I have asked this question elsewhere and haven't gotten the answers. Perhaps, I asked it the wrong way or forum.
I have a 40+ node cluster and I would like to make sure the data node scanning is aggressively done. I know 8192kb/sec is hard coded but was wondering if there was a better way to keep checking the data node status. Also, is there a document/wiki/blog page which will explain blockScanner report in detail, and what options are needed to tweak them? -- --- Get your facts first, then you can distort them as you please.--
