Hostname → IP resolution that works around BEAM's broken DNS path on iOS.
Why this exists
BEAM resolves hostnames by spawning an external helper called
inet_gethost (a port program). On macOS, Linux, Windows that
works fine. On iOS it doesn't — the iOS app sandbox forbids
execve of any binary the app didn't get a special pass for, and
there's no equivalent of Android's lib*.so escape hatch.
Result: :inet.getaddr/2 (and therefore Req, Finch, Mint,
ReqLLM, and basically every Elixir HTTP library) fails on iOS
the moment a request hits a hostname rather than a literal IP.
This module side-steps the problem by calling Darwin's
getaddrinfo directly via a NIF, then seeding :inet_db with
the result so subsequent BEAM-level lookups for the same host
succeed from the in-process file table.
Android isn't affected — mob_beam.zig ships inet_gethost as
libinet_gethost.so in jniLibs/, which the SELinux policy
allows to execve. The NIF here would work on Android too but
isn't wired up by default; the BEAM path is already functional
there.
How to use it
Resolve each hostname your app talks to before the first
Req / Finch / Mint call to that host. Once resolved, :inet_db
retains the mapping for the lifetime of the BEAM, so subsequent
HTTP calls go through without you doing anything else.
# At app startup, or before the first call:
{:ok, _ip} = Mob.DNS.resolve("api.example.com")
# Now this just works on iOS:
Req.get!("https://api.example.com/v1/things")For a small fixed set of hosts, the convenience helper
preresolve/1 does the whole list at once:
Mob.DNS.preresolve([
"api.example.com",
"auth.example.com"
])Scope and limitations
- IPv4 only. Most cloud endpoints serve A records; IPv6 is a follow-up if it becomes useful.
- One IP per host. If the hostname has multiple A records, the first one is used. BEAM caches the result; failover isn't automatic. If your endpoint cycles IPs frequently you may need to re-resolve.
- No automatic refresh. Mappings stay in
:inet_dbuntil the BEAM exits. If a backend's IP changes mid-session, the cached entry will be stale — callresolve/1again to refresh. - Doesn't help raw NIF networking. If a third-party NIF calls
libc
getaddrinfoitself, it never goes through BEAM's DNS layer and doesn't need (or benefit from) this fix — it already works on iOS. Only:inet-mediated lookups (which covers almost all Elixir HTTP libraries) need our help. - iOS only effectively. On Android and host (dev, macOS, Linux) the NIF works but is unnecessary; BEAM's built-in path is fine there.
Errors
{:ok, {a, b, c, d}} # success
{:error, :badarg} # host arg invalid
{:error, :nxdomain} # no such hostname
{:error, :timeout} # resolver TRY_AGAIN
{:error, :no_address} # resolved but no IPv4
{:error, {:gai, code}} # raw getaddrinfo error code
{:error, :nif_not_loaded} # called off-device (host tests)
Summary
Types
The error shapes resolve/1 can return.
Hostname to resolve. Latin-1 only — we're not in a domain that uses IDN.
Functions
Resolve a list of hostnames. Returns a map of host → result so the caller can see which ones failed.
Resolve host to an IPv4 address and seed :inet_db so subsequent
:inet.getaddr/2 lookups (and thus Req / Finch / Mint) find it.
True when host is already seeded in :inet_db.
Types
Functions
@spec preresolve([host()]) :: %{ required(host()) => {:ok, :inet.ip4_address()} | {:error, error_reason()} }
Resolve a list of hostnames. Returns a map of host → result so the caller can see which ones failed.
Useful at app startup for the known-fixed set of backends your app talks to.
%{
"api.example.com" => {:ok, {93, 184, 216, 34}},
"auth.example.com" => {:error, :nxdomain}
}
@spec resolve(host()) :: {:ok, :inet.ip4_address()} | {:error, error_reason()}
Resolve host to an IPv4 address and seed :inet_db so subsequent
:inet.getaddr/2 lookups (and thus Req / Finch / Mint) find it.
Idempotent — calling for the same host twice is harmless.
See module doc for usage, scope, and error shapes.
True when host is already seeded in :inet_db.
Useful for short-circuiting in caller code that wants to avoid an
unnecessary NIF call — but resolve/1 is idempotent, so calling
it again is also fine.