Uploaded image for project: 'Network Observability'
  1. Network Observability
  2. NETOBSERV-2460

Prevent / warn using Direct mode on large clusters

XMLWordPrintable

    • Quality / Stability / Reliability
    • False
    • Hide

      None

      Show
      None
    • None
    • None
    • None
    • None
    • NetObserv - Sprint 279, NetObserv - Sprint 280
    • None
    • None
    • Hide
      The network observability documentation explicitly recommends not using the Direct deployment model on large clusters, however, this recommendation wasn't visible other than in the documentation. It now generates a warning when Direct mode is used while running on a cluster bigger than 15 nodes.
      Show
      The network observability documentation explicitly recommends not using the Direct deployment model on large clusters, however, this recommendation wasn't visible other than in the documentation. It now generates a warning when Direct mode is used while running on a cluster bigger than 15 nodes.

      Using the Direct mode on large clusters may cause severe incidents/outage as FLP is deployed as a daemonset (including master nodes) and consumes a lot of memory.

      We should be pro-active to detect cluster size and trigger a warning when using Direct mode above 20-25 nodes

              jtakvori Joel Takvorian
              jtakvori Joel Takvorian
              None
              None
              None
              Amogh Rameshappa Devapura Amogh Rameshappa Devapura
              None
              Votes:
              0 Vote for this issue
              Watchers:
              3 Start watching this issue

                Created:
                Updated: