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.

HA Enable On/Off UI Differences

What is the effect of only enabling HA on the /ui/ha/settings page - What is the difference between these two settings? Are they independent of one another? Is HA effectively disabled if BOTH are not in an enabled state?

pastedImage_0.png

But on the /ui/ha/deployment/summary you leave it unchecked - does this mean that HA is enabled for the polling engines out bound polling, but the VHN/VIP is NOT enabled for incoming data?

pastedImage_4.png

  • '/ui/ha/settings' is how you disable HA globally. This is provided as a shortcut for customers that have dozens of HA pools, as it's required to disable when performing upgrades. You can also disable HA at an individual pool level. This is handy when you're troubleshooting an issue with support, need to restart services, etc. but don't want the server to failover.

  • Thanks, one last question, is the cause of my main polling engine being red here due to red criticals in the Deployment health page?

    The server is happy and green in NPM but red on all the HA pages. If its red, will a fail-over occur or is this just cosmetic due to 'perceived' critical health issues?

    pastedImage_0.png

  • That screenshot alone is not going to tell me why the server is in a critical state but I would recommend clicking on the Health tab to find out why.

  • Is there a SWIS verb to disable HA globally? Currently when doing upgrades I collect a list of pools and then use the enable/disable verbs for each.

    I know it's only one setting in the GUI but I have a script I use for upgrades that collects a list of all enabled alerts and then disables them, disables all the pools and then once everything is upgrades re-enables the alerts and turns the HA pools back on. Doing it once makes my script and error handling a lot cleaner.

  • There is currently no SWIS verb to globally disable HA. Currently, these SWIS verbs exist only at the individual pool level.

    pastedImage_0.png

  • My experience has been that if one of the nodes is red a failover event will not go as planned, finding out why they are red is still very much a game of hide and seek and the VAST majority of whats on deployment health is unrelated and like you said, many are red herrings

  • Hide and seek is a beautiful game in our industry. emoticons_happy.png

    "If the AD test found warnings in the logs, view the warnings and determine what, if anything, needs to occur to remove the warning."

  • What is the diag tool checking on this one? I'd like to verify this one manually.

    Check site bindings

    Last checked Feb 2, 2019 11:03:42 PM

    Orion website has no DB entry.

    Please run Config wizard and rebuild the site.

    Related KB articles

    MT7646

  • hpstech  wrote:

    What is the diag tool checking on this one? I'd like to verify this one manually.

    Check site bindings

    Last checked Feb 2, 2019 11:03:42 PM

    Orion website has no DB entry.

    Please run Config wizard and rebuild the site.

    Related KB articles

    MT7646

    It is validating your IIS Site Bindings to ensure that the 'SolarWinds NetPerfMon' website is bound to all adapters and interfaces.

  • Thanks. This is a totally new additional web server recently deployed that runs perfectly. Is this status not what I should be seeing below?

    pastedImage_0.png