We have the following search topology shown below. We shutoff server #12. My first question is how come the "Crawler" is still checked and how come when I try to execute a crawl on content source it gets stuck on starting. Even trying to reset the index it gets stuck. The moment #12 server is turned back on crawling continues.
Shouldn't crawl continue to work because of the redundancy here?
I experienced the same - there is bug showing an incorrect status in the UI. When it tells you "starting" it tries to connect to one server for one minute - if that does not work it starts crawling on the 2nd server - but still shows starting in the UI. The crawl then works as expected.
You can verify this behavior with the ULS log.
this is Crawl components are recognized and failure. default duration of 1 hour.
this setting is here.
HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Office Server\15.0\Search\Global\Gathering Manager DisableInterval
Detected failure and will continue to crawl.
