Fix race create sessione su CPX: ready solo dopo MediaMTX.

I nodi cloud restano in provisioning finché :9997 risponde; retry su create_path e 503 retryable se l’ingest è ancora irraggiungibile.

Co-authored-by: Cursor <cursoragent@cursor.com>
This commit is contained in:
2026-08-21 14:06:47 +02:00
co-authored by Cursor
parent 70fccc493e
commit b87a0f7bc0
12 changed files with 372 additions and 3 deletions
@@ -0,0 +1,52 @@
# frozen_string_literal: true
require "rails_helper"
RSpec.describe "API session create ingest unavailable", type: :request do
let!(:user) do
User.create!(email: "ingest-#{SecureRandom.hex(4)}@example.com", name: "Coach", password: "Password123", role: "coach")
end
let!(:club) { Club.create!(name: "IngestClub", sport: "volleyball") }
let!(:membership) { club.club_memberships.create!(user: user, role: "owner") }
let!(:team) { club.teams.create!(name: "Tigers", sport: "volleyball", slug: "ingest-tigers-#{SecureRandom.hex(3)}") }
let!(:match) { team.matches.create!(opponent_name: "Opp", scheduled_at: 1.hour.from_now) }
def auth_headers
post "/api/v1/auth/login", params: { email: user.email, password: "Password123" }
token = response.parsed_body["access_token"]
{ "Authorization" => "Bearer #{token}", "Content-Type" => "application/json" }
end
before do
plan = Plan.find_or_initialize_by(slug: "premium_full")
plan.name ||= "Premium Full"
plan.features = (plan.features || {}).merge(
"platforms" => %w[matchlivetv youtube],
"youtube_enabled" => true,
"concurrent_streams_limit" => 10,
"recordings_enabled" => true,
"recording_days" => 90
)
plan.save!
club.create_subscription!(plan: plan, status: "active") if club.subscription.blank?
allow_any_instance_of(Teams::Entitlements).to receive(:assert_can_stream_on!)
allow_any_instance_of(Teams::Entitlements).to receive(:assert_concurrent_stream!)
allow(Streams::CoverSlateEnsurer).to receive(:ensure_for!)
allow(Streams::SlateDistributor).to receive(:ensure_for!)
Streams::NodeRegistry.ensure_home_from_env!
end
it "returns 503 when MediaMTX connection fails after retries" do
allow_any_instance_of(Mediamtx::Client).to receive(:create_path)
.and_raise(Faraday::ConnectionFailed.new("Connection refused"))
post "/api/v1/matches/#{match.id}/sessions",
params: { platform: "matchlivetv", privacy_status: "private" }.to_json,
headers: auth_headers
expect(response).to have_http_status(:service_unavailable)
body = response.parsed_body
expect(body["error_code"]).to eq("stream_ingest_unavailable")
end
end
@@ -0,0 +1,50 @@
# frozen_string_literal: true
require "rails_helper"
RSpec.describe Mediamtx::Client do
describe "#create_path" do
let(:session) do
user = User.create!(email: "mtx-#{SecureRandom.hex(4)}@example.com", name: "M", password: "Password123", role: "coach")
club = Club.create!(name: "MtxClub-#{SecureRandom.hex(3)}", sport: "volleyball")
team = club.teams.create!(name: "T", sport: "volleyball", slug: "mtx-#{SecureRandom.hex(4)}")
match = team.matches.create!(opponent_name: "X", scheduled_at: 1.hour.from_now)
StreamSession.create!(match: match, user: user, platform: "matchlivetv", status: "idle")
end
it "retries Faraday connection failures then succeeds" do
ENV["MEDIAMTX_CREATE_RETRIES"] = "3"
ENV["MEDIAMTX_CREATE_RETRY_BASE_SECS"] = "0"
client = described_class.new(base_url: "http://mtx.test:9997")
conn = instance_double(Faraday::Connection)
client.instance_variable_set(:@conn, conn)
fail_once = Faraday::ConnectionFailed.new("Connection refused")
ok = instance_double(Faraday::Response, success?: true, status: 200, body: {})
expect(conn).to receive(:post).once.and_raise(fail_once)
expect(conn).to receive(:post).once.and_return(ok)
expect(client.create_path(session)).to eq(true)
ensure
ENV.delete("MEDIAMTX_CREATE_RETRIES")
ENV.delete("MEDIAMTX_CREATE_RETRY_BASE_SECS")
end
it "raises after exhausting connection retries" do
ENV["MEDIAMTX_CREATE_RETRIES"] = "2"
ENV["MEDIAMTX_CREATE_RETRY_BASE_SECS"] = "0"
client = described_class.new(base_url: "http://mtx.test:9997")
conn = instance_double(Faraday::Connection)
client.instance_variable_set(:@conn, conn)
allow(conn).to receive(:post).and_raise(Faraday::ConnectionFailed.new("Connection refused"))
expect { client.create_path(session) }.to raise_error(Faraday::ConnectionFailed)
ensure
ENV.delete("MEDIAMTX_CREATE_RETRIES")
ENV.delete("MEDIAMTX_CREATE_RETRY_BASE_SECS")
end
end
end
@@ -224,4 +224,41 @@ RSpec.describe Streams::Autoscaler do
expect(described_class.within_budget?(1)).to eq(false)
end
end
it "promotes provisioning nodes when MediaMTX becomes reachable" do
with_env(
"STREAM_AUTOSCALE_ENABLED" => "1",
"STREAM_AUTOSCALE_WARM_SPARE" => "0",
"STREAM_AUTOSCALE_SOFT_FREE_SLOTS" => "0",
"STREAM_AUTOSCALE_KIND" => "lab",
"MEDIAMTX_API_URL" => "http://mtx-home:9997",
"MEDIAMTX_RTMP_URL" => "rtmp://home.example:1935",
"HLS_PUBLIC_URL" => "https://home.example/hls"
) do
Streams::NodeRegistry.ensure_home_from_env!
