Hello @Ankit Awadhiya ,
There is extended documentation explaining transient fault handling.
The default solution is to work with a retry loop together with cool-off time.
If this happens a lot, you probably need to scale up your resources.
It's also valuable to check the query being executed and to find out why it fails. Perhaps the logic has too much freedom (gathering too much data) or the query is just not efficient resulting in extra costs (due to scaling up need).
If the response helped, do "Accept Answer". If it doesn't work, please let us know the progress. All community members with similar issues will benefit by doing so. Your contribution is highly appreciated.