cancel
Showing results for 
Search instead for 
Did you mean: 
Create Post

Orion Platform: Preparing for the Upgrade to 2018.2

Product Manager

Starting with NPM 12.2, SolarWinds has embarked on a journey to transform your Orion deployment experience with fast and frequent releases of key deployment components. The first step was revamping the legacy installer to the new and improved SolarWinds Orion installer. The installer was able to deploy new or upgrade an entire main poller in one seamless session. The second iteration of the installer released the capability to do the same for your scalability engines. In this release NTA has been updated to utilize a MSSQL database, allowing us to happily say that the SolarWinds Orion installer is truly an All-in-One installer solution for your Orion deployment. For NPM 12.3, we have made tremendous scalability improvements that allow you to utilize even more scalability engines. As a result, your Orion deployment upgrades gain in complexity, so the installer team is providing additional updates to how you can stage your environment for minimal upgrade time.

Normal Upgrade Process

Using the All-in-One SolarWinds Orion installer, your upgrade process will look like the following.

Step one:

Review all system requirements, back up your database and if possible snapshot the Orion deployment. This will be especially important in this release, as the NTA Flow Storage database requirements have changed. Note: Flow Storage database refers to the database instance that stores NTA collected flow data. In previous versions this was utilizing a Fastbit database, but in this release has been updated to use MSSQL with a minimum version of 2016. An Orion database is the primary database that stores all polled data from NPM and other Orion products.

Step two:

Download the NPM 12.3 installer, selecting either the online or the offline variant according to your system requirements. Note: the SolarWinds Orion installer is

Step three:

Run the installer on your main poller and upgrade it to completion. If you have any other Orion product modules installed, the installer will upgrade this instance to the latest versions of those modules at the same time to maintain compatibility with the new Orion Platform 2018.2. If there are new database instances to be configured, that will be handled during the Configuration Wizard stage of the main poller upgrade. This release of the installer has a new type of preflight check that requires confirmation from you before proceeding. In the example below, is one for the NTA upgrade. Click for details to see the confirmation dialog and select yes or no.

pastedImage_0.png

Configuration Wizard step for NTA:

pastedImage_1.png

Step four:

If you don’t have any scalability engines, e.g Additional Polling Engines, Additional Websites or HA Backups you’re ready to explore all of the new features available in this version!

Scalability Engines

For those environments utilizing scalability engines or for those who are looking to try them out, this section will guide you through the process of deployment. Even if you have not utilized scalability engines previously, trying them out to test the scale improvements is incredibly easy. Like every SolarWinds Orion product, they are available for an unlimited 30-day free evaluation.

Deploying a fresh scalability engine is handled with the same installer that you downloaded for the main poller.

1. Copy the installer to your intended server and Click to “Run as Administrator”

Note: If you downloaded the offline installer, which is about 2 GB, the download process to your server can take some time and does not currently stage the scalability engine for faster upgrade. In the future, this is something we’d like to improve but is not an available feature for this release.  if you’d like to shorten the initial download of installer file to server, you can always use the online installer to set up your scalability engine. This installer file is about 40 MB so the download of installer file time to the server is much shorter. This will still meet offline requirements because when selecting the “Add a Scalability Engine” option, it will choose to download from the main poller to maintain version compatibility and does not require internet access. As always, the 40 MB scalability engines installer is also available for download from the All Settings -> Polling Engines page.

2. Select the “Add a Scalability Engine” option.

Picture1.png

3. Similar to the main poller upgrade process, at this point system checks that are specific to scalability engines will be run.

Note: Anything tagged as a blocker may need confirmation or action from you before proceeding.  If this is the case, address those issues and run the installer again. Things that are tagged as a warning or informational message are simply for your awareness and will not prevent your installation from proceeding.

4. Select the type of scalability engine that you are looking to deploy, and then complete the steps in the wizard to finish your installation per your normal process.

Upgrading a scalability engine, is also handled through the same installer. However, this is where you have an opportunity to utilize our staging feature.

Note: If you were to proceed with your normal practice of putting the scalability engines installer on each server you need to upgrade, and then manually upgrading, that process will work perfectly well with no changes. Please read through the “Staging Your Environment for your Scalability Engines Upgrade” section below to see the alternative workflow that allows you to stage your environment.

