PeterH

Some previous thoughts below.
Especially @neilestrick
I agree that the organic binders/extenders are diminishing. Typically they put a biocide in there to preserve them, but if they are especially old it can wear out. Add some gum solution with a touch of copper carb in it to preserve it. I usually make gum solution 2 tbsp bum to a gallon of water, plus 1/8 teaspoon copper carb.
As your friend's underglaze seems to be having  a shortish life before thinning is there any special source of contamination/warmth that might be causing this? Are they using large bottles, and might decanting a working-stock into a smaller bottle help? (And maybe keeping the large bottles in a cooler location.)
 

My orange and red velvets get thin after a longggggg time.  In the pints.  The cost has gotten so prohibitive, I am no longer ordering pints, except for white.  And the Flame orange does have that sulphur smell.  after researching last year, I came to the conclusion that @Min stated, it's the CMC breaking down.  But I do not live in a humid area, nor is my studio humid.  Quite the opposite.  I live in the Sahara of Colorado.  Maybe humidity hastens the process.  

References to boiled green-tea, seven-up and sugar-water as additives in this and the following few posts.
 

Does the underglaze smell at all? Thinking the gum(s) are breaking down.

I prefer to use the term "durable" rather than food safe when talking about glazes. There isn't one agreed upon definition of what makes a glaze "food safe". In North America if a glaze does not contain lead or cadmium it can be labelled food safe. This is a very wide definition and adding something along the lines of what Mayco says "Finishedware producers bear responsibility for food safe testing their ware being sold into commerce." IMHO isn't good enough. Like Callie said Strontium Crystal Magic (SCM) is  fine on the outside of pots where your mouth or food won't be in contact with it.
Re lithium and toxicity, just adding a little bit. Topic of lithium and toxicity has come up in many discussions in many forums etc over the years. My impression of the reading I've done is it might not be an issue for most people but for those on lithium therapy it can be, both in the process of mixing glazes and leaching glazes.  Poison is in the dosage scenario.
With all due respect to Callie I found SCM quite rough, dry and unmelted. You could try it by itself on a test tile and see how you find it but it's a bit of a moot point because it isn't used by itself.  The overload of titanium is what makes the crystals/variegation when layered with other glazes. A good way to get interesting glaze reactions is to layer 2 or more dissimilar glazes, what one glaze lacks or has a surplus of reacts with the other glaze, this often results in interesting visual texture. SCM is loaded with fluxes to react with other glazes, that's part of the reason why it works.
Re silica and glaze durability, having as much silica and alumina in a glaze is one of the ways to ensure a durable glaze.  For cone 6 having at least 2.5 molar of silica is a good starting place, if the glaze can dissolve more then it's a good idea to add it if you are looking for a durable glaze. If you look at the screenshot above in the bottom half of the image you can see the silica is 1.68 so at the bare minimum it is about 1/3 short of where it should be. Overloading a matte glaze can quite often turn it into a gloss glaze  and really oversupplying it can turn it back into a matte and crazed glaze so there are parameters within which to work. 
Another aspect of silica that is often overlooked is the mesh size of the silica being used. When you mix glazes using a 325 mesh size silica more of the silica will likely be dissolved in the glaze than if you use 200 mesh. (not applicable for the SCM)
One more thing is to look at the colouring oxides used. Don't use crazy amounts of colouring/transition oxides, use the least you can to still achieve the effect/colour you are after.
There is a good article on toxicity that includes materials here if it helps.

Toxicity is a whole entire subject by itself. Mostly we have to start with definitions to help clarify. 
Tl;dr, yes you can put this glaze on the outside of a pot and it will be fine. It’s not touching food. It melts fine at cone 6, it’s just a “soft” glaze.
When a raw material is labelled toxic, that means it’s toxic in its current form. Once you put a material into a glaze with other materials and fire it, it becomes another material entirely, with different properties. If you’re using a commericallly mixed glaze that is labelled non-toxic, that means it’s not toxic in the unfired state. (It’s probably not toxic in the fired state either, but you’d have to test it to confirm, and manufacturers can’t say that it is for sure under all circumstances without legal problems. People do weird and unexpected things.)
When we’re talking about glazes, most often the finished glaze is more safe than, say, raw lithium. So potters should be concerned about their own health and safety first, and the end user second. Materials like lithium can be handled safely with sensible studio hygiene practices, and can be beneficial in a lot of good glazes.
When knowledgeable glaze chemists talk about food safety, we aren’t talking about whether a glaze will leach Bad Things (TM) into food. This is a very infrequent scenario outside of lead use, and lead hasn’t been used in most handmade pottery for decades now. But we are thinking about craftsmanship and durability. We’re asking questions like Does this glaze have flaws like crazing or pinholing, that can affect structural integrity or possibly harbour bacteria? Does this glaze scratch, or does it cutlery mark? Does it stain easily? Is it easy to clean? If those problems are solved, leaching is usually a non-issue. 
Re the silica question: You need 3 types of components in the right proportions in a base glaze for durability: silica to form glass, alumina for strength and to make it not run too badly, and different metals that make silica and alumina melt at much lower temperatures than they do by themselves (fluxes). Without enough silica or alumina, or if those things are in the wrong proportions, the glaze may react to acids found in many drinks, bases like in dishwasher detergent, or may scratch or cutlery mark. SCM doesn’t supply enough silica for durability, but the strontium is also a problem.
Within the flux category, there’s a bunch of stuff that’s a good idea to use in combination, because each material has beneficial and undesirable properties, and mixing them can bring out the best (or sometimes worst) of both worlds. For example, sodium is a strong flux, which is needed to bring down the melting temperature of silica and alumina, and is in most glazes. However, it tends to make glazes craze when by itself. But if you mix it with calcium (or barium or strontium or lithium and a few others), you can still get the melting, but less crazing.
The calcium/magnesium/strontium etc group of fluxes also give you different colour responses, or can help create matte effects if the proportions are high enough. One of the drawbacks of using strontium or barium though is that neither is very high on the Moh hardness scale, so if you use too much of it, the resulting glaze can be prone to scratching, staining or other flaws. But it helps create that really cool matte surface because it promotes crystallization, so sometimes creative folks make trade offs. 

