This discussion has been locked. The information referenced herein may be inaccurate due to age, software updates, or external references.
You can no longer post new replies to this discussion. If you have a similar question you can start a new discussion in this forum.

Uncovering Polling Problems and Issues in Your Environment

< BACK TO: The Magic Of SQL Scripts: Do What Counts, Don't Count What You Do

These 4 reports are extremely helpful and are those that I would always make sure get sorted first in any environment at any client site. Reports are built to highlight any issues with polling any objects in your environment that are not setup correctly and potentially missing on alerts, etc. You may well be surprised how many issues you don't even know are hidden from your vigilant eye. Just run them and share your thoughts here. I hope this will help to keep your environment and health monitoring more trustworthy and consistent. It is my wish that checks like that are available by default, out of the box

End Result:

4 mini on-page reports to highlight all polling issues in the environment

SnipImage.JPG

  • Components - will flag any components in Unknown state. There is an additional logic to exclude disabled and deleted components. It will also flag Down components as well, but you can remove this filter (read notes below)
  • Interfaces - will check to make sure interfaces are being polled. It will also check bandwidth is set. If bandwidth values are missing - your Utilisation alerts will not work. Have a go and run it. Then test your Utilisation alerts. If it is based on % usage - you are doomed until this is resolved. Easy way to resolve is just to set custom bandwidth
  • Nodes - will check polling from different angles, including the one recommended by adatole‌ through checking CPU utilization graph and making sure CPU stats are collected
  • Volumes - same as with node, will check volumes polling from different angles, including making sure that stats are being collected

Step-by-step:

(1)

Download reports attached and copy them over to your reports folder. Default location is C:\Program Files (x86)\SolarWinds\Orion\Reports

(2)

Open your Report-Writer App - these report will automatically be added there and will sit under "-THWACK" group:

001.PNG

(3)

Open them all and all you need to do is to change Database Name in SQL scripts to whatever name you have in your environment. It is used multiple times - make sure you change them all

example:

002.PNG

(4)

Now you can add them as resource on your desired page. Use resource called "Report from Orion Report Writer"

SnipImage1.JPG

Additional Notes:

  • With Polling Problems for Components I have chosen to report also on "Down" status, simply because in  my environment this status is never part of the workflow. I would roll all components to either Critical or Warning, but not Down. In case it is down - all it means to me - something is not working on the polling side. Not necessarily problem with SolarWinds though, might be poorly designed scripts, etc. If in your environment Down is OK - then simply remove one line in SQL script which does that check
  • Alternatively, simply use SQL scripts to create your web-based reports. I kind of like them to sit on my dedicated view instead

Thank you,

Alex Soul

attachments.zip
Parents Reply Children
No Data