Improve redis-trib replica assignment

[mattsta]

This tiny bit of code has gone through so many revisions. Hopefully
it's more correct now.

Fixes #2204

[badboy]

It is more correct now and catches a few more edge cases. I think we should go for it.
My hope is that in the long-run we can improve the tools even more (when the first people actually deployed this).
For the kind of perfect distribution of master/slaves we are better off relying on a human that pre-calculates everything instead of trying to support each and every edge case with our code.

[antirez]

Thanks! Merged. I believe for redis-trib wi'll soon start to see good external contributors once a reasonable amount of people deploy Redis Cluster.

[mkleins]

I have always the problem for the replica assignement with 6 instances (3 masters/3 slaves).

Adding replica 10.0.2.42:7001 to 10.0.2.41:7000
Adding replica 10.0.2.41:7001 to 10.0.2.42:7000
Adding replica 10.0.2.43:7001 to 10.0.2.43:7000

The master and the slave of the same node are hosted on the same host.

[sidd081-zz]

@mkleins
I made a small change in redis-trib.rb and it worked fine for me in most of the cases. In def alloc_slots
in script, after masters.each{|m| puts m} add one more new line interleaved.push interleaved.shift

....
579 masters = interleaved.slice!(0, masters_count)
580 nodes_count -= masters.length
581
582 masters.each{|m| puts m}
583 interleaved.push interleaved.shift
....

[alighanem]

@sidd081 It worked for me with this correction. It is necessary to integrate it with the master. thks!!

[sidd081-zz]

thanks @alighanem for trying it out :)