Is NAM available for Dynatrace Managed?

[AntonioSousa]

NAM has been launched. It replaces, although not said, the port & ping monitoring that existed before. It's also an implication of the move away from EF1.

The documentation talks a lot about the latest Dynatrace way to do it, but does not talk about Managed: https://docs.dynatrace.com/docs/platform-modules/digital-experience/synthetic-monitoring/network-ava...

I have a very big implementation of port monitoring on a Managed environment. 1.296 for Managed will start appearing next week. What should we expect?

[PedroDeodato]

As the Documentation is now split between SaaS docs and Managed docs , I found the same page mentioned above, but on its Managed version:

https://docs.dynatrace.com/managed/shortlink/network-availability-monitoring

The funny thing is: it seems to be an exact copy of the first one, as although being a Managed page, it mentions "Latest Dynatrace"

The documentation on that page seems a bit confusing (some of its links lead you to pages talking about "apps", but on managed??)...

So, I'm not 100% confident that NAM will be available on Managed or not...

Someone from Dynatrace should clarify if it will, as well as check the validity of these "Latest Dynatrace" mentions on the Dynatrace Managed docs.

Thanks 🙂

[Magda_Zarek]

Hi,

Thanks for your feedback 🙂 

This should not be there. We noticed this earlier as well and are issuing the fix.

As Jacek mentioned, we're working on a separate documentation for NAM for Managed. 

[Jacek_Janowicz]

Hi @AntonioSousa ,

Thank you for raising that question. I’m happy to assure you that we also offer Network Application Monitoring (NAM) for Managed customers. All three types of NAM monitors are available for Managed customers, although there are some differences:

1. The latest Dynatrace approach for handling NAM with the new Synthetic app is not available for Managed environments.
2. NAM Configuration and Reporting for Managed :
   • Configuration of NAM monitors is available through the API only for Managed and as an add-on option for the latest Dynatrace.
   • Reporting for NAM monitors relies on metrics produced by NAM. Additionally, we’ve added basic out-of-the-box (OOB) reporting based on the Unified Analysis screen.
   • Alerting also works the same way for both, Managed and SaaS.

Our team is already working on more detailed documentation for this aspect. Expect additional chapters of documentation to be available soon. I’ll also ping you within this thread to keep you informed.

Additionally, I wanted to inform you that next week we will be releasing a blog post about NAM. This post will include a dedicated section specifically focused on Managed customers.

Feel free to reach out if you have any further questions!

Best Regards,

Jacek

[AntonioSousa]

I'll have to wait for more information and see exactly what comes out with Managed. One more example of the bad things that came along with the EF1 -> EF2 migration.

Also, there are some unbelievable things, like using different metrics for SaaS / Managed, the nonexistence of UI configuration, documentation completely done wrong, ...

[jegron]

Hi @AntonioSousa !

For your information Dynatrace provide a migrator script  :

• https://github.com/dynatrace-extensions/dt-extensions-nam-migrator

[AntonioSousa]

@jegron,

BTW, where did you find that link? It might be a precious help, but not sure how it's supported, and if I can use it for the volume I have!

[Jakub_Surma]

Hello @AntonioSousa,
The script uses DT API that has a paging mechanism implemented, so in theory it should handle high numbers well.
The command get will convert the extension endpoints to monitor definitions one-by-one and save them locally. It will not modify the environment in any way, only the command post does. The worst case scenario is that it will take very long for a high number of endpoints.
Having said that, the script wasn't tested for huge loads and maybe there is some throttling on the DT side if there are too many requests done in short amount of time.

I'd recommend to run the get command and look if the number of definitions is correct.
Please feel free to reach out to me for any questions about the script or any problems you run into.

[jegron]

@AntonioSousa : We are working with Jacek since several month for a customer as Early adopter 🙂

For your information we did not test the script because we started directly with the new Synthetic.

[rohithsamudrala]

Is there any page which describes the scripts/ tools that Dynatrace is making available for their customers? or do we have just keep and eye on the github 

[AntonPineiro]

Hi,

You can check:

• New documentation entries.
• Release notes.

Best regards

[Jacek_Janowicz]

Hi all,

Thanks for the discussion and feedback shared already.

I want to provide some updates here:

We have released a blogpost about NAM, which, among other things contains:

• Section focused on offer for Managed users
• Link to script for converting extensions NAM monitors

I also know from good sources, that Managed documentation of NAM is about to be refreshed.

Additionally, I'd like to let you know that we're also exploring options of delivering basic option of using Unified Analysis based UI for creating NAM monitors. So far, I can not give you ETA for it or even promise that it is doable.
But let mu use that discussion as an opportunity to ask the question: do you see it as a good approach from your perspective?