Staging Your Environment for Your Scalability Engines Upgrade

For customers with more than a handful of scalability engines or with some distributed over WAN links, we noticed that they were occasionally experiencing extremely high download times from their main poller to their scalability engines. In addition, there was no centralized area where one could see the upgraded state of the scalability engines. Navigate to "All Settings", and click "High Availability Deployment Summary" and you will see the foundational pieces for an Orion deployment view.

pastedImage_2.png

The Servers tab contains the original High Availability Deployment Summary content, and is where you can continue to set up additional HA pools and HA environment.

pastedImage_0.png

Check out the new Deployment Health tab! You may not have heard of our Active Diagnostics tool, but it comes prepackaged with every install of the Orion Platform with test suites designed to test for our most common support issues. We've brought that in depth knowledge to your web console in the new Deployment Health view. With nightly run tests across your Orion Deployment, every time you come to this page you will see if there are any issues that could be a factor in the performance of Orion or your upgrades.

hadeployment.png

You are able to refresh a check if you're working on an issue and wish to see an updated test result. If there are tests that you don't want to address, silence them to hide the results from the web console. Click on the caret to the right and you'll be able to see more details and a link to a KB article that will give you remediation advice.

On the Updates tab is where you will be able to stage your scalability engines.

The first page of the wizard will let you know if there are updates that are available to be installed on your scalability engines. At this point you've upgraded your main poller, so there are definitely updates available!  Click "Start" to get started!

pastedImage_1.png

The second page is where we are testing the connection to each of the scalability engines. If we are able to determine the status of these engines, we'll give you the green light to proceed to the next step. Common issues that could prevent this from being successful could be that the SolarWinds Administration Service has not been updated to the correct version or is not up and running at this point. Click "Start Preflight Checks" to proceed.

pastedImage_3.png

Similar to the Deployment Health tab, these are running preflight checks across your Orion Deployment. You'll be able to see all of the same preflight checks that were available through the installer client, except centralized to one view. If there are blockers present on this screen, you can still proceed in this flow if at least one scalability engine is ready to go, but please note down those scalability engines with blockers. You will need to address those blockers before an upgrade can occur on those servers. Click "Start download" to start the staging process.

pastedImage_5.png

pastedImage_6.png

pastedImage_7.png

At this point, we are starting the download process of every msi needed to upgrade your scalability engines. In this example, I'm only staging one scalability engine, but if you  have multiple, you can see the benefits in time savings right away! All of the downloads will be triggered in parallel.

pastedImage_8.png

Sit back and relax as we stage your environment for you. You can even open up RDP sessions to those servers with one click from this page.

pastedImage_9.png

When everything has finished downloading, we will let you know which servers are ready to install. Click on the "RDP' icon to open your RDP session to the server.

pastedImage_10.png

On your desktop, you should see the SolarWinds scalability engines installer waiting for you to click on and finish the upgrade.

pastedImage_11.png

Visually you will run through the same steps that you normally would in clicking through the installer wizard. However, when you actually get to the installation part, you'll notice that there is no download appears in the progress bar. Finish your upgrade and move on to the next!

I hope you enjoy this update to how you can upgrade your Orion Deployment. I'm always looking for feedback on how we make this as streamlined as possible for you.

22 Comments
Product Manager
Product Manager

Looking for feedback in NPM 12.3 Orion 2018.2 Upgrade Feedback  on how your upgrade is proceeding

Level 13

This installer looks awesome. Excellent new features! Looking forward to the upgrade.

Level 10

Personally I think they spend too much time one changing from one thing to the other.

I think they need to focus on the management of the product as it's still very cluncky...

Level 9

The installer is great and upgrade was flawless.

Level 9

My upgrade included 6 pollers across two domains.

The installer correctly identifies and targets which components require update.  A small note here.  I might be in the minority, but I would prefer not to have to upgrade all components with this installer.  I'd like to be able to opt-out, assuming the components do not have dependencies.

The system check is a nice feature and called out that MS SQL 2016 SP1 was required for an NTA upgrade.  It allowed me to continue, but notified me that certain functions associated with the NTA component would not be available until I upgrade.  The check notified me about its need to upgrade the WPM player.  It allowed me to choose not to upgrade these, which I wanted for my case.  The check did its due diligence in notifying me about the support for Windows 2012 as well as MS SQL 2012 for the Orion DB.  All checks were appropriate and notified me of upcoming potential projects to plan for the next iteration.

The one item that the upgrade failed on and one where I am currently investigating is that all of the deployed nodes with agents in my environment did not update.  I had no notification of this and only found out when I began to receive alerts on 'down' nodes as the agent could not communicate between the node and poller.

Otherwise, the upgrade went well.

