So I tried Windows Loader and it successfully activated the PRO version I just installed. If you need any further information from me to troubleshoot please let me know. If you do get a fix for me to try I will get on it ASAP. -thx
Out of curiosity, since Windows Loader is not supposed to work on GPT format, and the process did work as you suggested. Did Windows Loader just do what MRP was supposed to do? i.e. if I understand correctly Windows Loader injects a script that runs each time windows boots, but in my case, this script injection did not happen because I have a GPT format.
Windows Loader works on GPT, ONLY when a valid slic is found. On GPT it won't install the bootloader emulation, only the oem:slp key and the corresponding certificate. And yes, that's what MRP is supposed to do too.
I edited a m/board to match your HP dmi data... Also added the HP SLIC... My debug results: Code: ==================== DMI Processing ==================== CSNAME : DMI + Database = Possible Match [HP] CSNAME {Conf} : DMI + Database = Match [HP] - via Conflict Resolution. DMI conflicts was detected. DMI query routine has completed. Checking any SLIC Data. HP [#08] SLIC v2.1 found for possible Windows 7 OEM activation. Bypassed Virtual Machine data checks as DMI [HP] was detected. HP [#01] manufacturer was detected for automated theme/branding. Checking database for OEM-SLP Product Key HP OEM-SLP Product Key was found and applied. Check for OEM XRM-MS Certificate. XRM-MS OEM Certificate will be applied. Which is correct... The only differences was that the board i used did not have UEFI bios or GPT HDD format and so only MBR was available, also no NVMe slot/capability. Not that those should of made any change to the results... I have now added more error checking in that area and the next MRP should show a code if it fails to find a SLP-Key etc for whatever reason caused the failure. I have had a friend who has a HP desktop to do a test w7 install and the results was activated, like my test was, however his newer desktop had UEFI/GPT but no NVMe. I doubt the NVMe drive could cause the strange glitch. Therefore i can only say it is one of those weird HP/Compaq scenarios where a combination of factors caused the project to fail activating Win 7 via the usual SLIC+Cert+Key methods. That manufacturer has always been a thorn in the side of MRP, with Lenovo being a close 2nd!
Appreciate the follow-up! So then could you simply have MRP run the same code that Windows Loader runs to insert the key+cert ? Also, is there any reason this couldn't be run on-demand by the user after Windows is installed?
Technically MRP will 99.9% of the time insert the key+certificate which matches the SLIC within the bios. The Loader i believe is written in c# or c++ which i am not familiar with, but MRP does do a similar job, but for some reason in your case it failed. The code within the project is quite complex because of all the changes over time, sadly there will always be something new that trips up the project, i try to eliminate as many bugs as possible. The project is designed to be only run during windows setup. I have not changed the process to allow manual activation etc later as to be honest there are other methods such as the loader which can insert the key+cert. At some time in the future i will be working on a 'after install branding' which may add SLIC activation but due to personal reasons i not have the time to do this yet.
I understand. Was just curious. I'm all set up on my end now so wasn't looking for a new solution from MRP. Good luck whenever you decide to undertake the after install branding project!
HP/Compaq has always been an adjustment from time to time. You would think they would use one damn name and stick to it. lol
MRP 117 is now in RC1 stage Mainly a 'bug' fix release for Win7/Vista/Servers for the SLP-Key/Cert routines which has new error checking. Also a W10 {19H1+} only option which is under testing... Disable 'Activity History' data collection, this should stop any of the user's activities being 'recorded' and potentially sent to m$ etc. Some of the MRP's internal code has been updated which should help to be more efficient in operation.
Re-checking MRP code to see why it is failing.... *EDIT* 7+ hours later Hopefully have now sorted the key/cert glitch. For some weird reason a variable was null'ed, (set as nothing), when it should of contained the 'Edition' name, took a few hours to sort it and several failed installs. It was not just on HP/Compaq it also could affect other OEM brands. Have now done a few installs on different machines i have - that have a SLIC 2.1 and now they all install the required OEM-SLP, (which can be a default one as they are not brand specific, such as Gigabyte which does not have a Professional SLP-Key,) also the Certificate routine is now used as it was bypassed before if the 'pidkey' or the 'Edition' variables was not set. This did not affect win 8.x/10 as they not use OEM-SLP/Cert's.