Last week I talk about a way to resolve endless crawl. Today I’ll talk about one of the possible cause.
In this case, the search configuration was to launch a full crawl every Saturday at 2AM.
Each time the full crawl was running endlessly while launching it manually was always a success.
I started to look on the error received during the crawl and saw nothing related (just some error on some pdf files indexation).
My next though was to think about a possible recycle of the application pool. I thought it was plan a 4AM so even if it causes some issue, some items should have been processed in the 2 hours between the start of the crawl and the recycle.
Looking at the configuration, I saw that the settings was set to 2AM too!
I switched the recycle time to 1AM and now everything is working fine.
Hoping this helps!