Here is a firmware that should take the Archer A7 out of the "not able to downgrade state":

https://static.tp-link.com/beta/2021/202104/20210414/a7v5_us-up-ver1-1-2-P1[20210413-rel54475]_to_1.0.2.zip

Once you do upgrade to the intermediate version you should then be able to downgrade to the version where you get the USB print server functionality back:

https://static.tp-link.com/2019/201904/20190417/Archer%20A7(EU)_V5_190403.zip

Thanks everybody, while I was working with Support I did not check this forum for awhile.

It turns out that Support offered me a novel solution.  It is exactly what Tony said in his update just above.

I did not check to see whether the links or their contents are the same, but the idea for sure is.

1. Apply an update to allow older updates to be made; then 2. Apply the older update to restore the function.

I asked Support several questions about this.  They did not fully answer.  I'm not sure whether they did not know the answers, did not want to tell me, or didn't understand the questions.

But one thing seems clear enough.  With this workaround, I would not be able to upgrade any more, and if an important fix came along (I imagine they're all important), I would not be able to apply it without losing the print function again.

I suggested to them, and I'll mention it here too (Tony?), that they should make the special update (the one that allows downgrades) "ageless."  That is, it should have an internal date of year 9999 or equivalent, so that it can be applied anytime, and at any then-current level, whenever someone wants to restore their print function.

As it is now, it represents the most current upgrade available, so it could be applied on top of any level now in the field.  But I suspect if I saved this for a rainy day, then a year from now tp-link delivered a new upgrade, this special update would fail, since it would be older than any update which might come in the future.

In the end, I want to keep my router current, so I thanked Support for their help and informed them that I plan to purchase a standalone print server device.

(Of course they were not willing to absorb the cost of this.  No surprise, but it doesn't hurt to ask!)

We'll never know exactly why they "had" to remove the existing function, but I'm sure they did the math and though it is probably not the case that nobody is using the function, it likely is the case that their business will not suffer from taking it away.  Whatever potential new revenue stream or whatever liability is saved by the functional or security updates they plan in the future that needs all the space they can reclaim, it's probably more than the hypothetical lost business that will result.

There are still a couple things that bother me, which I've let slide once I knew the print function was removed.  I wish my memory was better, and I admit I could be mistaken, but I could swear that:

1. The problem did NOT happen on both computers at the same time.

2. In fact, the problem didn't occur like a switch was flipped.  The function seemed to degrade over a period of a few days or a couple of weeks, where certain aspects failed intermittently in an unpredictable way until it was clear that it didn't work at all.

3. I know everyone says upgrades do not happen automatically; I have to explicitly download them.  But I do not remember doing this anytime near when the problem first appeared.  I think it did happen automatically.

4. Not all possible update levels are available on tp-link's website.  They only list three.  The one which restores the print function is not even one of the three.  (Of course I do not expect the special one to allow downgrades to be there.)

Thank you all for your help and insights.

@Coach-D 

Well, I suspect what the private beta does is use a version number once installed as an older date of any that exist? When you go to install it, it represents itself as a newer f/w than any produced?

When you installed it, the beta, the GUI of the router should have shown the f/w date at the bottom. I guess you didn't notice that?

It might not be the date either, could be the f/w version. From the filename it isn't clear the date is the key?

Coach-D wrote

But one thing seems clear enough.  With this workaround, I would not be able to upgrade any more, and if an important fix came along (I imagine they're all important), I would not be able to apply it without losing the print function again.

Yes, and it is more than likely to be a SECURITY fix, that is not applied leaves your router exposed. This was the reason that I ditched my Netgear R8000. Netgear removed function I required. TP-Link seems to not issue f/w updates that often? Compared to Netgear at least? They seemed to do it every few months. The README's always had some operational fix in the f/w and a line stating 'Security fixes'. 

If you look at https://routersecurity.org/bugs.php, you'll see a LOT of Netgear routers are listed. There is a Dec. 2020 item for TP-Link listed. It is for these routers::

====================

A password-disclosure issue in the web interface on certain TP-Link devices allows a remote attacker to get full administrative access to the web panel. This affects WA901ND devices before 3.16.9(201211) beta, and Archer C5, Archer C7, MR3420, MR6400, WA701ND, WA801ND, WDR3500, WDR3600, WE843N, WR1043ND, WR1045ND, WR740N, WR741ND, WR749N, WR802N, WR840N, WR841HP, WR841N, WR842N, WR842ND, WR845N, WR940N, WR941HP, WR945N, WR949N, and WRD4300 devices.

====================

I don't have that one, but I'll assume that TP-Link has released new f/w (usually before this is made public possibly) to correct the f/w flaw.

It is flaws like that you'll miss correcting if you can't upgrade, more liable to happen than an 'operational' fix or added new feature.

Why was the function removed? Doubt they would tell you? Like I said before, either the code expanded due to fixes and it no longer would fit in RAM or a NEW feature added was doing that and something had to go? Of course, the code itself for the Print Server could have been 'licensed' and they didn't renew the license? Only TP-Link would know for sure. Matters not, they removed it, and did work so you could back-level, I'm sure if it was just removed for no reason, the code could have been put back.

Coach-D wrote

 

There are still a couple things that bother me, which I've let slide once I knew the print function was removed.  I wish my memory was better, and I admit I could be mistaken, but I could swear that:

 

1. The problem did NOT happen on both computers at the same time.

 

2. In fact, the problem didn't occur like a switch was flipped.  The function seemed to degrade over a period of a few days or a couple of weeks, where certain aspects failed intermittently in an unpredictable way until it was clear that it didn't work at all.

 

3. I know everyone says upgrades do not happen automatically; I have to explicitly download them.  But I do not remember doing this anytime near when the problem first appeared.  I think it did happen automatically.

You might not have noticed it on both computer's. It might have been some time between trying to print on each computer?

I say that because if the code was removed from the Router, there is no way it could work on one computer and not the other?

My Archer A20 doesn't have a F/W Update feature? Some routers do though. Still, if code changes in a router, ALL devices use the same code. Basically I think this means you had to manually update the router, and that is done via the GUI and a browser. Maybe someone else in the house who had the router login info did it?

This is what your F/W update page should look like (from the emulator):

No Auto-Update possibility. Also, see the bottom like, wonder what it would look like with the Beta f/w? That might answer all the other questions you had about the date/version and if it might work on newer f/w even?

@IrvSp ,

1. My "ageless" update suggestion was a conceptual idea.  I was not trying to suggest a particular implementation, except to explain the idea.

2. There's no use speculating further on why they removed the function.  You and I have both given the same short list of possibilities three times already, and I think we're both right.

3. I'll repeat that I leave open the possibility that my memory might have been faulty about the events and timeline when things started failing.

4. I never even tried their beta, so I cannot answer any of your questions about it.  I don't even know for sure whether it works, though I have no reason to doubt it.

5. Not being allowed to upgrade was a showstopper for me.  It wasn't worth it.  I since received my print server device from Amazon.  $39.99 w/ free delivery.  It's already set up and it's working great!

----------

Thank you all!  As far as I'm concerned, this is the end of the story for me.................