Level 9

Update:  The agents failed to update as the Solarwinds AgentUpdater service on the node itself did not restart to allow SW to update the agent.  Once the service was restarted, the agent updated.

Product Manager
Product Manager

greg.remer@spe.sony.com  wrote:

Personally I think they spend too much time one changing from one thing to the other.

I think they need to focus on the management of the product as it's still very cluncky...

Hi Greg,

On the platform side, we've been consistently investing in the installer experience for the last few years and will continue to do so in the future. We've got big plans, as indicated by the presence of "Centralized Upgrades" on What We're Working on for NPM (Updated June 1st, 2018) 

Management of the product is also something that we have put some time and energy into with various projects like changing the way you customize views Radically Improved Dashboard View Customization Workflow  and adding integrated search Search Integrated into the Global Navigation  Is there some management functionality you're looking to improve? Have you added a feature request for what you're looking for? Network Performance Monitor Feature Requests  I'd love to hear more.

Product Manager
Product Manager

cha-chang!  wrote:

My upgrade included 6 pollers across two domains.

The installer correctly identifies and targets which components require update.  A small note here.  I might be in the minority, but I would prefer not to have to upgrade all components with this installer.  I'd like to be able to opt-out, assuming the components do not have dependencies.

The system check is a nice feature and called out that MS SQL 2016 SP1 was required for an NTA upgrade.  It allowed me to continue, but notified me that certain functions associated with the NTA component would not be available until I upgrade.  The check notified me about its need to upgrade the WPM player.  It allowed me to choose not to upgrade these, which I wanted for my case.  The check did its due diligence in notifying me about the support for Windows 2012 as well as MS SQL 2012 for the Orion DB.  All checks were appropriate and notified me of upcoming potential projects to plan for the next iteration.

The one item that the upgrade failed on and one where I am currently investigating is that all of the deployed nodes with agents in my environment did not update.  I had no notification of this and only found out when I began to receive alerts on 'down' nodes as the agent could not communicate between the node and poller.

Otherwise, the upgrade went well.

You nailed it with the "assuming components do not have dependencies" comment. Being that all of our products are on the Orion Platform, it's fair to say that they all have dependencies.

I'm curious with your set up of 6 pollers across two domains - how long did it take for you to upgrade your Orion deployment? did you run any pieces of the upgrade in parallel?

Level 10

Hi,

Here is the feature Request I submitted.

To be honest, I've been a long time admin on Solarwinds and it just feels like Solarwinds don't always listen to us as users.

I get the upgrade installer is a good thing, but in the past it still worked? we just had more to install. If you think about what an admin has to do daily, that isn't one of them.

My guess is resources are being put into too many baskets, when in fact they need to go back to basics and look at what they offer to us admins.

Adding Nodes should be one of the easiest parts to NPM especially with all the modules, yet if feels like thing are just put here and there and the management is just a nightmare to add nodes.

Keep things simple, and tidy and then work on the back end stuff.

Please review my feature request as I have spoken to lots of admins and feedback I've had personally is that's a great idea, would help us all so much.

Sorry to go on, but thanks.

Product Manager
Product Manager

greg.remer@spe.sony.com  wrote:

Hi,

Here is the feature Request I submitted.

 

To be honest, I've been a long time admin on Solarwinds and it just feels like Solarwinds don't always listen to us as users.

I get the upgrade installer is a good thing, but in the past it still worked? we just had more to install. If you think about what an admin has to do daily, that isn't one of them.

My guess is resources are being put into too many baskets, when in fact they need to go back to basics and look at what they offer to us admins.

Adding Nodes should be one of the easiest parts to NPM especially with all the modules, yet if feels like thing are just put here and there and the management is just a nightmare to add nodes.

Keep things simple, and tidy and then work on the back end stuff.

Please review my feature request as I have spoken to lots of admins and feedback I've had personally is that's a great idea, would help us all so much.

Sorry to go on, but thanks.

Thanks for sending the feature request link. I took a look and it looks to be a feature under the purview of some other teams so I'll chat with their PMs to bring some additional attention to it.

Going back to basics is definitely something we're doing, but with less fanfare then our new installer. For instance, do you have VMAN? In the latest releases you can see a new version of "Add a Node" flow that is being built as a first step. As we prove it out for the virtualization wizard flow, we're looking to pull the new add a node flow to the other products. If you don't have VMAN, you can try it out here http://oriondemo.solarwinds.com/ui/add-node

Level 9

Hi Serena,

Generally, I do not run the upgrade in parallel.  I prefer the option of having a poller available to assign nodes to if I hit an issue with the poller being upgraded.  I will also note that after running the upgrade on the main poller, the wording suggest to run the upgrade to APE's in the order and does not specifically call out that you can run them in parallel. 