There is some where a black which iz  acheived by a mixture of iron oxide, cobalt and Manganese dioxide. 
This doesn't discount the above info but certainly gets to a black black. Some primitive and ancient potters get a great black.

Interesting, thanks for the pićtures. They are super melted but hard to tell, it appears the cone on the left in the picture (cone 5)  is facing cones 6 & 7 so it will melt on top of them, and it did! Since this is an important calibrating type fire I would suggest setting cones so they melt away from each other. I am not a fan of self supporting cones being elevated on clay as it often changes the tip height slightly and can change the angle slightly if not installed perfectly flat. Maybe instead of a wad of clay, kiln wash or a touch of alumina for the shelf beneath the cone and set cones so they fall away from each other. Maybe for a test fire with some extra shelves or simulated loading to see how this performs with a typical load.
So for accurate calibration I suggest you test where this really ends before entering offsets. The cone pack in your picture likely does not confirm accurately because they melted on top of each other.
Typical cone pack below, Guide ( melts first), Firing, Guard ( should not melt much) Sue McLeod has a nice pdf on cone packs, the picture below is from Sues website and can also be found here: https://suemcleodceramics.com/how-to-make-a-cone-pack/
 

Interesting ...when I read your post last night, sparked a memory, but not sure what...
...perhaps the mill where I once worked, a machine governed by a PLC* would revert to metric.
We found an initialization that ran on cold start up only, then made provision to run the necessary bits on a bounce (power blip).
Perhaps there's a similar situation with your pi? I did a bit of reading last night - they are Centigrade  by default, yes?
 
*Programmable Logic Controller, iow, "real-time" operating system machine, where the programs run on specific clock intervals.

If we have a look at the recipe and formula mentioned in the thread above there are some alterations to make to the glaze to get it to melt at cone 6.(info from that thread at the bottom of this post) Either boron (from a frit or Gerstley Borate) or zinc will be needed to flux it enough for midrange.
For boron the obvious choice would be Ferro Frit 3269 as its a boron frit with very low calcium with no other alkaline earths but it also contains zinc. However, it is very high in sodium,( Na2O), and potassium, (K2O), which is likely to increase glaze crazing from their high Coefficient of Expansion (COE). So to adjust the glaze to reduce the COE there must be either Lithium added at the expense of some of the K2O / Na2O or one of the low COE alkaline earths, magnesium, (MgO), would be a good choice. Problem with this is then you are back to alkaline earths washing out the iron, albeit less so with MgO than CaO.
Second problem to address is the colour response zinc can cause with iron. Granted there isn't a great deal of it with Ferro 3269 so this might or might not be an issue if you can come up with a formula using it with lithium to counteract the high COE plus increase the Al2O3 and SiO2 would be necessary to make a workable midrange glaze. Might be somewhere to start if you want a deep dive into glaze chem and testing.

 
44% North Cape nepheline syenite
34% Potash felspar
20% AB clay
2%  Kaolin
... which is attempting to achieve
0.8 K2O
0.2 Na2O
1.17 Al2O3
0.05 Fe2O3
5.35 SiO2
edit: for the purposes of this discussion I have left the flux ratio out of the equation but it should be considered.

