Assuming your log-files of the needed time (~3 months) still exist & weren't deleted due to log storage capacity (log maintenance), then it's fairly easy.
follow sk111766 (https://supportcenter.checkpoint.com/supportcenter/portal?eventSubmit_doGoviewsolutiondetails=&solut...),
and add these lines After stopping the Indexer (evstop) & configuring the no. of days you choose (-days_to_index <90> or beyond) to have it completely re-index with your chosen no. of days.
cp $INDEXERDIR/data/FetchedFiles{,.Orig}
rm -f $INDEXERDIR/data/FetchedFiles
then start it (evstart)
Also make sure to disable/up the daily index files deletion to avoid it from being deleted again.
This will cause a re-Indexing of these last 3 months of logs (or as many days back as you've configured).
which has a performance impact during the re-indexing process which should take roughly several days (depending on your log-rate vs. HW strength).
if you need a better estimation, you can send us your log-rate (or size of log-files) & HW CPU/memory details to better estimate.