My upgrade involved upgrading the following items to get to Orion Platform 2018.2.1.SP1, CloudMonitoring 2.0.1, SRM 6.6.0, IPAM 4.7.0, SAM 6.6.1, DPAIM 11.1.0, NetPath 1.1.3, NTA 4.2.3, and WPM 2.2.2.  I have 4 pollers (the main poller and DB are in this domain) in one domain and 2 others in the other. 

As far as timing, I found that upgrading all of the components I needed took about 20 minutes to complete per poller in my main domain (the one with the main poller and DB).  BTW, the scalability engine that is provided for you after the initial upgrade to the main poller, is a fantastic value add. 

Unfortunately, the time it took to run the upgrade with the scalability engine on the other domain took much, much longer.  It took about 4 hours to download and start the upgrade for each.  I do not put fault on SW for that, as that is a product of my environment.  Hence, my preference to have different components as I could manually move them over one at a time before running the upgrade at once.  As a side note, on a previous upgrade version, I was one of the folks that had an issue with the downloading process hitting a hard stop of 30 minutes to download the necessary upgraded components.  This forced me to have to rebuild new pollers after running the install locally and copying over that poller to its permanent destination.  So I am glad that the timeout is removed, even if it does take me hours for my APEs.  What I appreciate with the new installer is that each component that the installer is downloading is called out and you have a 'moving' icon that tells me that the process is still going and not hung. 

Product Manager
Product Manager

cha-chang!  wrote:

Hi Serena,

Generally, I do not run the upgrade in parallel.  I prefer the option of having a poller available to assign nodes to if I hit an issue with the poller being upgraded.  I will also note that after running the upgrade on the main poller, the wording suggest to run the upgrade to APE's in the order and does not specifically call out that you can run them in parallel. 

My upgrade involved upgrading the following items to get to Orion Platform 2018.2.1.SP1, CloudMonitoring 2.0.1, SRM 6.6.0, IPAM 4.7.0, SAM 6.6.1, DPAIM 11.1.0, NetPath 1.1.3, NTA 4.2.3, and WPM 2.2.2.  I have 4 pollers (the main poller and DB are in this domain) in one domain and 2 others in the other. 

As far as timing, I found that upgrading all of the components I needed took about 20 minutes to complete per poller in my main domain (the one with the main poller and DB).  BTW, the scalability engine that is provided for you after the initial upgrade to the main poller, is a fantastic value add. 

Unfortunately, the time it took to run the upgrade with the scalability engine on the other domain took much, much longer.  It took about 4 hours to download and start the upgrade for each.  I do not put fault on SW for that, as that is a product of my environment.  Hence, my preference to have different components as I could manually move them over one at a time before running the upgrade at once.  As a side note, on a previous upgrade version, I was one of the folks that had an issue with the downloading process hitting a hard stop of 30 minutes to download the necessary upgraded components.  This forced me to have to rebuild new pollers after running the install locally and copying over that poller to its permanent destination.  So I am glad that the timeout is removed, even if it does take me hours for my APEs.  What I appreciate with the new installer is that each component that the installer is downloading is called out and you have a 'moving' icon that tells me that the process is still going and not hung. 

You're absolutely correct about the wording, and I have taken a note down to improve that to  let users know that parallel installs are acceptable. Thank you for pointing that out! The details about your scalability engine upgrade was exactly what I needed, and I'll be tweaking some details on our upcoming changes to upgrade to accommodate. Thanks for taking the time to let us know.

Level 9

So is the My Orion Deployment - Health and Update check only available with a HA setup?  I have 1 poller, 1 NTA fastbit server and 1 DB server, and it would be nice to see the Health and Updates for a smaller environment. 

I installed in our Dev environment to test functionality, but Deployment Health and Updates Available are blank.

This is what I have installed in our env.

Orion Platform 2018.2.1 SP1, WPM 2.2.2, VNQM 4.5.0, NCM 7.8, CloudMonitoring 2.0.1, NPM 12.3, DPAIM 11.1.0, NTA 4.2.3, NetPath 1.1.3

In response to a different comment, it would be nice to update/install a product without the installer forcing an update to all products.  Last fall we added SAM to our install, the Smart Start resource said we could just install SAM, without impact, but installer would not let us install without also deploying to NPM, NCM and NTA.  we had to postpone project while we went and tested features in our dev environment.

