you CAN have multiple versions of this one and display different lists.

      Dropdown list of Security Groups
      Dropdown list of Security Groups
    • After a VMis provisioned, verify in the vSphere Web Client that ithas been assigned to the expected security group

      VM added to Security Group
      VM added to Security Group

Next

In the next parts of this series, I plan to address the problems of maintaining the dropdown list manually and having a single security group per machine.

Many thanks to my friend Grant Orchard for his article on selecting a security group in a blueprint . It was the inspiration for this series.

vBrownBag – Custom Properties in vRealize Automation

I was honored with the opportunity to present a discussion on how you’d use custom properties in vCAC/vRA to accomplish three use cases;

  • Select a standard VM size
  • Select a Network Profile (same list for everyone)
  • Select a Network Profile (different lists per blueprint)

Quick and Dirty workaround for vCAC/vRA MSDTC issues

Please note that this is a workaround, not a fix. I couldn’t get the damn thing working reliably, so I bypassed authentication. I wish I knew why I got “DataBaseStatsService: ignoring exception: Error executing query usp_SeclectAgent Inner Exception: Error Executing query usp_SelectAgentCapablities” all the frigging time. Followed John’s advice and reinstalled MSDTC on e.v.e.r.y.t.h.i.n.g, no change. Made sure that the settings match VMKB2089583, no change.

As a workaround, I just selected “No Authentication required” on the Clustered MSDTC and the IaaS Manager Service services. Don’t do this as a long-term thing.

This works, but don't do it in production
This works, but don’t do it in production

Connecting vRealize Automation Gugent in a Distributed deployment – some notes

Configuring a distributed vCAC 6.x or vRealize Automation 6.2 installation, ran into a few issues and thought I’d document them.

  • The Gugent points to the vCAC IaaS Manager Service, not the vCAC Web/Model Manager
  • Test by browsing to https://[vcacmanager]/VMPS
  • If you are using vCNS or NSX as your load balancer, use these settings for the virtual address (these worked for me):
    • Application Profile: TCP, no persistence
    • Pool: IP-HASH algorithm, Monitor: NONE.  One enabled member, port 443, not transparent
    • Virtual Server: TCP 443, Pool and App Profile from above
  • Confirm the Gugent is working properly before configuring a template for Application Services

The Value of EMC Enterprise Hybrid Cloud

EMC Enterprise Hybrid CloudFull disclosure: I work for EMC and am a lead architect in the EMC Enterprise Hybrid Cloud “SWAT” team.

What is it?

Much like VCE Vblock converged infrastructure systems, EHC is an engineered solution.  This means that its components have been tested and integrated to work together and it is supported as a unit.

The Value

I compare the solution to Vblock frequently, because in the early days of VCE/Acadia, many people did not understand the value of converged infrastructure.  It took a while for many folks to understand how it saves time, frustration and money over building a solution in-house from selected components.  Nowadays, most enterprise IT shops understand that converged infrastructure brings a level of integrated support that is not available when you support individual components of the solution.

EMC Enterprise Hybrid Cloud: Federation SDDC Edition (just “EHC” for this discussion) brings many VMware and EMC components together with integration and support.  Sure, the front-end of EHC – what you’ve seen pictures of – is VMware vRealize Automation.  What you may not see in that picture is the integration of EMC ViPR that allows selected admins to provision a new datastore to hosts in a vSphere cluster from the same portal.  You may not see in that picture the tight Backup-as-a-Service integration with IaaS; while requesting a new VM, the requestor can select a backup policy for the new VM.  It is automatically added to a backup job and the machine owner can choose to perform a backup or restore on demand – from the same portal.

Another important feature of EHC that cannot be “seen” is the vast engineering time that has been put into ensuring the components and their versions work together flawlessly.  You cannot see the countless hours spent testing and retesting the workflows to ensure they behave as expected.

It is this time that the EMC EHC team has invested in the solution, so that customers can rely on the solution for Enterprise IT operations and not have to worry about version x.xx of a component working correctly with version y.yy of another.