My opinion / experience
12 Gauge SEW wire is very similar to SMRL rated about 200C and braided over silicone. SEW more of a Canadian specification I believe. Neither SMRL or SEW meet or exceed the MG wire that your kiln likely was built with (-450C) so in that sense the MG is superior. All will likely work fine though, I am just an old engineer. (Very old).
With respect to the buss bars, they look corroded, so I would wire brush (Dress) the bars and ends to remove all the corrosion and reinstall with new bolt hardware (low cost, mainly labor). The bus bars are better than any wire and crimp or clamp connection IMO (for several reasons) and if replacing with wires these connections would need to be thoroughly cleaned anyway and new bolt and nut hardware for ring terminal type wire termination (Presently on your kiln element leads) when replacing with wire. One could butt splice the wire directly to the element as well but this introduces an additional splice to the connection and is not easily removable during future element change out.
Since the element tales will be well beyond 400c at points in the firing, the mg in my mind is superior. The buss bars cool themselves because of the metal surface and are rated well beyond 400C. To me clean buss bars = best, MG wire good, lesser rated wire would be my third choice. All should work though.
Looking at your picture, this kiln appears to have been rewired removing the old time delay relays and installing two infinite switches which is good news IMO.
For me I would:
Clean all buss bars and connections and reinstall with new nuts and bolts Check all remaining connections, clean and install new crimp or nut and bolt hardware. Check the infinite switch operation, maybe anticipate installing two new because they are fairly economical Measure the element resistance to see if they are worn. This would be the most costly and I am going to guess likely needed and you ought to check first from a budget perspective. For future digital control you would need a controller, power transformer, relays to replace the infinite switches,  thermocouple(s) misc wire and assorted mounting hardware. If you are handy or know someone (reasonably skilled) that can help, I would guess $250.00 - $500.00 using an older Bartlett V6cf control and fitting it into the control box. @neilestrick probably has a preference here based on ease of retrofit install and available space in the control box. I have in the past drawn up a solution for friends  that uses a separate control box for the controller, relays, transformer, etc… that gets bolted right to the front of this panel. It is economical but definitely takes a bit of effort and wiring skills.
Late add - this is the diagram for the DK kiln it can be found in the appendix of their service manual and can be downloaded https://corp.paragonweb.com/wp-content/uploads/LX_809_Duncan_Kiln_Service_Manual-1.pdf
Its a simple kiln - no electronic timers

I also use the clay bag in the bucket method which works well.  One caution, make sure your clay bag has no holes.  That's a sure way to get slop!  Lin

I've also read of "flat" seven-up as an alternative to green tea as a solvent/vehicle. I suppose that the sugar acts as a binder as the "ink" dries.

Some ideas in ...
 

Some ideas in ...
 

There is some conflicting information about the first firing floating around, both from different kiln manufacturers, for different purposes, and even within the same manufacturer's instructions. L&L's printed manual has long instructed owners of their new kiln to do an initial break-in firing of the empty kiln (but with the furniture) slow bisque to cone 5 (yes, five, not oh-five) with a 3 hour preheat for the dual purpose of seasoning the elements to develop a protective oxide coating and to set the cement used during manufacturing to hold the kiln bricks together. Their newest manual indicates these instructions are for both the Dynatrol and Genesis controllers.
Conventional wisdom by some other kiln manufacturers instruct owners of their new kilns to do the initial firing slow to 04. This is consistent with the recommendation from the manufacturer of the Kanthal elements for seasoning newly installed replacement elements, again to develop a protective oxide coating on the wire. Out on the interwebs, many commenters who own these other brands of kiln will adamantly (but incorrectly) assert that new owners of all brands of kilns should do the initial firing to 04.
And now we have this new video from L&L for programming the first firing on a Genesis controller that instructs a glaze firing to 04 with a 1 hour preheat. However, if one looks around on the L&L website, there is another video for programming the older Dynatrol controller for the first firing that is consistent with the printed instructions, i.e., slow bisque to 5. Is the basic kiln constructed differently for a Genesis vs. Dynatrol controller that it would need a different initial firing? I don't work there, so don't take my word for it, but the conflict seems fishy to me. With 3 instruction sources to choose from (the printed manual and 2 videos), I would go with the 2 that are consistent, i.e, the printed manual and the Dynatrol video. But maybe that's just because I am a recovering accountant...

I've also read of "flat" seven-up as an alternative to green tea as a solvent/vehicle. I suppose that the sugar acts as a binder as the "ink" dries.

Brent did get back to me. They sent a video and instructions. I tried attaching the paper instructions but the file was to large. But, here is the youtube format. 
 

A weird glaze result is one thing, but it's really hard to believe that an under-glaze could completely disappear on firing. Let alone four different under-glazes: red, orange, black & light green.
Sorry, but to err is human.

A weird glaze result is one thing, but it's really hard to believe that an under-glaze could completely disappear on firing. Let alone four different under-glazes: red, orange, black & light green.
Sorry, but to err is human.

A weird glaze result is one thing, but it's really hard to believe that an under-glaze could completely disappear on firing. Let alone four different under-glazes: red, orange, black & light green.
Sorry, but to err is human.

Hi, tired potter?
I have v late at night early morning glazed only one half of bowls ..light not the best, brain fatigue...if rest of kiln perfect, human error?
But worth investigating.

To leak slowly.
A quick test is to fill the teapot and leave it on a paper towel overnight. The towel should remain dry.

Just to add to @PeterH I have left vases for several days for that really tiny leak. Nothing worse than giving someone a vase they use at work and after several days they tell you their vase leaked …… on their desk! No harm in testing longer.

To leak slowly.
A quick test is to fill the teapot and leave it on a paper towel overnight. The towel should remain dry.