Collecting Java Heap dump from vCloud Director Cells

You just need to generate the java heap dump from one of the cells.  What you’ll need to succeed:

  • JCONSOLE
  • IP tables disabled on the cell you are connecting to.
  • Disk space available on the cell to accommodate the dump – I believe these can be between 8 and 10 GB in size
  • Unless an emergency, do this operation outside of normal business hours as it will be CPU intensive for up to 3 minutes, can impact API call performance, and can potentially cause the VCD cell inventory service to hang.

Step #1: Disable iptables on the cell

  • ssh to the desired cell and run the following command:

# service iptables stop

Step #2: Connect with jconsole (java console)

  • domain credentials should work here depending on your environment
  • connect to port: 8999
  • connect to desired cell

vcd9

  • If you get this message “Secure connection failed. Retry Insecurely?” just click the ‘insecure’ button to continue

Step #3: Generate the heap dump

  1. On the MBeans tab, in the com.sun.management/HotSpotDiagnostics object, select the Operation section.
  2. In dumpHeap parameters, enter the following information:

    p0: [heap-output-path]

    p1: true – do a garbage collection before dump heap

    For example:

    p0: /opt/vmware/vcloud-director/xx01-1-vcdc1_heap-dump-file.hprof

    p1: true

  3. Click the dumpHeap button.

image2017-5-31_15-19-56.png

  • There will be no indication that the heapdump completes.  I just watch the size of the file until the growth stops on the cell.  This process typically takes less than two minutes.

Step #4: Cleanup and send-off

  • Locate the heap dump in /opt/vmware/vcloud-director/ and move off to a location where you can compress and upload to VMware FTP site as you would for logs.
  • Start the iptables on the cell: # service iptables start

Upgrading VMware vCloud Director to 8.20

This document was creating while upgrading an existing vCloud Director 8.10.1 environment with an Oracle database, and multiple cloud cells.

After downloading the latest version of vCloud Director 8.20 for service providers,  SCP the upgrade to all VCD cells.  You can review the release notes here.

What you’ll need to do before getting started:

  • SSH into each cell and ‘sudo su -‘ to root
  • move the bin to the root directory
  • chmod +x vmware-vcloud-director-distribution-8.20.0-5515092.bin
  • I strongly advise opening an support request with VMWare before proceeding with the upgrade.  You may not need it, but it comes in handy having one logged beforehand.

Maintenance – Shutdown the cells

1. Open an SSH session into each VCD cell

 

2. Sudo to root using the following command:

# sudo su -

3. Change to the vcloud-director/bin/  directory

# cd /opt/vmware/vcloud-director/bin/

4. Use the Cell Management Tool to quiesce the cell.  This will move active jobs over to another cell.

# ./cell-management-tool -u administrator cell --quiesce true

5. Get the status of any running jobs on each cell.   ** Verify Job count = 0   |  Is Active = false  | In Maintenance Mode  = false

# ./cell-management-tool -u administrator cell --status
Example Output:

vcd6

6. Shut the cell down to prevent any other jobs from becoming active on the cell.  This command will also allow active jobs to cleanly finish

# ./cell-management-tool -u administrator cell --shutdown

Example Output:

vcd7

7. Get a status on the cells to be sure everything is down

# service vmware-vcd status

8. Now complete steps 4 – 7 on the remaining cells to cleanly shutdown the vCD service on all cells.

9. Here is where I would shutdown the VCD cell virtual machines, and database to get a clean snapshot while the environment is powered off

10. Once the database virtual machine is fully up, power-on the VCD cell virtual machines.

11. Log back into the vCloud Director environment to verify functionality before the upgrade.

12. SSH to all VCD cell virtual machines and use the following command to stop the service again on each cell.  Here there is an assumption made that we are now well within a maintenance window.

# service vmware-vcd stop

Starting The vCloud Director Upgrade

1. Start with the first cell, and run the first half of the upgrade.  DO NOT upgrade the database yet.

# ./vmware-vcloud-director-distribution-8.20.0-5515092.bin

Example Output:

vCD1

2. Respond with: y

Example Output:

vcd2

3. Stop.  Now you need to run steps one and two on the rest of the vCloud Director Cells, and install the upgrade.  Do them one at a time.  DO NOT upgrade the database yet.