Lastly, the EHC solution is ready to use very quickly.  This means that customer’s time-to-value is very short and they can begin realizing the benefits almost immediately.

The Future

The EHC Federation SDDC edition will continue to evolve, using newer versions of components only once they’ve been fully vetted.  I predict more integration with network services, a solution integrated with RecoverPoint, VPLEX and Site Recovery Manager.

 Conclusion

Jeez, this does sound like a marketing pitch, sorry.  I really do thing this is a great solution and a great direction from EMC.  I won’t give away any of the secrets in the solution, but you can bet that the fix to some of the bumps-in-the-road we hit will be blogged about here.

Configuring Highly Available vCenter SSO with SSL certificates

*** UPDATE 12/18/14 ***
Instead of this blog, I strongly suggest you use and follow the
Configuring VMware vCenter SSO High Availability for VMware vRealize Automation Technical Whitepaper. It is somewhat more comprehensive and authoritative. For VMware documentation, it’s really good.
*** UPDATE ***

I love the simplicity of the vCenter Server Appliance and the VMware Identity Appliance for vCAC, but neither offer a high availability option better than vSphere HA. There are use cases where you’d need your SSO service to offer better uptime and resilience. In addition, there is some SSL certificates to be configured and for that, we’ll follow the instructions in KB2034833,  KB2061934 and KB2034181.

Notes, caveats, warnings
AFAIK, this will only work with vSphere 5.5. v5.1 handles SSO differently. I’m only using two nodes, if you have more, there will be extra steps. I do not have intermediate CAs, if you do, consult the KBs for the additional steps. I’m going to use a vCloud Networking and Security Edge Gateway as my load-balancer.  It does not offer SSL offload like some other load-balancers do, so you may have to take extra steps to configure SSL offload.

Here's what I have in mind; load-balanced SSO Servers
Here’s what I have in mind; load-balanced SSO Servers

Prerequisites
Reserve the IP addresses for your actual SSO servers, plus the Virtual IP address.
Add A or CNAME records to your DNS for the SSO servers and the virtual IP.
The DNS name of the virtual IP is what the SSL certificate must match (vcsso in my case)
You should have an edge gateway already configured with an interface in the same networks as your virtual IP and actual SSO servers.

First SSO Server

I’m starting with two freshly deployed Windows Server 2008 R2 VMs, joined to the domain and named vcsso1 and vcsso2.  On vcsso1, install the Single Sign-on service.  Be sure the prerequisites are all ok.

vCenter SSO Prereqs

On the deployment mode step, choose “vCenter Single Sign-On for your first vCenter Server” First SSO Server

Next,next,finish your way through the installation. You’ve set up an SSO server, YAY!

Second SSO Server

On the second server, vcsso, also install the SSO service.  We’re going to make a few different selections than we did on vcsso1 though.  On the deployment mode step, here we’re going to select “vCenter Single Sign-On for an additional vCenter Server in an existing site”.Second SSO

 

Next, we’re prompted for information about the first, or partner, SSO server.  SSO Partner

 

We have to confirm that the information obtained from the first SSO server is correct, so click Continue.Certificate Verification

Then we select the site name configured on the first SSO server.  I named mine “Lab”, but you can leave yours as “Default-First-Site” or whatever makes sense for your environment. Select Site Name

 

From here, you’ll Next,Next,Finish your way to completion.

Generating the Cert
Prerequisites: Either the VMware ssl-certificate-updater-tool or OpenSSL Win32 v0.9.8

  1. Log on to the first SSO server (vcsso1), extract the VMware SSL certificate updater tool to C:\ssltool or similar.  Create folders named “C:\certs\sso“.  Open notepad and paste the following:

    [ req ]
    default_bits = 2048
    default_keyfile = rui.key
    distinguished_name = req_distinguished_name
    encrypt_key = no
    prompt = no
    string_mask = nombstr
    req_extensions = v3_req


    [ v3_req ]
    basicConstraints = CA:FALSE
    keyUsage = digitalSignature, keyEncipherment, dataEncipherment
    extendedKeyUsage = serverAuth, clientAuth
    subjectAltName = DNS:ServerShortName, IP:ServerIPAddress, DNS:server.domain.com, DNS:ServerIPAddress

