A client that is already gone leaves no monitor to demonitor. Crashing
the worker takes the shared connection down with it and fails every
other client using it.
Three paths obtained a client ref and abandoned it without closing:
the content-length rejection in call/3, the invalid-status clause in
request/4, and the success branch of response/6.
Under Hackney 4 an abandoned ref is a permanent process leak. A conn
that finishes or errors without an explicit stop parks in `closed` or
`connected` and never terminates: closed/3 only arms its stop timer for
pooled conns, `connected` gets no state_timeout when idle_timeout is
infinity, and a non-pooled conn monitors hackney_conn_sup rather than
its caller, so the owner-DOWN cleanup never fires. Each parked process
pins every refc binary it read, which for proxied video is tens of MB.
Releasing on every outcome means the gun client must stop releasing in
stream_body/1, or the ref would be removed twice.
Hackney does reclaim a pooled connection itself for a response that
cannot carry a body (204, 304, HEAD), so those refs are marked and left
alone; stopping one would evict a connection another request may
already hold. The mark is gated on the request actually being pooled,
since a direct or proxy-tunnelled conn is never checked in and leaks
unless we stop it.
Checking against the default port of the Endpoint URL scheme is
redundant as normal instances will have the combination https/443
by default created by pleroma.instance gen, Tor-only instances should
have combination http/80 and local testing instances httt/XXXX.
The default scheme port doesn't add anything usefull in these configs.
increase_vote_count/3 was incrementing votersCount on every vote
activity, causing inflation when a single voter picks multiple options.
Now only increments when the actor is a new unique voter, and preserves
existing votersCount otherwise.
Also adds is_integer guard to voters_count/1 to handle nil safely, and
adds tests for the voters_count clause ordering and edge cases.
Reject unknown remote Update targets and invalidate signed payloads when their signer identity cannot be mapped, avoiding crashes and fail-open signature state.
