How to perform simple scanning of files instead of a full scan?

There are too many files, and each scan takes too much time, making it very inefficient. Are there detailed configurations available?

thx

it;s ai trans

If by full scan you mean hashing, then that is only done when a file is either new or it has been modified.

Is there a way to disable the hash verification process for new files? Most of the time, I want the specified folders to be quickly synchronized across devices, and performing a full scan on the files is just too slow.

thank you

Syncthing splits files into blocks, and hashing is an integral part of the process, so no, you cannot disable it.

You probably want to use rsync in your use case. This tool can decide whether to copy a file just looking at the modification dates and file sizes.

1 Like

How do I set it up?

thank you

Sorry, this is a Syncthing forum, so it is out of scope to expect rsync support here. There are a ton of rsync guides and examples out there. As usual, proceed with caution: Try it out on test data first.

Can it be set so that it synchronizes one file or folder as soon as each one is verified?

thank u

Once again, this is not where you get support on rsync.

But let’s go back to what you wrote early on:

To me it is not 100% clear what you think is the problem.

Syncthing performs (at least) two types of scanning:

  1. As has been described already, when a file modification (or a new file) is detected, the file content must be scanned. Not ALL files, but the modified (or new) file. This scanning cannot be avoided.
  2. There is also another type of scanning: A scheduled scanning of all files. For each folder, there is an advanced setting “Full Rescan Interval” with a default value of 3600 seconds. This scan does this: Compare metadata for all files in the file system with metadata in the database. The purpose of this scan is to find changes which might have been missed by the file system watcher, which is not 100% reliable under all circumstances. Note that only metadata is read from the file system, not the files content. (This scanning can be disabled by setting Full Rescan Interval” to 0.)

Which of these scans do you think takes too much time?

For me, both methods are the same, because I need to frequently add and delete files.

tank you

Ok. Use rsync instead then, as suggested by @Martchus. Good luck!

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.