SharePoint Search Service Instance Disables itself

Question

I have a small SharePoint 2019 farm (1 WFE/Cache + 1 App/Search + 1 SQL). We've been running Search with no issues for a few years but earlier this year we started having issues with the Full Crawl (which runs on Saturday evenings) getting "stuck". It would still be "running" when I come in the office on Monday so I would have to cancel it. I found that one of the Search Service instances would go "Disabled". Starting the instance back up, restarting the search server and then starting the crawl again would resolve it. It only happened occasionally at first but then over the last 6 months it started happening every weekend. We noticed the CPU and Memory was consumed during full crawl. We allocated more CPU and memory and that seemed to resolve it but then found that it made performance for end-users terrible. SharePoint was unusable during the Incremental crawls (runs every 15 min). I then read up on "Crawler impact rules" and so created a crawler impact rule to limit to 4 documents per request. This helped dramatically. Crawls run great now. User performance is great. Except, I noticed last weekend the "Search Service Instance" went disabled again.

We have two instances. Is this normal with a single Search server with a single Index and a single crawl database? Is one of these uneeded and that's why it keeps going disabled? If it's normal to have two, do you know what I should look for in the ULS logs? I'm not finding anything in particular. Anything in particular I should enable verbose logging on?

Answer 1

Hi @Dominique Graves,

It is not normal for a Search Service Instance to go disabled frequently. It could be due to various reasons such as insufficient resources, configuration issues, or other underlying problems. Since you have already allocated more CPU and memory and created a crawler impact rule, it is recommended to check the ULS logs for any errors or warnings related to the Search Service Instance. You can also enable verbose logging to get more detailed information. As for having two instances, it is not necessary for a single Search server with a single index and a single crawl database. You can try disabling one instance and see if it resolves the issue. If the issue persists, it is recommended to further investigate the root cause.

---

If the answer is helpful, please click "Accept Answer" and kindly upvote it. If you have extra questions about this answer, please click "Comment".

Note: Please follow the steps in our documentation to enable e-mail notifications if you want to receive the related email notification for this thread.