    [ req_distinguished_name ]
    countryName = Country
    stateOrProvinceName = State
    localityName = City
    0.organizationName = Company Name
    organizationalUnitName = vCenterSSO
    commonName = server.domain.com
  2. Replace the values in red with those appropriate for your environment. Be sure to specify the server name and IP address as the Virtual IP and its associated DNS record. Save the file as c:\certs\sso\sso.cfg
  3. At a command prompt, navigate to the folder containing openssl.exe (eg: C:\ssltool\tools\openssl). Run this command to create the key and certificate site request (CSR):

    openssl req -new -nodes -out c:\certs\sso\rui.csr -keyout c:\certs\sso\rui-orig.key -config c:\certs\sso\sso.cfg

    Generate CSR
    Generate CSR
  4. In the same command prompt, run this to change the key to the necessary type.


    openssl rsa -in c:\certs\sso\rui-orig.key -out c:\certs\sso\rui.key

  5. Follow the steps in KB2062108 to create the appropriate certificate template in you Active Directory Certificate Authority.
  6. Log into the Microsoft CA certificate authority Web interface. By default, it is http://servername/CertSrv/.
  7. Click the Request a certificate link.Click advanced certificate request.
  8. Click the Submit a certificate request by using a base-64-encoded CMC or PKCS #10 file, or submit a renewal request by using a base-64-encoded PKCS #7 file link.
  9. Open the certificate request (rui.csr) in notepad. Copy the content between —–BEGIN CERTIFICATE REQUEST—– and —–END CERTIFICATE REQUEST—–
  10. Paste the copied content into the “Base-64-encoded certificate request” textarea. Select VMware Certificate as the Certificate Template. See KB2062108 if you don’t have the “VMware Certificate” template
  11. Click Submit to submit the request.
  12. Click Base 64 encoded on the Certificate issued screen. Click the Download Certificate Chain link.
  13. Save the package as C:\certs\certnew.p7b.
  14. Double-click the p7b to open it in certmgr.  Navigate to Certificates – Current User\C:\Certs\Certnew.p7b\Certificates.

    Certs in P7b
    Certs in P7b
  15. You’ll see two certificates here (unless you have intermediate certificates, then you’ll have more).
  16.  Right-click the one for the SSO server, choose All Tasks|Export.  Save the file as Base-64 encoded X.509 (.CER) to c:\certs\sso\rui.crt
  17. Right-click the one for root CA server, choose All Tasks|Export.  Save the file as Base-64 encoded X.509 (.CER) to c:\certs\root.cer ..  Close certmgr.
  18. Generate the pfx by running this command:

    openssl pkcs12 -export -in c:\certs\sso\rui.crt -inkey c:\certs\sso\rui.key -certfile c:\certs\Root64.cer -name "ssoserver" -passout pass:changeme -out c:\certs\sso\ssoserver.p12

    Note: The certificate store password must be changeme and the key alias must be ssoserver. Do not change these parameters.

Install and Configure the Certificate

  1. While logged on to the first SSO server (vcsso1) as an administrator, make sure this folder exists: C:\Program Files\Common Files\VMware vCenter Server – Java Components If it doesn’t, you’ll need to check your SSO installation
  2. Open an elevated command prompt (as administrator) and enter the following

    SET JAVA_HOME=C:\Program Files\Common Files\VMware vCenter Server - Java Components
    SET PATH=%PATH%;C:\Program Files\VMware\Infrastructure\VMware\CIS\vmware-sso;%JAVA_HOME%\bin

  3. In the command prompt, cd to the folder containing openssl.exe (eg: C:\ssltool\tools\openssl)
  4. Generate a subject hash from the certificate using this command:

    openssl x509 -subject_hash -noout -in c:\certs\root.cer

    This will return an 8-character hash. Record it, we’ll need it later

  5. On both SSO servers, create the folder C:\ProgramData\VMware\SSL
  6. On both SSO servers, copy c:\certs\root.cer to C:\ProgramData\VMware\SSL renaming it to ca_certificates.crt
  7. On both SSO servers, copy c:\certs\root.cer to C:\ProgramData\VMware\SSL again, this time renaming it to <subjecthash>.0 (replacing <subjecthash> with your hash value from above and appending dot zero)
  8. Just on the first SSO server, paste the following into a text file named c:\certs\gc.properties. Replace the red text with appropriate values.

