Stale records

[Am0din]

So how do records that aren't being used anymore fall off the list? Do they go stale after so many days, or something else?

[11notes]

There is no renovate process present. Are you asking that the database should have a cleanup after {n} timeout value? Records that are older than {n} timeframe get removed?

[Am0din]

Yes, there may be computers that are no longer available, have been formatted and redone, etc. Those listed on the KMS main page showing the summary of listed activated KMS licenses should be deleting after so long as they no longer exist.

It was my understanding with KMS licenses, is they contact their activating server from time to time, as they have to not go stale. Is that not the case with them any longer?

[11notes]

This is the case. Question is how to define stale. Any client that has not contacted the KMS longer than KMS_RENEWALINTERVAL is not activated anymore, but would be activated if connected again. So, what defined an inactive client? KMS_RENEWALINTERVAL + a custom timeout? An option in the GUI to remove a client? Both?

[Am0din]

I think the simplest answer would be to just give an option to manually remove stale records, so we don't have to wait on them to 'go stale' and remove themselves, only to not risk that possibility of them returning. This also shouldn't be a huge list for most people using your fantastic project, but some of us with OCD makes us whine because we can't scratch that itch, lol.

I think that if you were to also add a stale record timer, it would be something we would define, i.e., a manual entry timeout in days.

[mcrook250]

Im still testing before I upload my source code and compile, could you look over this and see if this looks good? I looked into it and according to MS KMS, they should drop off from the RENEWALINTERVAL + 30 days as thats the timeout for the os license, for example 180 days and stays valid for another 30. Basically it gets a 30 day grace period.

new features are enabled via environment variables.
Server
-- AUTO_PURGE=True

GUI
-- ENABLE_DEL=True

Please see images below for gui updates.

11notes, if you haven't figured this out for your code, feel free to head over to my repo to copy it over for yours.

[11notes]

I’ll give it a try.

using your fantastic project

Thanks, but don't forget it's a copy of py-kms with some adjustments and changes like IP address and different GUI. Plus, all the people who helped with their inputs. It’s not just me 😉.

[Am0din]

Yeah I'll check it out, probably tomorrow or Monday, for sure.

…

On Sat, Jun 21, 2025, 00:05 mcrook250 ***@***.***> wrote: Im still testing before I upload my source code and compile, could you look over this and see if this looks good? I looked into it and according to MS KMS, they should drop off from the RENEWALINTERVAL + 30 days as thats the timeout for the os license, for example 180 days and stays valid for another 30. Basically it gets a 30 day grace period. new features are enabled via environment variables. Server -- AUTO_PURGE=True GUI -- ENABLE_DEL=True Please see images below for gui updates. kms-auto-purge.png (view on web) <https://github.com/user-attachments/assets/2b011812-e2b7-44e1-8f74-6549362d91ad> kms-delete.png (view on web) <https://github.com/user-attachments/assets/22c4a9c1-40ee-4c21-ba47-4ee83e276486> 11notes, if you haven't figured this out for your code, feel free to head over to my repo to copy it over for yours. — Reply to this email directly, view it on GitHub <#31 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BAXTNDRA6FU6S3ZCTL3EYZL3ETR2ZAVCNFSM6AAAAABZL6PF4OVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTGNJTGU3TSMI> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[11notes]

@Am0din just be careful with the code from @mcrook250, he seems a little off the charts:

Because the upstream loser 11notes likes to leave breaking bugs in his code

[Am0din]

Even if that were the case, that's why I have nightly backups of my VM/LXC environment, lol. But thanks for the warning, I didn't see the post.

…

On Sun, Jun 22, 2025, 00:28 ElevenNotes ***@***.***> wrote: @Am0din <https://github.com/Am0din> just be careful with the code from @mcrook250 <https://github.com/mcrook250>, he seems a little off the charts: Because the upstream loser 11notes likes to leave breaking bugs in his code — Reply to this email directly, view it on GitHub <#31 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BAXTNDTU73CU6YFHLKNEU7T3EY5I7AVCNFSM6AAAAABZL6PF4OVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTGNJUGA4DCOI> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[11notes]

And just that we clear, there is no bug @Am0din. That user could not activate Windows Server 2008 SP2 32bit. I downloded the iso, and could activate this version without any issues. I explained this to the particular user but he simply ignored my explanation. He is now using a different account with a similar name to discredit what I do.

[Am0din]

I think I'll just wait on your development to this stale record update, @11notes. If we're gonna be like that to a developer, then I don't see any reason to support that behavior testing their own code. Find another monkey.