4. Now that all cells have been upgraded, go back to the first cell and run the database upgrade.

# ./opt/vmware/vcloud-director/bin/upgrade

Example vCD Database upgrade output:

vcd3

5. Respond with: y

vcd4

6. Start the the first cell by responding with ‘y’

vcd8

7. Manually start the VCD service on the remaining cells

# service vmware-vcd start

8. Get the VCD status of all cells by running the following command on each

# service vmware-vcd status

9. Log into the cell, and watch/wait for vCenter to sync with vCD under the Manage & Monitor section → vCenters.  This normally takes 30 minutes or so.  Once done the status will change from a spinning circle to a green check mark.

10. Run some environment validation tests to be sure everything is working and is proper, and then delete those snapshots taken earlier.

 

Closing Out VMWORLD 2017

vmware’s annual party was held at the T-Mobile arena in Las Vegas this year.

Great entertainment, however the convention ran out of food pretty quick at several kiosks.  This year’s entertainment was a band from New Jersey called Bleachers who put on a decent show.

The main headliner for the night was Blink-182.  Certainly not the band I remember growing up, but they hit a lot of the songs from my youth.

On Thursday, the last day of the conference, vmware invited several guests on stage to discuss the future of science and technology.

The first speaker discussed the advancement of facial recognition, and the desire to track human emotions.  The ability to have artificial intelligence (A.i) understand empathy, and the ability to deliver digital content based on the emotional preferences of the viewer.  Although this session was cool, I personally feel that this technology is dangerous to pursue.

  • Delivering digital content to the viewer based on their viewing habits feels as if it would lead us to an era where people become emotionally numb.  If people are not shown content that exposes them to the full spectrum of emotions, it feels as if society would become even more disconnected from the world around them.
  • Giving A.I the ability to show empathy, and more huminized emotions may not be a bad thing if the proper requlations are in place.  I think as we become more escustomed to having our robotic companions, this would certainly make thss frustrating to use.

The second speaker discussed the advancement in prosthetics.  This was probably the most exciting keynote I’ve watched.

  • The speaker discussed his double leg amputation in his early twenties due to frostbite while mountain climbing.  He discussed his research at MIT where not only did he design his own prosthetics, but began a journey with colleagues that will ultimately change the future of prosthetics.  They identified a problem with the current generation and the lack of true innovation.
  • The speaker also discussed the crude method surgeons still currently use that prevents the rewiring of the human body, to feel the prosthetic, and to be able to flex and use it as if it were a real limb.  He then went on to show a new surgical procedure that allows the human body to do just that, and it was truly inspiring.

The final speaker of the keynote was a pediatric surgeon discussing the advancement of surgical “batting practice” where the teams practicing surgical procedures on patients is now finally becoming a thing of the past, and instead surgeons have teamed up with Hollywood special affects organizations.

  • This partnership has allowed the creation of life like surgical dummies that allows surgeons to practice in preparation for the real thing.  This allows surgeons to tackle complex cases, theories procedures, and then order these surgical dummies to suit these theories, and the patients specific medical condition.
  • They can then perform the multiple procedures, attend after action meetings to discuss, and ultimately find the best surgical solution for the patient.  This ultimately allows the surgeon to discuss these findings with families before even asking them to come to the hospital for the procedure, giving them the confidence that the surgical teams have practiced of their child’s specific condition.

VMWORLD 2017 is off to a great start

VMware Unveils New Security Model VMware AppDefense ( Link Here ) .  vmware also had several partners join them on stage to discuss how they use vmware’s portfolio of platforms, including the Red Cross.  The Red Cross discussed efforts underway in Texas to help those affected by the storm.  Finishing out the presentation, vmware hinted at more exciting announcements tomorrow when Michael Dell joins the stage.

Upgrading NSX from 6.2.4 to 6.2.8 In a vCloud Director 8.10.1 Environment

We use NSX to serve up the edges in vCloud Director environment currently running on 8.10.1.  One of the important caveats to note here, that when you do upgrade an NSX 6.2.4 appliance in this configuration, you will no longer be able to redeploy the edges in vCD until you upgrade and redeploy the edge first in NSX.  Then and only then will the subsequent redeploys in vCD work.  The cool thing about that though, is VMware finally has a decent error message that displays in vCD if you do try to redeploy an edge before upgrading it in NSX, you’d see an error message similar to:

—————————————————————————————————————–

“[ 5109dc83-4e64-4c1b-940b-35888affeb23] Cannot redeploy edge gateway (urn:uuid:abd0ae80) com.vmware.vcloud.fabric.nsm.error.VsmException: VSM response error (10220): Appliance has to be upgraded before performing any configuration change.”

