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

Would you sit down and chat with a SolarWinds developer? What questions would you have for them?

Loading question...
37 Replies
Level 16

I would show the developers the single pane of glass front end we wrote in house and ask them to develop something similar out of the box as a replacement for the Message Center. 

0 Kudos

I would have some questions around the platform as a whole. I won't post spoilers here, no prep, Mr Developer! 😛

- Jez Marsh
0 Kudos
Level 12

Do you plan a true RBAC design for the whole products ? it would be nice to have something constant from one SolarWinds product to another one. (admin, operator, read-only, none, ...)

For exemple to manage agent we must be administator 🙂

 

 


@MathieuJM wrote:

Do you plan a true RBAC design for the whole products ? it would be nice to have something constant from one SolarWinds product to another one. (admin, operator, read-only, none, ...)

For exemple to manage agent we must be administator 🙂


Plus all of the ones for this question/request!!!!!

Tags (2)
0 Kudos
Level 17

Why do we not have an option to Import Network Atlas Maps into Orion Maps?

When will the Account Limitation Builder be migrated to the Web Console, and work with all Orion applications?

0 Kudos
Level 7

Why can't the software updater do an automatic backup of the database?

Good question @rixter13 !  I think the general thought process is that it might be better to leave that to the professional DBAs.  Yes, not every company has one on staff, but one of the problems I saw was that people would arbitrarily fire a backup job and nuke all the remaining disk space on a server, thereby causing the server to stop presenting data.

The other issue is that the difference between simple, full, and failover backups.  Depending on the configuration of your SQL server, they have very (and I mean very) different outcomes.

Regardless, I think it's worth bringing this up for the simple backup/recovery question.  Keep questions like this coming!

"Shoot for the stars to reach the moon"
0 Kudos

That's the best workaround I've been able to find, but it still feels a bit like cobbling the monitor together. The statistic message value saves our bacon for now.

0 Kudos

I get it, but if we introduce "string matching" then you have all kinds of other issues.  "Starts with," "ends with," "contains," "does not contain", etc.  It's just easier (at the moment) for the authors of the script templates to choose how to work with the variables.  After all 'value != 1" is pretty easy logic.  This is why it's so important to understand the content in a script monitor as well as how to write up the alerts.

"Shoot for the stars to reach the moon"
0 Kudos

Can we hope to ever be able to evaluate non-number results in custom script monitors?

@sturdyerde , for Script returns, I always "alias" an plain text answer to a number (up = 1, etc.), but I'm also big on using the "messages" that you can return.  I've done this for my pi-hole and (at my previous company) include this as part of the alert.

KMSigma_0-1597083097937.png

In the example above, "Status" and "Gravity Enabled" are actually text fields that are returned which I convert to a numeric answer.  I did the same for the "SolarWinds Backup" templates.

"Shoot for the stars to reach the moon"
Level 9

Is there a way to view other internal SolarWinds users that are currently viewing the same Node/Interface?

Having remote workers and similar naming conventions this would validate who is looking at this element and show them you too are reviewing. Also would allow techs to skip over an element if seeing other(s) on it to work another issue. Minimizing redundant efforts.

0 Kudos

I'm looking for

more efficiency v resource. We already have more db & engines than any one of our business apps - perhaps a Raspbery Pi for probes, or even small engines? 

more Rabbit & less MSMQ

fewer SW queries highlighted when I run DPA against the Orion databases

more integration with IPAM, its tables still handle like an add-on

0 Kudos

  1. How much more is going to move into Cortex engine?
  2. What would be the one feature you would implement if you were to be PM for the day?
  3. What is the reason that Dark Theme has taken so long to implement 🤣
0 Kudos
Level 12

 

Yes but...

 

I would like to get the upgrade to v2020.2 first.

 

However a before and after may also be good to have.

0 Kudos
Level 7

I already have meetings scheduled.

0 Kudos

One thing that is currently interesting for me is that a lot of the Alerts I have configured are SWQL based to the complex nature of the event or because a certain field within the DB is being targeted. 

Will there be further improvements in the following:

1) Making more complex alert commands easier to be applied via the web console / GUI

2) Upgrades on the SWQL Product line to enhance the current feature set and adding additional search functionality.... I.E, I know the value due to an alert, but I do not know the DB table it is stored in... Like reverse searches

0 Kudos

How do you assess if you are taking advantage of all the feature available in any product?  

In Virtualization Manger Version 2019.4, is there a way to set a dashboard similar to the appliance?

I really liked that dashboard, would you believe that I am still running the old appliance along with the new version!

 

zennifer_0-1596834671882.png

And, I want to change my maybe to a yes @yumdarling  - thanks - Zennifer

0 Kudos
Level 11

Not really sure that some these are to be directed at the developer, but rather at the development/product team.

Here goes:

- Why is it that you cannot assign your own API Poller templates to nodes (rather it is just out-of-the box templates)?

- Could we please make error messages more intuitive/informative? Today I got something along the lines: an error has occurred and the detailed description was that one or more errors has occurred. Or "Object referent is not set to an instance of an object" and then the error report contains absolutely no useful information (IDs, queries, source etc) 

- I love the look of the new PerfStack inspired graphs, but is there any way to make them more snappy? If you put more of them on the same page you end up with spinning cubes for a good chunk of time before any data appears.

- I like the versatility of the Custom table widget, but seeing the data source name (for [Data_Source_Name]) displayed on top of the table can be a bit annoying; if that would not show up the custom tables would be practically indistinguishable from  the out-of-the box resources.

- Kudos for the new 'modern' dashboards. Any plans to integrate the new widget stiles into 'classic' dashboards? Will they both converge to a hybrid version in future releases?

- When the new 'Manage Entities' page came out I was very excited but then I realized that although it had some upside (multiple filters, modern look) it also had quite a few drawbacks that have kept me from using it instead of 'Managed Nodes' to this day (lack certain options from the more actions menu, no way to select the columns to be displayed, spinning cubes 🙂 ). Are there any plans to address some of these?

- APIs are all around us these days and have become an absolute must have. Are there any plans to add the possibility to auto-discover nodes based on API queries (i.e. mass GET from CMDB)?

I'm sure that if I thought about it I could add a few times as many questions/suggestions to the list. 

Bottom line is that you have a great product suite and an awesome all-round approach, but some of the little things can annoying sometimes.

Cheers!

LE: 

- Why is it that you cannot select custom SQL/SWQL query as a data source/entity type for a PerfStack graph. I could think of tons of use cases where this could come in handy. Right now I can run a custom SWQL query through an API poller that passes the call to an API Gateway that deals with SSL validation (pre 2020.2; I know that API pollers can do that out of the box now, thanks!) and add that as a data source in PerfStack.

- When is dark mode coming to Orion? That seems to be the hype these days when it come to 'new' features 🙂

0 Kudos
Level 10

I would like a way to easily strip the domain information off the node name. I have to manually do that for all nodes, and it is time consuming. What I mean is changing something like ibmnode1.ibm.com to ibmnode1, for all the nodes at the same time. It makes the maps cleaner if the node names are short. Thanks!