Instead, if the DM accidentally picks a monster that has resistance to all nonmagic weapon damage, it is definitely more difficult to notice if this is wrong...
IIRC, the policy in early 3e was to balance monsters on the assumption that PCs could
not bypass damage reduction or other resistances. IMO, that's the right policy for 5e as well - and that should eliminate this case entirely.
(Later in the lifespan of 3e, I'm pretty sure that policy was reversed, as part of the overall power creep that that edition suffered as it went on. IMO, a bad move on all fronts, though I seem to be in a minority on that one.)
... the DM may just think the monster is fine and only realize after a few rounds that the PCs are not going to make it. Then it will be clearly much less nice to remove the immunity or handwave some discount in the middle of the fight.
I actually wrote a snarky first draft of this reply, because of that single line.
In all honesty, in the case like this, my impulse is to argue that it is the responsibility of the
players to realise they are overmatched and to retreat. And that applies
even if the DM accidentally throws a too-tough encounter at them. Regardless of resistances, or immunities, or the lack of magic weapons, the overpowering encounter is a valid scenario that PCs really should take into account during play.
So, yeah, your suggestion that it was for the DM to realise the problem and to fix it so that PCs don't die got me a little riled. Then I realised I was being unreasonable.