Best Regards,

Jacek

[AntonioSousa]

@Jacek_Janowicz,

Quite frankly, when I read the first paragraph of the blog post, I was disappointed. For anyone that reads that first paragraph, it seems that the Dynatrace platform is doing this for the first time? I have been doing ICMP latency detailed analysis with Dynatrace for more than 3 years now, and the official Dynatrace port&ping extensions have been out there for some time, too! And yes, they were a good solution, you could do it by API (I was doing it), only the cost model was wrong... Quite frankly again, this is the change that makes more sense to me!

Unified Analysis seems good to me, but it won't be a good option for Managed, because Synthetics have a view that is not aligned with Unified Analysis. So, it will be another hack... In SaaS, let's see.

Regarding other aspects, I'll have to check my wish list better. But it seems that the one on the top has not been contemplated; I'll put it as a Product Idea next.

[Jacek_Janowicz]

Hi Antonio,

Thank you for your feedback. I truly appreciate it and consider your comments valuable guidance for our next steps.

I also wanted to share some updates with you:

Regarding the product idea you mentioned, I have already shared it with my colleagues in the infrastructure monitoring domain. You should expect an update from them soon.

As for the Unified Analysis view being referred to as a “hack,” I can comprehend why you described it as such. I hadn’t mentioned it earlier, but the link to the UA list is available on the synthetic screen (only for Managed).

NAM is critical for a lot of our customers hence it was imperative that we have a solution for both SaaS and managed customers. Creating NAM monitors on both, classic and the new Dyntrace via APIs is quite a powerful solution that gives our users a lot of control to monitor the availability of their network devices.

Given Dynatrace’s direction to facilitate migration to Grail and the new Dynatrace UI, it was imperative to incorporate network availability monitors and their results in the new interface (whereas a similar view to analyze results already exists in Dynatrace classic as Unified Analysis).

Best Regards,

Jacek

[tim_gerlach]

At first I was excited to see NAM coming to Dynatrace Managed. However, the user experience is absolutely unacceptable for a commercial product. I really don't understand why the creation of NAMs has not been integrated with the Synthetic screens. For Managed customers, the product is getting worse and every update makes the product feel more like a neglected solution with some screens stitched together.

[Jacek_Janowicz · Accepted solution]

Hi @tim_gerlach 

Thanks for reaching out and raising that important point

I do understand the concern you are sharing.

The fact is that modern availability & performance monitoring and reporting requires enriched data and lengthier retention which is not possible without a data lake like Grail; thus, we can't offer certain enhancements and innovations for Managed.

Luckily, NAM is an exception to that rule. We realize that functionality is critical for all our customers, including Managed. Thus, NAM is also available for Managed.
We also hear your feedback that an API-only approach to configuration is insufficient, and we want to react to it immediately. Thus, I’m happy to share that we decided to build a simple UI for configuring  NAM monitors on Managed. It will be available in version 301. Let me share a couple of screenshots with you:

 

 

I realize that the user experience is not the same as for SaaS customers and is a bit different from what is offered for handling HTTP and Browser monitors in Managed environments. But our main goal was to react to collected feedback swiftly. So, recognizing that a UI tool for managing NAM tests, complementary to API, is needed urgently, we prioritized this simplified solution, as it can be delivered quickly without leaving our users with an API-only option for a longer period. 

Best Regards,

Jacek

[kalle_lahtinen]

Hi,

Has anyone successfully setup NAM monitors for Managed? I tested this today, and posted the monitor via API, assigning it to a synthetic-enabled Cluster AG. But nothing is executed. When polling the private locations via API, I see that the capability for MULTI_PROTOCOL is not listed:

"capabilities": [
"BROWSER",
"HTTP",
"ADVANCED_HTTP"
],

Could that be the issue, or something else...? The AG version is new enough that it should be supported.

[Jacek_Janowicz]

Hi @kalle_lahtinen 
Yes, we do have customers using NAM from Managed. Just to make sure: Is your ActiveGate 295+ ?

Best Regards,

Jacek

[kalle_lahtinen]

Hi @Jacek_Janowicz 

Yes the AG version is new enough, like I mentioned. I already got a response from support that this is a bug, and "will most likely be fixed in version 1.300, but it's not a final decision".

So it's due to the available capabilities I listed above, here's support's response:

"Your assumption seems to be correct. The lack of the mentioned capability can cause the monitor not to run.
This capability should be added automatically with the cluster and AG update."

Seems to affect Cluster AGs at least, I haven't tested with Env. AG yet because I will then need to change the AG role from extensions 2.0 to synthetic-only, which means that the old monitoring can not be executed anymore - I don't want to take the risk that it won't work, which is why I'm testing with Cluster AG first.

[kalle_lahtinen]

Support implemented the fix manually for our cluster, and the monitoring works now. FYI the list of capabilities for the synthetic AG you're planning to use should look like this:

"capabilities": [
"TCP",
"BROWSER",
"DNS",
"HTTP",
"ADVANCED_HTTP",
"ICMP"
],

[Jacek_Janowicz]

Thanks for the confirmation.
And sorry for the trouble. Good news is that the fix has been already delivered by development team, so, soon the functionality will be available without need to involve support.

Best Regards,

Jacek