I've been googling for the last few days to try and find a solution to a problem I'm having on site here. We've got a current branch (was 1610 now 1706) SCCM single-server install, now back when we were on 2012 R2 (and possibly earlier builds of 2016/CB, I wasn't aware of this problem until recently) both PXE booting from legacy systems and UEFI machines was tolerable, neither were what I'd call blistering, but at least it would take <5 minutes to get to the task sequence selection screen.
Now we appear to be in a situation OPPOSITE to that of most google search results, in that my Legacy machines are taking HOURS to boot into WinPE (even boot.sdi is taking minutes to download), whereas my UEFI machines are booting as quick as they ever were.
I've tried messing with TFTP Block Size and Window Size to no avail. I've tried disabling VMQ on all the network adapters (SCCM is installed on a Hyper-V VM inside a 2016 cluster), but it's still garbage.
Anyone got any ideas?
Now we appear to be in a situation OPPOSITE to that of most google search results, in that my Legacy machines are taking HOURS to boot into WinPE (even boot.sdi is taking minutes to download), whereas my UEFI machines are booting as quick as they ever were.
I've tried messing with TFTP Block Size and Window Size to no avail. I've tried disabling VMQ on all the network adapters (SCCM is installed on a Hyper-V VM inside a 2016 cluster), but it's still garbage.
Anyone got any ideas?