Looking for patch to fix "High hopcount" problem addressed in CTDB 2.4

[Martin Schwenke]

On Wed, 21 May 2014 18:16:26 +0000, Kevin Osborn
<kosborn at overlandstorage.com> wrote:

> One of our customers has experienced the "High hopcount" issue while
> running CTDB 2.2. I can see from the release notes that this issue
> has been fixed in version 2.4. We will have a new release of our
> cluster software in a few months that will incorporate CTDB 2.5.1
> (or later) but I am wondering if I might be able to devise a patch
> fix for CTDB 2.2 in the interim. Can you direct me to a patch or the
> section of code that addresses the "High hopcount" issue?

If you mean the bug where hop count for a record increased forever
then I think it was these 2 patches in the CTDB master branch:

  a610bc3 vacuuming: Fix vacuuming bug where requests keep bouncing between nodes (part 1)
  http://git.samba.org/?p=ctdb.git;a=commitdiff;h=a610bc351f0754c84c78c27d02f9a695e60c5b0f

  f085301 vacuuming: Fix vacuuming bug where requests keep bouncing between nodes (part 2)
  http://git.samba.org/?p=ctdb.git;a=commitdiff;h=f0853013655ac3bedf1b793de128fb679c6db6c6

If they're seeing lots of high hopcount messages (without an endless
spiral upwards) for different records then this might just be due to
contention.

peace & happiness,
martin