node = StreamNode.create!(
slug: "ingest-lab-prov",
hostname: "prov.lab",
role: "lab",
status: "provisioning",
provider: "local",
rtmp_base_url: "rtmp://h:1935",
hls_base_url: "https://h/hls",
api_base_url: "http://h:9997",
max_publishers: 2,
max_relays: 2
)
allow(Streams::NodeHealth).to receive(:promote_if_healthy!) do |n|
n.update!(status: "ready", last_health_at: Time.current)
true
end
provisioner = instance_double(Streams::NodeProvisioner)
expect(provisioner).not_to receive(:provision_lab!)
result = described_class.reconcile!(provisioner: provisioner)
expect(result.actions).to include(:"ready_ingest-lab-prov")
expect(node.reload.status).to eq("ready")
end
end
end
@@ -0,0 +1,39 @@
# frozen_string_literal: true
require "rails_helper"
RSpec.describe Streams::NodeHealth do
let!(:node) do
StreamNode.create!(
slug: "ingest-health-01",
hostname: "ingest-health-01.mltv-stream.net",
role: "cloud",
status: "provisioning",
provider: "hetzner",
rtmp_base_url: "rtmp://h:1935",
hls_base_url: "https://h/hls",
api_base_url: "http://203.0.113.10:9997",
max_publishers: 4,
max_relays: 4
)
end
after { node.destroy }
it "promotes provisioning node when MediaMTX is reachable" do
client = instance_double(Mediamtx::Client, reachable?: true)
allow(Mediamtx::Client).to receive(:new).with(base_url: node.api_base_url).and_return(client)
expect(described_class.promote_if_healthy!(node)).to eq(true)
expect(node.reload.status).to eq("ready")
expect(node.last_health_at).to be_present
end
it "does not promote when MediaMTX is down" do
client = instance_double(Mediamtx::Client, reachable?: false)
allow(Mediamtx::Client).to receive(:new).with(base_url: node.api_base_url).and_return(client)
expect(described_class.promote_if_healthy!(node)).to eq(false)
expect(node.reload.status).to eq("provisioning")
end
end
@@ -24,11 +24,15 @@ RSpec.describe "Streams::NodeProvisioner cloud" do
ENV["STREAM_CLOUD_DNS_SUFFIX"] = "mltv-stream.net"
ENV["STREAM_CLOUD_MAX_PUBLISHERS"] = "4"
ENV["STREAM_CLOUD_PUBLIC_CONTROL"] = "0"
ENV["STREAM_NODE_READY_TIMEOUT_SECS"] = "0"
allow(Streams::NodeHealth).to receive(:promote_if_healthy!).and_return(false)
node = Streams::NodeProvisioner.new(cloud: cloud, dns: dns).provision_cloud!
expect(node.slug).to eq("ingest-01")
expect(node.role).to eq("cloud")
expect(node.provider).to eq("hetzner")
expect(node.status).to eq("provisioning")
expect(node.hostname).to eq("ingest-01.mltv-stream.net")
expect(node.rtmp_base_url).to eq("rtmp://ingest-01.mltv-stream.net:1935")
expect(node.api_base_url).to eq("http://10.0.0.9:9997")
@@ -39,6 +43,45 @@ RSpec.describe "Streams::NodeProvisioner cloud" do
%w[
MEDIAMTX_API_URL MEDIAMTX_RTMP_URL HLS_PUBLIC_URL
STREAM_CLOUD_DNS_SUFFIX STREAM_CLOUD_MAX_PUBLISHERS STREAM_CLOUD_PUBLIC_CONTROL
STREAM_NODE_READY_TIMEOUT_SECS
].each { |k| ENV.delete(k) }
end
it "marks cloud node ready when MediaMTX answers during wait" do
cloud = instance_double(
Streams::CloudProviders::Hetzner,
create_node: Streams::CloudProviders::Instance.new(
id: "100",
name: "mltv-stream-ingest-01",
public_ip: "49.13.9.9",
private_ip: "10.0.0.9",
status: "running",
raw: {}
)
)
dns = instance_double(Streams::DnsProviders::Hetzner)
allow(dns).to receive(:upsert_a)
ENV["MEDIAMTX_API_URL"] = "http://mtx-home:9997"
ENV["MEDIAMTX_RTMP_URL"] = "rtmp://home.example:1935"
ENV["HLS_PUBLIC_URL"] = "https://home.example/hls"
ENV["STREAM_CLOUD_DNS_SUFFIX"] = "mltv-stream.net"
ENV["STREAM_CLOUD_PUBLIC_CONTROL"] = "0"
ENV["STREAM_NODE_READY_TIMEOUT_SECS"] = "5"
ENV["STREAM_NODE_READY_POLL_SECS"] = "0.01"
allow(Streams::NodeHealth).to receive(:promote_if_healthy!) do |node|
node.update!(status: "ready", last_health_at: Time.current)
true
end
node = Streams::NodeProvisioner.new(cloud: cloud, dns: dns).provision_cloud!
expect(node.status).to eq("ready")
ensure
%w[
MEDIAMTX_API_URL MEDIAMTX_RTMP_URL HLS_PUBLIC_URL
STREAM_CLOUD_DNS_SUFFIX STREAM_CLOUD_PUBLIC_CONTROL
STREAM_NODE_READY_TIMEOUT_SECS STREAM_NODE_READY_POLL_SECS
].each { |k| ENV.delete(k) }
end
end