    [service]
    friendlyName=The group check interface of the SSO server
    version=1.5
    ownerId=
    productId=product:sso
    type=urn:sso:groupcheck
    description=The group check interface of the SSO server


    [endpoint0]
    uri=https://SSOserver.domain.com:7444/sso-adminserver/sdk/vsphere.local
    ssl=c:\certs\Root64.cer
    protocol=vmomi

  9. Paste the following into a text file named c:\certs\admin.properties. Replace the red text with appropriate values.

    [service]
    friendlyName=The administrative interface of the SSO server
    version=1.5
    ownerId=
    productId=product:sso
    type=urn:sso:admin
    description=The administrative interface of the SSO server


    [endpoint0]
    uri=https://SSOserver.domain.com:7444/sso-adminserver/sdk/vsphere.local
    ssl=c:\certs\Root64.cer
    protocol=vmomi

  10. Paste the following into a text file named c:\certs\sts.properties. Replace the red text with appropriate values.

    [service]
    friendlyName=STS for Single Sign On
    version=1.5
    ownerId=
    productId=product:sso
    type=urn:sso:sts
    description=The Security Token Service of the Single Sign On server.


    [endpoint0]
    uri=https://SSOserver.domain.com:7444/sts/STSService/vsphere.local
    ssl=c:\certs\Root64.cer
    protocol=wsTrust

  11. Next, we need the service ID for each of the three services SSO uses. To get these, run the following command, replacing the red text with the FQDN to your first SSO server:

    ssolscli.cmd listServices https://vcsso1.domain.local:7444/lookupservice/sdk

    SSO Services
    SSO Services
  12. The service ID for each service should be saved to a file. Use quickedit to copy the service id for each and echo it to a file:

    Echo service ID to files
    Echo service ID to files
  13. Update the Group Check service:

    ssolscli updateService -d https://ssoserver.domain.com:7444/lookupservice/sdk -u SSO_administrator -p password -si c:\certs\gc_id -ip c:\certs\gc.properties

  14. Update the Admin service:

    ssolscli updateService -d https://ssoserver.domain.com:7444/lookupservice/sdk -u SSO_administrator -p password -si c:\certs\admin_id -ip c:\certs\admin.properties

  15. Update the STS service:

    ssolscli updateService -d https://ssoserver.domain.com:7444/lookupservice/sdk -u SSO_administrator -p password -si c:\certs\sts_id -ip c:\certs\sts.properties

  16. Copy the new SSL files to C:\ProgramData\VMware\CIS\runtime\VMwareSTS\conf on both/all SSO servers:


    copy C:\certs\SSO\ssoserver.p12 C:\ProgramData\VMware\CIS\runtime\VMwareSTS\conf\ssoserver.p12
    copy C:\certs\Root.cer C:\ProgramData\VMware\CIS\runtime\VMwareSTS\conf\ssoserver.crt
    copy C:\certs\SSO\rui.key C:\ProgramData\VMware\CIS\runtime\VMwareSTS\conf\ssoserver.key

  17. Stop and restart the VMware Secure Token Service on both servers

Preparing to load-balance

  1. Navigate to http://firstssoserver.domain.local:7444/sts/STSService/vsphere.local notice that the certificate gives an error, but look at the cert. The certificate should return the “common” name (in my case, “vcsso” instead of “vcsso1”. Repeat this for the second and subsequent SSO servers, verifying that they provide the same certificate
  2. Using vCNS Manager, locate the appropriate edge gateway, click Actions|Manage to open it for editing
  3. On the Configure Tab, edit the interface that will listen on the virtual IP
  4. Edit the Subnet and add the Virtual IP. It’s probably not the primary IP. Save and publish those changes