I also hope there is an effort to clean up UI inconsistencies across different products, widgets etc.  and the classic Manage Nodes view never goes away.

Level 9

I felt the same way with having to update all components vs. the component I want, but I have changed my tune and for me, I have just adjusted to additional testing.  With that said, I would say that the documentation as well as support does not always make this clear.  Even with the most latest updates, the indication is to just be able to update a component, when that is not really the case.  The wording should be consistent across the board. Just my two cents. 

Product Manager
Product Manager

lsalinas  wrote:

So is the My Orion Deployment - Health and Update check only available with a HA setup?  I have 1 poller, 1 NTA fastbit server and 1 DB server, and it would be nice to see the Health and Updates for a smaller environment. 

I installed in our Dev environment to test functionality, but Deployment Health and Updates Available are blank.

This is what I have installed in our env.

Orion Platform 2018.2.1 SP1, WPM 2.2.2, VNQM 4.5.0, NCM 7.8, CloudMonitoring 2.0.1, NPM 12.3, DPAIM 11.1.0, NTA 4.2.3, NetPath 1.1.3

In response to a different comment, it would be nice to update/install a product without the installer forcing an update to all products.  Last fall we added SAM to our install, the Smart Start resource said we could just install SAM, without impact, but installer would not let us install without also deploying to NPM, NCM and NTA.  we had to postpone project while we went and tested features in our dev environment.

I also hope there is an effort to clean up UI inconsistencies across different products, widgets etc.  and the classic Manage Nodes view never goes away.

Hi Larry,

No sir, you have access to this page whether you have HA set up or not, it's simply put side by side with the HA server screen. If it's blank that means you're running into an unexpected issue. If you have an opportunity, can you check if you can see if the page is available via other web browser types?

The All in One installer will upgrade your to the latest based on the Orion platform simply because it will ensure that everything is compatible with the version of the Orion platform you are installing on your box, while ensuring key security fixes are installed in a compatible manner.

There is definitely an effort to clean up UI inconsistencies. Do you have examples of some that you'd like to see? meech​ is collecting these types of feedback so we'd love to get your point of view.

Level 13

I always want to hear about UI inconsistencies!  Please either reply here or email them to me directly at any time, direct email is kellie.mecham@solarwinds.com.

Level 12

Good Afternoon Serena,

Just double checking prior to upgrading that the offline installer packages are synced up and have the latest available online updates/ Orion 2018.2 hot fix 3 etc? The offline installer in the portal shows an update date of May 31st, which was before hotfix 3 I believe?

npm12.3-upgrade.png

Thanks.

Product Manager
Product Manager

prawij  wrote:

Good Afternoon Serena,

Just double checking prior to upgrading that the offline installer packages are synced up and have the latest available online updates/ Orion 2018.2 hot fix 3 etc? The offline installer in the portal shows an update date of May 31st, which was before hotfix 3 I believe?

npm12.3-upgrade.png

Thanks.

Yes sir, the offline installer packages have been synced up. A few of the dates were not updated yet but I've confirmed with our release management team that the offline installers are all good to go with the HF3.

Level 11

serena​, Could you help with the latest Offline Orion platform 2018.2((NPM 12.3, NTA 4.4,IPAM 4.7,NCM 7.8) downloader url, We would like to start with a fresh evaluation Installation.

Product Manager
Product Manager

dhinagar_j  wrote:

serena , Could you help with the latest Offline Orion platform 2018.2((NPM 12.3, NTA 4.4,IPAM 4.7,NCM 7.8) downloader url, We would like to start with a fresh evaluation Installation.

You can always grab a new copy of the offline installer via our http://solarwinds.com/downloads  page. Simply click on any Orion product for instance NPM, fill out the registration form and the offline link will be to the right.

pastedImage_3.png

Level 9

I like the new installer and certainly speeds things up when updated all products.

I wish there was still a way to pick and choose or maybe ignore a product if there are no dependencies.

We had a product that we dropped maintenance for as we had no product updates for 2 years, we had no choice but to uninstall the product just to apply a hotfix to another product.

Level 9

Fentstic serena . Feature of storing the flow data in Orion SQL DB which we were waiting for long term is introduce in this version its a good news.

We are planning to upgrde next month will post you the status.

About the Author
Serena has been a software developer, program manager, product owner, and has found her passion in being a product manager. She loves building solutions for customer problems. Currently she's a Senior Product Manager at SolarWinds, enthusiastically responsible for Server & Application Monitor.