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

NPM 12.3 Orion 2018.2 Upgrade Feedback

What has your upgrade to NPM 12.3 on Orion Platform 2018.2 looked like? We on the product manager team would like to hear about it all, the good the bad and the ugly! For a starting point here is a quick getting started blog post on upgrading to 2018.2 Orion Platform: Preparing for the Upgrade to 2018.2

Tags (1)
310 Replies
Level 12

Hi Serena,

The main product installs went smoothly and in a multi-server environment this was a massive improvement from the previous workflow.

However I did have to create a number of new servers in order to facilitate the install, and i've had several problems since. Still pinning down what's caused what so I wont blame the product for everything.

The current version of NTA needing a higher database version than the rest of the suite threw me off, and I'm operating without NTA for a bit due to that.

One problem which I believe is directly related to the solarwinds install is that alert emails which included graphs have all broken.

I believe this is due to the processing of ampersands somewhere in the software. All the pictorial, non-interactive graphs I was aware of included operators like so:

"....CustomChart.aspx?ChartName=PercentDiskUsage&Title=&SubTitle=..."

These now get translated into their html text symbol equivalent instead of being passed through:

"...ChartName=IfAggrUtilizationIn&NetObject=&..."

Which isnt procced in the same way by the web server, and so the images fail to generate

I've got an open query with our vendor before it escalates up to yourself, but do you have either a fix, or guidance for how best to implement similar resources in NPM 12.3?

Interesting; though I have never seen that happen I know what you are talking about. You do see a lot of "Adjusted" characters on read. There is a lot of data in the system that gets translated to or from web/html syntax. Though the only time I have ever had this be an issue is when I read the data. Only in reports, or breaking out system data to decipher structure and build have I had to either replace, or cast those specific characters back to what they are supposed to be &<>;= in most cases, but I believe there are a few others to consider if you dive deep. I would not expect that to happen unless you copy or parse manually.

adam.beedell  wrote:

Hi Serena,

The main product installs went smoothly and in a multi-server environment this was a massive improvement from the previous workflow.

However I did have to create a number of new servers in order to facilitate the install, and i've had several problems since. Still pinning down what's caused what so I wont blame the product for everything.

The current version of NTA needing a higher database version than the rest of the suite threw me off, and I'm operating without NTA for a bit due to that.

One problem which I believe is directly related to the solarwinds install is that alert emails which included graphs have all broken.

I believe this is due to the processing of ampersands somewhere in the software. All the pictorial, non-interactive graphs I was aware of included operators like so:

"....CustomChart.aspx?ChartName=PercentDiskUsage&Title=&SubTitle=..."

These now get translated into their html text symbol equivalent instead of being passed through:

"...ChartName=IfAggrUtilizationIn&amp;NetObject=&amp;..."

Which isnt procced in the same way by the web server, and so the images fail to generate

I've got an open query with our vendor before it escalates up to yourself, but do you have either a fix, or guidance for how best to implement similar resources in NPM 12.3?

Adam, I'm not aware of any other cases where this has been reported. That's not to say it's not a bug, but perhaps you're the first to notice. Please open a case with support so we can investigate this issue.

Hi Jeremy,

Will do - Am i to take it then that the processes for adding graphical information to alert emails shouldn't have changed, and that linking to content like that is still kosher?

0 Kudos

adam.beedell  wrote:

Hi Jeremy,

Will do - Am i to take it then that the processes for adding graphical information to alert emails shouldn't have changed, and that linking to content like that is still kosher?

We have done nothing purposful to prevent it if that's what you're asking. There's something clearly wrong as the HTML is not being properly escaped. Why however, is unknown as I'm unaware of any obvious changes which would have caused this behavior to change.

0 Kudos
Level 12

Someone had mentioned on the first page:

"My team and I have found 2 other bugs so far.

1.) In alert actions sending an email using the ${DefaultEmailTo} or ${DefaultEmailCC} or ${DefaultEmailBCC} or ${DefaultEmailFrom} these values get corrupted in the database. The issue is something is causing the values to start iterating additional brackets like this ${{DefaultEmailTo}} or ${{{DefaultEmailTo}}} or ${{{{DefaultEmailTo}}}}. We have had to create sql scripts to update the DB and replace the corrupt values with what we are expecting. This is a temporary fix until next time we create a new alert email action and it starts all over again."

I'm experiencing this myself - ticket has been opened, just wondering if anyone else is experiencing it and what the fix was? In our case I'm seeing extra { }'s around any variable in an alert action.

0 Kudos

Hi Prawij,

I have experienced this and opened a ticket on 15 Jun 2018 11:03 PM, but SolarWinds was not able to re-create it. Good Luck...

