ReFS partition index is forced to rebuild every boot

[roBsonn]

I have two partitions, one formatted as NTFS, other as ReFS (without using Storage Spaces). When I open Everything, the ReFS index is being rebuilt from start, and because it's 2 TB large, it takes a very long time and undesirable amounts of repeated disk IO. This happens every boot.

I had no such problem with NTFS partitions - having installed the Everything service, only the latest changes were updated, which only took seconds. Entire partition was scanned anew ONLY when i forced rebuilt from Options.

Is it normal behaviour? I hope not, as I have the same USN Journal config options enabled for both NTFS and ReFS.

[void]

Could you please send the Everything debug output to support@voidtools.com after restarting your computer:

• In Everything, type in the following search and press ENTER:
  /debug_log=1
• Restart your computer.
• After Everything reindexes your ReFS volume, please send your %TEMP%\Everything Debug Log.txt to support@voidtools.com.

To disable debug logging:

• In Everything, type in the following search and press ENTER:
  /debug_log=0

[roBsonn]

I managed to capture this situation in a log file! Sorry it took so long, it wasn't 100% repeatable at boot (maybe some other factors are in play). It happened two times in a row (after boot, then after an hour later). Here are both logs (I will also send them at supplied e-mail):
[removed links -void]

[void]

Thanks for the debug output.

2017-09-23 11:23:44.199: read NTFS usn journal failed 1181

2017-09-23 11:23:44.199: refs monitor out of date

Both the C: drive and D: drive USN Journals have changed too much since Everything was closed and restarted.

Are you performing a backup to these drives on shutdown or other procedure that may be changing your file systems on shutdown (eg: large windows update)?

Please try increasing your USN journal size:

• In Everything, from the Tools menu, click Options.
• Click the NTFS tab on the left.
• Select your C: drive.
• Change Maximum size to 65536 KB.
• Click the ReFS tab on the left.
• Select your D: drive.
• Change Maximum size to 65536 KB.
• Click OK.

2017-09-23 11:23:44.199: read REFS usn journal failed 87

Please see if Everything 882b helps.

[roBsonn]

Are you performing a backup to these drives on shutdown or other procedure that may be changing your file systems on shutdown (eg: large windows update)?

No.

Please try increasing your USN journal size to 65536 KB.

Did not help, unfortunately. Nor increasing more, up to 500 MB at one point.

Please see if Everything 882b helps.

I'm really sorry, but I read this topic and I don't know what can I relate to. Is there any specific sentence with I fix, that I don't see?

BUT

I am however happy to report I found a fix for this issue. Based on your suggestions that USN Journal is overfilling between Everything's exit and next start, I found my Run at startup option was disabled. This was intended, because I already had Run service at startup enabled and I thought the service manages updating of the database file in the background. Turns out only the main app does it (am I right?), so when now I start Everything app + service at startup, the partition is not being rebuilt, but only updated from USN Journal. I am happy with two instances running in the background, if that's the only disadvantage

Thanks for support, void!

[void]

Turns out only the main app does it (am I right?)

That is correct, the Everything service is rather dumb, as it only allows the Everything search client to read your ReFS volumes USN Journals.
The Everything service does not manage the Everything database.