At VMworld 2011 in Copenhagen VMware unveiled a significant revamp of their management suites, including a new version of vCenter Operations Manager (v5 to align with the vSphere release). vCenter Operations is now a suite of tools which includes vCenter Configuration Manager, the new vCenter Infrastructure Navigator (which I’ll cover in a later blogpost) and vCenter CapacityIQ (which is now fully integrated into vCOps, the standalone CapacityIQ is now end of life).
Although announced at VMworld it wasn’t publicly available until Jan 2012 when VMware formally launched vCOps v5. Coming less than a year after the release of the first version it’s apparent that VMware see this as an important product which is evolving fast. Steven Herrod, VMware’s CIO stated recently at the Italian VMUG (around the 5 minute mark) that vCOps ‘is becoming the most adopted new technology that VMware has ever had’. The vCenter Operations suite is still aimed at infrastructure monitoring as opposed to application monitoring (despite the addition of Infrastructure Navigator) – VMware’s solutions aimed at the application tier belong to the vFabric suite. For a good overview of where vCOps and vFabric Hyperic fit into VMware’s cloud suite read Dave Hill’s blogpost on the subject.
If you aren’t familiar with vCenter Operations here are the kind of problems it aims to address;
- Is your virtual infrastructure healthy?
- What serious problems should I address immediately?
- Is the workload in my environment normal?
- Am I using the resources in my environment efficiently?
- How long do I have before resources run out?
- What impact did a recent change have?
A few people have already posted articles which I’d recommend reading;
- David Davis has a great review including available versions, top 5 new features and his conclusions
- Julian Wood posted a good summary along with his initial thoughts
- VMware’s official blogpost about the new features
- My review of the first version of vCOps. A year on it’s still got a few good points and is a good intro to the product
With v1.0 I concluded that it was a great product but there were a few reasons why it wasn’t for me, primarily the lack of email notifications and pricing. In this post I’ll cover the requirements and deployment considerations for the new version and in part two I’ll cover day to day use and new features. The final part will cover the capacity features along with info about pricing and my conclusions.
UPDATE APRIL 2012 – VMware have just launched 2.5 hrs of free training for vCOps.
Requirements and editions
You can get a 60 day evaluation of the Enterprise version, giving you more features compared to the previous version’s trial which was the Standard edition. Also similar to the previous version, vCOps is packaged as an appliance in the .OVF format but instead of a single VM there are now two VMs bundled as a vApp. The first is a front end server for the UI and the second is dedicated to the patented analytics engine. There is a standalone install as opposed to the vApp but it seems to be an edge case and comes with a few drawbacks (as per this discussion on the vCOps forums).
NOTE: The 60 days start from when you register for the trail, NOT when you install the vApp so only register when you’re ready to start your eval.
Before we get into the specifics it’s worth bearing in mind a summary of the different versions of vCOps. NOTE: there is a single download – the license determines the edition and hence the features;
- Standard – vSphere performance and basic notifications
- Advanced – as above plus capacity management
- Enterprise – as above plus chargeback, reporting and advanced alerting, and configuration management (limited)
- Enterprise plus – as above plus enhanced configuration management and extensibility (third party plugins, customizable dashboards etc)
You can see a full feature breakdown via VMware’s feature summary and version matrix. Both Standard, Advanced and Enterprise are licensed ‘per VM’ whereas Enterprise+ is negotiable! There is a version of vCOps especially designed for View environments but it’s currently in beta with no official release date.
The Enterprise and Enterprise+ editions can work with third party plugins which allow them to operate with a variety of alternative products. The list of adapters as of March 2012 includes products from Microsoft, BMC, Oracle, Netapp, EMC, HP, IBM, and VMware’s Hyperic suite. As a Netapp customer I’d love to see what value combining vCOps with Netapp metrics could deliver although I’m not likely to go Enterprise anytime soon due to the cost. Enterprise+ licensing costs aren’t published on the VMware website but Enterprise is £195 per VM so for a 1000VM estate you’re looking at nearly £200,000. At those prices you’re competing with the big boys (Microsoft’s System Center suite, Oracle Enterprise Manager, IBM Tivoli, Netuitive etc) all of whom have more mature products with an existing customer base. I know I’d be up against Oracle’s offering and I know which will win (even if only for political reasons) and it’s not VMware!
NOTE: The Netapp adapter notes detail the metrics it’ll collect and a rough architecture for those who are interested!
