Power Systems 2009 – PowerHA Networking

As usual for those blogentries, they will not describe the topic in detail, but more in a punctual way.

HACMP networking and some best practices might be mentioned here 🙂

  • HACMP or PowerHA has a better smitty implementation (complete re-order of the menue)
  • Be aware of the DNS settings on the lpar. It should be : local,DNS,NIS
  • PowerHA can have up to 64 resourcegroups, each RG with many ip addresses (256 in total per cluster)

PowerHA fastfailover

=> set failure detection time = FAST

lssrc -ls topsvcs

Fast: sensetivity = 5 missed heartbeats

Normal: sensetivity = 10 missed heartbeats

IPAT

  • has a big disadvantage => ARP
    • after change, IP is still known via old MAC-Address to the switches
    • as of this, a GARP is send out to the network making the announcement that the MAC changed
      • broadcast messages have to be allowed for GARP to be working

IP collocation
collocation with persistent ip address

  • set all service_labels on NIC with persistent address

IPAT via replacement

  • max 1 service label per RG
  • ARP solution: HWAT

Heartbeat over IP aliasing

  • boot and service ip labels in the same network => very tricky and complex
  • alternate address will be placed on boot addresses with increments on other node

Gigabit fast failover on etherchannels

  • 2 dual etherchannel environment the failover takes < 1ms

With fixpack 1 on PowerHA 5.4 netmon.cf changed to the format “!REGD ipaddress”

Advertisements
This entry was posted in aix, HACMP and tagged . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

w

Connecting to %s