>Hi all,
>
>I have been reading a lot of the posts on SMB2/3 settings until I can read no more...want to summarize what I think others are doing - with a couple questions.
>
>1) it seems that this arises in heavy use situations and typically manifests itself as corrupted indexes (although this article [ http://fox.wikis.com/wc.dll?Wiki~Windows10MultiUserIssue-CorruptionofCDXsandDatarecords~WIN_COM_API ] also pointed to missing child records of a form which is what my symptom is i.e. I have not had index corruption)
>
>2) the best solution seems to be do add some registry settings to each client PC (published in threads here and via Alaska software)
>
>Questions:
>
>- are these settings necessary on the server itself (some seem to do so, others do not)

No, UNLESS:

- The server is a Remote Desktop Services (Terminal Server) host running sessions within which your VFP app runs, or
- You run your VFP app directly on the server console

It's worth pointing out that, as I understand it, the registry changes must be applied to ALL workstations/sessions running your app; if you miss any the problem could recur. Also as I understand it, making those changes on a server that doesn't otherwise need them will NOT "fix" the issue on workstations where it has not been applied.

>- looking at the registry setting names, it seems they might help with my issue because it seems the settings would force the OS to re-read the files every time. Any opinions on this from others?

I don't think there's any value trying to read anything into the names of registry keys/values. Best to look up documentation for them if you're so inclined.

>(to reiterate my issue, a user saves a parent record and some child records; when they go back into it a few minutes later, the child records are missing - which is I think maybe a data lag showing up as an outdated index which then shows itself up as the child records not being found)
>
>I know I posted stuff on this last year and the year before but some coding changes (FLUSH...FORCE) seemed to fix it but now a user is reporting it again.

I remember seeing a post about a networking-related Group Policy setting whose default value was apparently changed in Windows 8.1 and later. Changing this has helped some people: Message#1642076

Do I recall correctly that you're running this app/seeing this problem in only one location/environment? One thing I've been saying for years is that multi-user VFP apps using native VFP tables is a stern test of the network environment - it only works properly if everything is exactly correct. If anything is wrong it often first manifests as an issue with a VFP app.

If I were you I'd apply the registry changes to the workstations/sessions and test with that. If the problem persists I'd try to get more sets of eyeballs checking everything in your environment.

For my own reference:

- SMB2+ summary: Message#1620274