It’s worth noting that the resource http://premier-pharmacy.com/product/crestor/ requirements have increased although you tailor them by selecting one of three sizes (based on the number of VMs you need to analyze);
- small (up to 1500 VMs) – 16GB RAM, 4 vCPUs, 900GB storage with 1500 IOps
- medium (1500 – 3000 VMs) – 25GB RAM, 8vCPUs, 1.8TB storage with 3000 IOps
- large – (3000+ VMs) – 34GB RAM, 16 vCPUs, 3.6TB storage with 6000 IOps
The significant storage requirement is for the performance heavy metrics store (which is why VMware recommend eager-zeroed thick provisioning). vSphere5 includes over a hundred new metrics and vCOps analyses more than its predecessor which means more information (and more storage required!). By using a proprietary database specially designed to store time-series information VMware claim to have improved performance compared a traditional relational database (by a factor of ten or more) or even Hadoop!. Interestingly Kit Colbert states (in VMworld presentation CIM2452) that vCOps queries the hosts directly NOT the vCenter database as I previously thought although permissions are assigned in vCenter.
A new feature is the ability for a single vCOps install to monitor multiple vCenters (both Windows and the Linux appliance) which is certainly welcome. The requirements for this aren’t overly clear from the documentation – while it will detect vCenters using Linked-Mode during the setup wizard it doesn’t use the Linked-Mode to get statistics and you can monitor multiple vCenters even if they aren’t in Linked-Mode. The single set of credentials you provide must have access to every vCenter you register so potentially two vCenters in the same domain but not in Linked-Mode would be fine. I was concerned that collecting statistics from a remote vCenter (or hosts as it goes direct) over a WAN might degrade performance but after some tests I found it made a negligible difference (my WAN latency was <10ms). Nice!
The product integration has been greatly improved for the capacity, configuration, and chargeback features and you can now upgrade from one version to another more easily. This change does mean there are some caveats around upgrades however, especially if you previously had multiple products such as CapacityIQ and Configuration Manager running separately.
The deployment is straightforward as it’s wizard driven but you will need to create IP pools in vCenter BEFORE you start the deployment. IP pools are covered in the install video from VMware and the official deployment docs which also cover further requirements (vCenter and vSphere versions, port requirements etc). You can read more about how network attributes are passed to vApp VM’s but bear in mind that changing the IPs for a vApp VM isn’t always straightforward.
NOTE: The use of a vApp does mean deployment in environments without DRS (a requirement for resource pools and hence vApps) is more convoluted. See VMwareKB2013695.
With the IP details sorted you then follow the usual vApp deployment process. Once the vApp is deployed you can configure various settings from the web administration page (https://<UI VM’s IP address>/admin/), although only the first of the below actions is required;
- Register with one or more vCenter servers
- Configure SNMP and SMTP.
NOTE: There is no way to easily test SMTP settings during setup (such as a ‘Test’ button) which would be nice.
- Configure SSL certs. The default option is for a self signed certificate but if you want your own follow this blogpost by Eric Bussink (with help from Julian Wood)
- Configure permissions (vCOps adds new roles to vCenter). As with vCOPS v1 you can use permissions within vCenter to restrict the inventory covered (as per VMwareKB1036195, this blogpost, and this forum discussion) which allows you to license only parts of your virtual estate. Care needs to be taken as you can restrict the analysis and invalidate some results this way. I’m guessing this would pose problems in a multi-tenant installation using vCD for example!
- Update the appliance with an offline bundle
- Generate a diagnostic bundle to send to VMware support
- Reset passwords for the administrator and root accounts.
The last step is to add your vCOps license (using the VI client) in the vCenter licensing portal. Make sure you’ve deployed the appliance before you try adding the license key otherwise you’ll get an error as per VMwareKB1000609.
If you run into any issues check the release notes as there are a few gotchas for certain environments.
vCOPS in 5 mins – an amusing overview from VMware! (marketing rather than technical)
vCOps learning resources (Gregg Robertson)
5 Minutes with vCenter Operations Manager 5 (Bob Plankers)
VMworld 2011 sessions (login required);
- CIM2452 – VMware vCenter Operations DeepDive (Kit Colbert). NOTE: This covers v1 not v5 but the fundamentals still apply.
- CIM2285 – Automated Infrastructure and Operations Managment with VMware vCenter Operations