—————————————————————————————————————–

Now we get to the fun part – The Upgrade…

A little prep work goes a long way:

  • If you have a support contract with VMware, I HIGHLY RECOMMEND opening a support request with VMware, and detail with GSS your upgrade plans, along with the date of the upgrade.  This allows VMware to have a resource available in case the upgrade goes sideways.
  • Make a clone of the appliance in case you need to revert (keep powered off)
  • Set host clusters DRS where vCloud Director environment/cloud VMs are to manual (keeps VMs/edges stationed in place during upgrade)
  • Disable HA
  • Do a manual backup of NSX manager in the appliance UI

Shutdown the vCloud Director Cell service

  • It is highly advisable to stop the vcd service on each of the cells in order to prevent clients in vCloud Director from making changes during the scheduled outage/maintenance.  SSH to each vcd cell and run the following in each console session:
# service vmware-vcd stop
  • A good rule of thumb is to now check the status of each cell to make sure the service has been disabled.  Run this command in each cell console session:
# service vmware-vcd status
  • For more information on these commands, please visit the following VMware KB article: KB1026310

Upgrading the NSX appliance to 6.2.8

  1. Log into NSX manager and the vCenter client
  2. Navigate to Manage→ Upgrade

nsx4

  1. Click ‘upgrade’ button
  2. Click the ‘Choose File’ button
  3. Browse to upgrade bundle and click open
  4. Click the ‘continue button’, the install bundle will be uploaded and installed.

 

nsx1

nsx2

  1. You will be prompted if you would like to enable SSH and join the customer improvement program
  2. Verify the upgrade version, and click the upgrade button.

nsx5

  1. The upgrade process will automatically reboot the NSX manager vm in the background. Having the console up will show this.  Don’t trust the ‘uptime’ displayed in the vCenter for the VM.
  2. Once the reboot has completed the GUI will come up quick but it will take a while for the NSX management services to change to the running state. Give the appliance 10 minutes or so to come back up, and take the time now to verify the NSX version. If using guest introspection, you should wait until the red flags/alerts clear on the hosts before proceeding.
  3. In the vSphere web client, make sure you see ‘Networking & Security’ on the left side.  If it does not show up, you may need to ssh into the vCenter appliance and restart the web service.  Otherwise continue to step 12.
# service vsphere-client restart

12. In the vsphere web client, go to Networking and Security -> Installation and select the Management Tab.  You have the option to select your controllers and download a controller snapshot.  Otherwise click the “Upgrade Available” link.

nsx8

13. Click ‘Yes’ to upgrade the controllers.  Sit back and relax.  This part can take up to 30 minutes.  You can click the page refresh in order to monitor progress of the upgrades on each controller.

nsx9

14.  Once the upgrade of the controllers has completed, ssh into each controller and run the following in the console to verify it indeed has connection back to the appliance

# show control-cluster status

15. On the ESXi hosts/blades in each chassis, I would run this command just as a sanity check to spot any NSX controller connection issues.

 esxcli network ip connection list | grep 1234
  • If all controllers are connected you should see something similar in your output

nsx10

  • If controllers are not in a healthy state, you may get something similar to this next image in your output.  If this is the case, you can first try to reboot the controller.  If that doesn’t work try a reboot.  If that doesn’t work…..weep in silence. Then call VMware using the SR I strongly suggested creating before the upgrade, and GSS or your TAM can get you squared away.

nsx11

16.  Now in the vSphere web client, if you go back to Network & Security -> Installation -> Host Preparation,  you will see that there in an upgrade available for the clusters.  Depending on the size of your environment, you may choose to do the upgrade now or at a later time outside of the planned outage.  Either way you would click on the target cluster ‘Upgrade Available’ link and select yes.  Reboot one host at a time that way the vibs are installed in a controlled fashion. If you simply click resolve, the host will attempt to go into maintenance mode and reboot.

