Large log data download - Memory issue🐛

[eliasbruvik]

When downloading a large log, memory issues might cause the API to crash. This mostly happens when downloading several months of log data at a time.

As a quick fix, I suggest to set a limit of how much data you can download at a time. This limit should be dependent on both the size (difference between start and end indexes of the range), and the number of mnemonics you are trying to download. Find a balance where you can download fairly large logs as long as it does not cause the server to crash. As this could run on personal computers and different servers for each company, we should be a little on the safe side in case the system runs on low memory.

Possible solutions for the future if people request downloading even larger files:

• Utilize the Azure integration to continuously upload the files to cloud while fetching the data from the WITSML servers. Then pass a download link from the cloud back to the client.
• Save the data to temporary file(s) on the server and then stream it as a FileStream instead of keeping it in memory.