2

Even though split brain scenario occurs in both Oracle RAC and Percona's XtraDB Cluster, a two node cluster is allowed and split brain scenario is resolved in RAC but a two node is not recommended in Percona Cluster ( 3 nodes is recommended ). Why is it like that? Doesn't Percona resolve split brain happening in two node cluster? I know I'm comparing two different technologies.. but still if anyone could clear this out.

Thanks!

1
  • You actually answer many questions but you do not upvote the ones which interests you.. Kidding :) will do it henceforth.
    – tesla747
    Aug 13, 2015 at 3:44

1 Answer 1

3

I can see why Oracle RAC can handle split brain and PXC cannot.

What separates them in their architecture and data storage. In what ways ???

Oracle RAC

  • All RAC Instances deal with only one set of database files
  • Each RAC Instance has its own Log Buffer and LGWR (Log Writer) process
  • Writes to any Oracle RAC instance, even on nodes that go down temporarily, can have their log sequence numbers coalesced and serialized

Percona XtraDB Cluster

  • Each PXC node has its own complete set of data
  • Writes to one PXC node must be rolled back or committed on all nodes in unison
  • Writes to PXC nodes that are split in communication will only update/commit to its local copy of the data. Awareness of such writes by other unconnected PXC nodes does not exist when communication has been established again.

According to Percona's Frequently Asked Questions on PXC

Q: How would it handle split brain?

A: It would not handle it. The split brain is hard stop, XtraDB Cluster can’t resolve it. That’s why the minimal recommendation is to have 3 nodes. However there is possibility to allow a node to handle the traffic, option is:

wsrep_provider_options="pc.ignore_sb = yes"

SUMMARY

3
  • so split brain in PXC can be overcome if I use 3 nodes instead of 2 nodes?
    – tesla747
    Aug 13, 2015 at 11:16
  • @tesla747 - Yes. At least 3 nodes gives a "split" situation a quorum wherein the majority decides that it is still the "true" cluster, and the minority understands that it is no longer a valid part of the cluster. Upon reconnection, the nodes will converse, and the minority will be updated with changes that happened during the split.
    – Rick James
    Aug 17, 2015 at 20:46
  • For true HA, the 3 nodes should be in 3 geographically dispersed data centers. (Think, flood, earthquake, tornado, etc.)
    – Rick James
    Aug 17, 2015 at 20:47

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.