For your reference here is my ticket number:

Case # 00117980 - After NPM 12.3 upgrade we see dup brackets around the email fields

-Phil

0 Kudos
Level 13

Curious, whether to use the "online" vs "offline" installer for this upgrade.  It seems the consensus from this thread  (I've tried to catch up on it since the 2018.2 platform came out) is to use the "offline" installer even though the "online" installer is recommended by SolarWinds.

Is there any difference between them (what they will install) if I downloaded both and ran the installs immediately (on separate, identical servers) after download?  If the "offline" one behind the "online" one at the moment I download them (like it's a few days behind, takes awhile to add the latest patches, etc.)?

My belief from what I've been reading is the "online" one will perform a check and download *all* the latest of all applicable software (say, if platform Hotfix 6 came out 1 second after I downloaded the installer, I would get HF 6 when I ran the online installer), wheras the "offline" one seems to be locked down to the ?current? software available the moment I download the installer (e.g. I would not get HF 6 if it were released 1 minute after I downloaded the installer) as the offline installer does not even attempt to pull down new software if it can.   Possible issues with the online installer I'm seeing seem to be related to how long the install can take to download everything, esp. if something goes wrong and you have to run the online installer multiple times, but I just wanted to clear up if both options are going to install the same software (at the moment of downloading the installers, assuming you install immediately after download, assuming some patch wasn't made available seconds after downloading them).

0 Kudos

tigger2  wrote:

Curious, whether to use the "online" vs "offline" installer for this upgrade.  It seems the consensus from this thread  (I've tried to catch up on it since the 2018.2 platform came out) is to use the "offline" installer even though the "online" installer is recommended by SolarWinds.

Is there any difference between them (what they will install) if I downloaded both and ran the installs immediately (on separate, identical servers) after download?  If the "offline" one behind the "online" one at the moment I download them (like it's a few days behind, takes awhile to add the latest patches, etc.)?

My belief from what I've been reading is the "online" one will perform a check and download *all* the latest of all applicable software (say, if platform Hotfix 6 came out 1 second after I downloaded the installer, I would get HF 6 when I ran the online installer), wheras the "offline" one seems to be locked down to the ?current? software available the moment I download the installer (e.g. I would not get HF 6 if it were released 1 minute after I downloaded the installer) as the offline installer does not even attempt to pull down new software if it can.   Possible issues with the online installer I'm seeing seem to be related to how long the install can take to download everything, esp. if something goes wrong and you have to run the online installer multiple times, but I just wanted to clear up if both options are going to install the same software (at the moment of downloading the installers, assuming you install immediately after download, assuming some patch wasn't made available seconds after downloading them).

You are correct! The online installer will check with our release system and see if there has been a HF at the time of update, or even a newer version of the installer that could fixes any known bugs. The reason why the online installer is recommended because that is the best way for you to receive all of the bug fixes / patches that we release so the likelihood that you run into issues that some customers report through thwack is much less. It allows us to deliver stable software to you in the fastest way possible.

The two most common complaints that I have received about customers not wanting to use it is

1. my infrastructure to download is too slow and I don't want to watch it download or my internet is locked down.

2. My company has policies that prevent me from going to the latest versions available without doing X things to validate, so I need to constrain it to what is released at the time that I have started the process to upgrade.

If those concerns match your environment, then yes the offline installer would be a better fit. However, the benefits of the online installer are enough for us to make it the recommended and preferred option.

1. much smaller downloadable file (40 MB vs 2-3GB) that you can easily and quickly transport it from server to server

2. updated frequently

3. always provides the latest versions and security patches

4. can handle a mix of requirements in a more resilient way because it is not limited to the set of bits that was packaged at the time of release ( in some rare instances this may mean that the offline installer does not have all the dependencies needed)

5. efficiently only downloads what is necessary versus bits that you may or may not need to be spending time to download, so your overall time to installer is actually more time efficient. When you have larger environments, this small difference can add up to huge gains.

... I could go on but that covers the basics

Hope that helps!

Level 9

Orion Platform 2018.2 Hotfix 5 Updated August 20, 2018 Someones tried it ?

0 Kudos

jerominus  wrote:

Orion Platform 2018.2 Hotfix 5 Updated August 20, 2018 Someones tried it ?

If you update to any of the systems products that have recently been released SolarWinds | SolarWinds Broadens Systems Portfolio with Unified Infrastructure Management Enhancemen...  then this will contain Orion Platform Hotfix 5.

0 Kudos

Hi Serena,

For the link you sent: SolarWinds | SolarWinds Broadens Systems Portfolio with Unified Infrastructure Management Enhancemen...

It seems to include components that haven't been released yet? Specifically SAM 6.7, VMan 8.3 and SRM 6.7? Also is Server Configuration Monitor (SCM 1.0) an add-on like NCM, that will run in addition to NPM?

0 Kudos

prawij  wrote:

Hi Serena,

For the link you sent: SolarWinds | SolarWinds Broadens Systems Portfolio with Unified Infrastructure Management Enhancemen...