    Add the virtual IP to the Edge Gateay
    Add the virtual IP to the Edge Gateay
  5. On the Load Balancer tab, on the Pools page, click “Enable”, then “Publish Changes”Enable Load Balancer
  6. Click the green plus to add a load-balancing pool
  7. Enter a recognizable Name and Description, click “Next”LB2-AddPool1
  8. On the Services step, check HTTPS, set Balancing Method to “ROUND_ROBIN” and the Port to 7444. Clck “Next”LB2-AddPool2
  9. On the Health Check step, set it as shown. Click “Next” when done.LB2-AddPool3-HealthCheck
  10. On the members step, click the green plus to add the IP address of you SSO servers to the pool. I suggest keeping the weifght for each at 1, unless you have a reason to send more requests to specific nodes. Keep the HTTPS port and Monitor Port at 7444 for each. Click “Next” once all you members are added.LB2-AddPool4-Members
  11. Review the Ready to complete step and click “Finish” if it all correct
  12. Click the Publish Changes Button before proceeding
  13. Click the “Virtual Servers” link, then the green plus to add a Virtual Server
  14. Enter a meaningful name and description, provide the Virtual IP adddress that you added to the edge earlier, select the Pool created in the steps above and Enable HTTPS on port 7444. Set the Persistence Method to SSL_SESSION_ID and make the “Enabled” box is checked. Click “Add” then “Publish Changes”LB2-AddVirtualServer
  15. Test by navigating to https://ssovirtual.domain.local:7444/lookupservice/sdk and https://ssovirtual.domain.local:7444/sts/STSService/vpshere.local verifying that the certificates match.
  16. YAY, load-balanced SSO with matching SSL certs!

One more thing….

Using your favorite web browser, navigate to http://ssovirtual.domain.local:7444/websso/SAML2/Metadata/vsphere.local you’ll be prompted to download and save an XML file named vsphere.download. Now open the XML file in notepad or Notepad++.  First, make sure you received a readable XML file.  Second, noticed that the EntitiesDescriptor/EntityDescriptor entityID property is server-specific.  We’ll need both servers to respond with the same information.

<EntitiesDescriptor xmlns="urn:oasis:names:tc:SAML:2.0:metadata" xmlns:saml="urn:oasis:names:tc:SAML:2.0:assertion" xmlns:vmes="http://vmware.com/schemas/attr-names/2012/04/Extensions" Name="vsphere.local" validUntil="2014-08-12T23:54:04Z">
<Extensions>
<vmes:ExportedOn>2014-08-11T23:54:04Z</vmes:ExportedOn>
<vmes:ExportedBy>Exported by VMware Identity Server (c) 2012</vmes:ExportedBy>
</Extensions>
<EntityDescriptor entityID="https://VCSSO1.domain.local:7444/websso/SAML2/Metadata/vsphere.local">
<IDPSSODescriptor WantAuthnRequestsSigned="false" protocolSupportEnumeration="urn:oasis:names:tc:SAML:2.0:protocol">
<KeyDescriptor xmlns:ds="http://www.w3.org/2000/09/xmldsig#" use="signing">
<ds:KeyInfo>
<snip...>

Warning This is not in a VMware KB, and may not be best way to do it. Having the value in the EntitiesDescriptor/EntityDescriptor entityID property match the FQDN is going to be very important in the near future. Trust me.

  1. On each server, open C:\ProgramData\VMware\CIS\cfg\vmware-sso\hostname.txt. It only contains the resolved hostname, so update it to the virtual hostname (vcsso.ragaazzi.lab in my case) save the file
  2. Retrieve the XML file from http://ssovirtual.domain.local:7444/websso/SAML2/Metadata/vsphere.local again open it and confirm that it contains the virtual hostname

Conclusion
This was such a lengthy post, I considered breaking it up, but there was no good break-point. Thanks for sticking with it. This is mostly for my own benefit, hopefully you’ll find it helpful too.