Exercise 1: "Too Many Connections" With Identical Prior-Day Traffic — Possible Solution
====================================================================

The most likely first thing to check: whether a NEW DEPLOYMENT went
out between yesterday and today -- specifically, one that could have
changed connection pooling behavior or introduced a connection leak
(connections not being properly closed or returned to the pool).

Why this is the most likely first check:

The key detail in the scenario is that the SAME traffic level was
handled fine yesterday. If connection exhaustion were simply a matter
of the database's connection limit being too low for the traffic
volume, it would presumably have already been a problem yesterday
too, under the same load. Since it wasn't a problem then and IS now,
something must have CHANGED between the two points in time --
directly matching this chapter's own point that "a recent code
deployment changing connection pooling behavior" is a genuinely common
root cause specifically because it explains why previously-fine load
suddenly becomes a problem.

This also directly applies Chapter 4's own "what changed and when"
discipline -- rather than assuming the traffic level itself is
inherently too high (which the identical-yesterday detail argues
against), the investigation should start by checking recent deployment
history around the time the errors began, before assuming the fix is
simply "increase the connection limit."

Confirming this: per the chapter, check the database's current
connection count against its configured maximum directly -- if
connections are climbing toward the max without a proportional
increase in actual query throughput, that pattern is consistent with a
leak (connections being opened but not properly released) rather than
simply more legitimate concurrent usage.

WHY THIS WORKS AS AN ANSWER
------------------------------
The reasoning hinges specifically on the "same traffic, different
outcome" detail given in the scenario, which is exactly the signal
that should point an investigator toward "what changed" (Ch.4) rather
than "the traffic is just too high" -- this chapter's own material
names a recent deployment as the specific, common explanation for
precisely this pattern.