17. After the new vibs have been installed on each host, run the following command to be sure they have the new vib version:

# esxcli software vib list | grep -E 'esx-dvfiler|vsip|vxlan'

Start the vCloud Director Cell service

  • On each cell run the following commands

To start:

# service vmware-vcd start

Check the status after :

# service vmware-vcd status
  • Log into VCD and by now the inventory service should be syncing with the underlining vCenter.  I would advise waiting for it to complete, then run some sanity checks (provision orgs, edges, upgrade edges, etc)

 

 

 

 

Performing A Database Health Check On vRealize Operation Manager (vROPS) 6.5

In a previous post I showed how you could perform a healthcheck, and possibly resolve database load issues in vROPs versions from 6.3 and older. When VMware released the vROPS 6.5, they changed the way you would access the nodetool utility that is available with cassandra database.

 $VCOPS_BASE/cassandra/apache-cassandra-2.1.8/bin/nodetool --port 9008 status

For the 6.5 release and newer, they added the requirement of using a ‘maintenanceAdmin’ user along with a password file.  The new command to check the load status of the activity tables in a vROPS 6.5+ is as follows:

  $VCOPS_BASE/cassandra/apache-cassandra-2.1.8/bin/nodetool -p 9008 --ssl -u maintenanceAdmin --password-file /usr/lib/vmware-vcops/user/conf/jmxremote.password status

Example output would be something similar to this if your cluster is in a healthy state:

vrops51

If any of the nodes have over 600 MB of load, you should consult with VMware GSS or a TAM on the next steps to take, and how to elevate the load issues.

Next we can check the syncing status of the cluster to determine overall health.  The command is as follows:

$VMWARE_PYTHON_BIN /usr/lib/vmware-vcops/tools/vrops-platform-cli/vrops-platform-cli.py getShardStateMappingInfo

Example output:

vrops52

The “vRealize Ops Shard” refers to the data nodes, and the Master and Master Replica nodes in the main cluster. The available status’ are RUNNING, SYNCING, BALANCING, OUT_OF_BALANCE, and OUT_OF_SYNC.

  • Out of Balance and Out of Sync should be enough to open an SR and have VMware take a look.

Lastly, we can take a look at the size of the activity table.  You can do this by running the following command:

 du -sh /storage/db/vcops/cassandra/data/globalpersistence/activity_tbl-*

Example Output:

vrops53

If there are two listed here, you should consult with VMware GSS as to which one can safely be removed, as one would be left over from a previous upgrade.

 

Removing Old vROps Adapter Certificates

I’ve come across this issue in previous versions of vRealize Operations Manager prior to the 6.5 release, where you delete an adapter for data collection like vSphere, NSX or VCD, and immediately try to re-create it.  Whether it was a timing issue, or vROps just didn’t successfully complete the deletion process, I’d typically get an error that the new adapter instance could not be created because a previous one exists with the same name.  Now there are two ways around this.  You can connect the adapter to whatever instance (VCD, NSX, vSphere) you are trying to collect data from using the IP address, instead of the FQDN (or vice-versa), or you can cleanup the certificate that was left behind manually as I will outline the steps below.

To resolve the issue, delete the existing certificate from Cassandra DB and accept the new certificate re-creating adapter instance.

1. Take snapshots of the cluster

2.  SSH to the master node.  Access the Cassandra DB by running the following command:

 $VMWARE_PYTHON_BIN $VCOPS_BASE/cassandra/apache-cassandra-2.1.8/bin/cqlsh --ssl --cqlshrc $VCOPS_BASE/user/conf/cassandra/cqlshrc

3. Access the database by running the following command:

use globalpersistence;

4.  We will need to look at the entries in the global persistence certificate store.  To do this, first list all the entries in globalpersistence.certificate store by running the following command:

SELECT * from globalpersistence.certificate;

5. From the list, find the desired certificate.  Now select that specific certificate with the following command:

 SELECT * from globalpersistence.certificate where key = 'Certificate.<ThumbprintOfVCCert>' and classtype = 'certificate' ALLOW FILTERING;

For example:

 SELECT * from globalpersistence.certificate where key = 'Certificate.e88b13c9e346633f94e46d2a74182d219a3c98b2' and classtype = 'certificate' ALLOW FILTERING;

6.   The tables which contains the information:

namespace | classtype | key | blobvalue | strvalue | valuetype | version
——————-+—————+——-+—————–+————–+—————–+———