It seems to include components that haven't been released yet? Specifically SAM 6.7, VMan 8.3 and SRM 6.7? Also is Server Configuration Monitor (SCM 1.0) an add-on like NCM, that will run in addition to NPM?

If you check in your customer portal you will see the RC available for SAM 6.7, VMAN 8.3, & SRM 6.7.

SCM is a new product that you can evaluate and purchase that would be run in addition to NPM. This is similar to NCM, and if you're interested talk to your account rep for more details.

When did HF 5 come out and what issues does HF 5 address?  I have done HF 3 and HF 4 due to all the odd issues going on since the upgrade. 

0 Kudos

HF5 came out yesterday. Here are the release notes:


Orion Platform 2018.2 Hotfix 5 - SolarWinds Worldwide, LLC. Help and Support

Level 10

Upgraded back in early July.  Noticed after awhile that some of the built-in Solarwinds Pollers are 'missing' from the webpage UI--no results on nodes pages, and not showing up at all anymore in settings -> Manage Pollers.  Biggest change was we added SAM to existing system during upgrade.  Looks like maybe the 'native'/built-in Solarwinds Pollers from NPM were 'removed' during the upgrade?  Can anyone else confirm what built-in pollers they see in the latest version?

Anyone else see anything like this?  Here's a screenshot of ALL pollers I can see grouped by 'Technology':

Pollers3.PNG

For some reason I can only see 39 of them, although I don't know exactly how many I'm supposed to have; the "Switch Stack" poller and others such as: “Virtual Port Channels”, “EnergyWise”, “Fibre Channel”, “Multicast Routing”, “Routing”, “Virtual Routing and Forwarding”, etc. are all missing since upgrading...

Have since re-ran Configuration Wizard with no positive effect. and upgraded to HF4 (and another Config wizard run), again with no change.

Support Ticket #00154983

But they don't seem to have any idea about what could cause such a problem.  We did verify the database pollers table still contains all the poller assignments (even the ones for the 'missing' pollers to the nodes that had them), and no other related errors are apparent in the diagnostics.

Upgrade was from: Orion 2017 HF2, NPM 12.1HF1, NCM 7.6, NTA 4.2.2 and integration with DPA 11.0.387 to:

Orion 2018 HF3 (now HF4), NPM 12.3, NCM 7.8, NTA 4.2.3, SAM 6.6.1 (new addition) and integration with DPA 11.1.x.

0 Kudos

slebbon  wrote:

Upgraded back in early July.  Noticed after awhile that some of the built-in Solarwinds Pollers are 'missing' from the webpage UI--no results on nodes pages, and not showing up at all anymore in settings -> Manage Pollers.  Biggest change was we added SAM to existing system during upgrade.  Looks like maybe the 'native'/built-in Solarwinds Pollers from NPM were 'removed' during the upgrade?  Can anyone else confirm what built-in pollers they see in the latest version?

Anyone else see anything like this?  Here's a screenshot of ALL pollers I can see grouped by 'Technology':

Pollers3.PNG

For some reason I can only see 39 of them, although I don't know exactly how many I'm supposed to have; the "Switch Stack" poller and others such as: “Virtual Port Channels”, “EnergyWise”, “Fibre Channel”, “Multicast Routing”, “Routing”, “Virtual Routing and Forwarding”, etc. are all missing since upgrading...

Have since re-ran Configuration Wizard with no positive effect. and upgraded to HF4 (and another Config wizard run), again with no change.

Support Ticket #00154983

But they don't seem to have any idea about what could cause such a problem.  We did verify the database pollers table still contains all the poller assignments (even the ones for the 'missing' pollers to the nodes that had them), and no other related errors are apparent in the diagnostics.

Upgrade was from: Orion 2017 HF2, NPM 12.1HF1, NCM 7.6, NTA 4.2.2 and integration with DPA 11.0.387 to:

Orion 2018 HF3 (now HF4), NPM 12.3, NCM 7.8, NTA 4.2.3, SAM 6.6.1 (new addition) and integration with DPA 11.1.x.

I see in your referenced support case you sent in the SNMP walk, at this point this will be the information that will help us to determine some more information about the environment.

0 Kudos

45 qty pollers here in recently upgraded 12.3.

hpstech  wrote:

45 qty pollers here in recently upgraded 12.3.

Would you be interested in participating in a dev assisted upgrade when we are running sessions targeted towards Centralized Upgrades?

0 Kudos

Serena -

We already upgraded Production to 12.3. So I guess the next DAU for us would be for 12.4.

Unless of course you'd like to help design our HA initiative for AWS.

Our main focus now is doing HA between on-prem and AWS with our newly acquired HA license.

From the looks of it all, there are many considerations for this idea but its something that's been mandated.