I’ve got a few decently-specced Synology servers out there that seem to be constantly running with nearly 100% CPU usage caused by synoelasticd. This task is the main process behind Synology Universal Search.
From what I’ve read, the issues often occur when you have a large number of files stored on your NAS (this one in particular has around 4 – 5 million files) and there are a large number of results configured to be cached.
To fix this, ssh into the NAS and then edit the file var/packages/SynoFinder/etc/settings
In this file, find the line with max_search_hit_count and change the value from the default down to something considerably lower. In my case, I changed it from 100,000 down to 10,000 and this seems to have fixed the issue.
Once you’ve saved the changes, reload the config via /var/packages/SynoFinder/target/tool/synoelastic -a reload_settings or reboot your NAS.
Give it another 5-10 minutes to see if the CPU usage has returned to an acceptable level.