7.  Select the Key which matches the Thumbprint of the Certificate you wish to remove and run the following command:

 DELETE FROM globalpersistence.certificate where key = 'Certificate.<ThumbprintOfVCCert>' and classtype = 'certificate' and namespace = 'certificate';

For example:

 DELETE FROM globalpersistence.certificate where key = 'Certificate.e88b13c9e346633f94e46d2a74182d219a3c98b2' and classtype = 'certificate' and namespace = 'certificate';

8.  Verify that the Certificate has been removed from the VMware vRealize Operations Manager UI by navigating to:

Administration > Certificates

9.  Click the Gear icon on the vSphere Solution to Configure.

10.  Click the icon to create an new instance. Do not remove the existing instance unless the data can be lost.  If the old instance has already been deleted prior to this operation, then this warning can be ignored.

11.  Click Test Connection and the new certificate will be imported.

12.   Upon clicking Save there will be an error stating the Resource Key already exists. Ignore this and click Close and the UI will show Discard Changes?. Click Yes.

13.   Upon clicking Certificates Tab the Certificate is shown for an existing VC Instance.  Now you should have a new adapter configured and collecting.  If you kept the old adapter for the data, it can safely be removed after the data retention period has expired.

Get VM Tools Version with VMware’s PowerCLI

I had an engineer visit me the other day asking if there was an automated way to get the current version of VMtools running for a set of virtual machines, and in this case, it was for a particular customer running in our vCenter.   I said there most certainly was using PowerCLI.

Depending on the size of the environment, the first option here may be sufficient, although it can be an “expensive” query as I’ve noticed it takes longer to return results.  Using PowerCLI, you can connect to the desired vCenter and run the following one-liner to get return output on the console.  Here I was looking for a specific customer in vCloud Director, so in the vCenter I located the customers folder containing the VMs.   Replace the ‘foldername’ inside the asterisks with the desired folder of VMs.  This command would also work in a normal vCenter as well.

Get-Folder -name *foldername* | get-vm | get-vmguest | select VMName, ToolsVersion | FT -autosize

Example output:

vmtools01

You can see that this example that folder has a mix of virtual machines running, some not (no ToolsVersion value returned), and has a mix of VMtools versions running.

What if you just wanted a list of all virtual machines in the vCenter, the whole jungle?

 Get-Datacenter -Name "datacentername" | get-vm | get-vmguest | select VMName, ToolsVersion | FT -autosize

In either case, if you want to redirect output to a CSV add the following to the end of the line

 | export-csv -path "\path\to\file\filename.csv" -NoTypeInformation -UseCulture

Example:

Get-Folder -name *foldername* | get-vm | get-vmguest | select VMName, ToolsVersion | export-csv -path "\path\to\file\filename.csv" -NoTypeInformation -UseCulture

 

Another method/example of getting the tools version, and probably the fastest is using ‘Get-view’. A much longer string of command-lets, but this would be the ideal method for large environments if a quick return of data was needed, lets say for a nightly script that was least impactful to the vCenter.

 Get-Folder -name *foldername* | Get-VM | % { get-view $_.id } | select name, @{Name=“ToolsVersion”; Expression={$_.config.tools.toolsversion}}, @{ Name=“ToolStatus”; Expression={$_.Guest.ToolsVersionStatus}}

Example Output:

vmtools02

If you are after a list of all virtual machines running in the vCenter, a command similar to this can be used:

 Get-VM | % { get-view $_.id } | select name, @{Name=“ToolsVersion”; Expression={$_.config.tools.toolsversion}}, @{ Name=“ToolStatus”; Expression={$_.Guest.ToolsVersionStatus}}

VMware has put together a nice introductory blog on using get-view HERE

Just like last time, if you want to redirect output to a CSV file just take the following on to the end of the line for either method ie specific folder or entire vCenter:

 | export-csv -path "\path\to\file\filename.csv" -NoTypeInformation -UseCulture

 

 

 

VMware Certified Professional 6 – Data Center Virtualization

VMW-LGO-CERT-PRO-6-DATA-CTR-VIRT

Certificate

I do apologies for being MIA these past couple of weeks.  Anyone who has taken the VCP exam knows, it can be a brutal test to study for.  I thought it best to keep my head down, and study hard so I can pass the VCP6-DCV exam on the first go around.

As I wait for VMware Education to finalize my records, I will be readying new material to share with my fellow virtualization geeks in the coming weeks ahead.

All the Best,

Cory B.