fix(github-sync): retry transient network errors during GraphQL batch (#144)

GitHub's GraphQL endpoint occasionally closes connections mid-response,
raising EOFError (and siblings like Errno::ECONNRESET / OpenSSL::SSL::SSLError)
out of Net::HTTP. The retry loop in User::GithubSyncable#github_graphql_request
only caught Net::OpenTimeout/Net::ReadTimeout, so any connection reset
escaped the rescue and failed the entire batch of users without retrying —
observed in production as EOFError "end of file reached" from
UpdateGithubDataJob.

Extract the full set of transient network error classes into
TRANSIENT_NETWORK_ERRORS and rescue them uniformly so they trigger the
existing backoff/retry instead of aborting the batch.

Added regression tests covering both paths: retry-then-succeed and
retries-exhausted.

Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: newstler

app/models/concerns/user/github_syncable.rb

@@ -1,11 +1,26 @@
 require "net/http"
 require "json"
+require "openssl"
+require "socket"
 
 module User::GithubSyncable
   extend ActiveSupport::Concern
 
   GITHUB_GRAPHQL_ENDPOINT = "https://api.github.com/graphql"
 
+  TRANSIENT_NETWORK_ERRORS = [
+    Net::OpenTimeout,
+    Net::ReadTimeout,
+    Net::WriteTimeout,
+    EOFError,
+    IOError,
+    SocketError,
+    Errno::ECONNRESET,
+    Errno::ECONNREFUSED,
+    Errno::EPIPE,
+    OpenSSL::SSL::SSLError
+  ].freeze
+
   # Sync from OAuth callback data
   def sync_github_data_from_oauth!(auth_data)
     api_token = auth_data.credentials.token
@@ -155,12 +170,12 @@ def github_graphql_request(query, api_token, retries: 3)
           else
             return { errors: [ "HTTP #{response.code}: #{response.message}" ] }
           end
-        rescue Net::OpenTimeout, Net::ReadTimeout => e
+        rescue *TRANSIENT_NETWORK_ERRORS => e
           if attempt < retries - 1
             sleep(2 ** (attempt + 1))
             next
           else
-            return { errors: [ "Request timed out: #{e.message}" ] }
+            return { errors: [ "Network error: #{e.class}: #{e.message}" ] }
           end
         end
       end

test/models/concerns/user/github_syncable_test.rb

@@ -0,0 +1,51 @@
+# frozen_string_literal: true
+
+require "test_helper"
+require "webmock/minitest"
+
+class User::GithubSyncableTest < ActiveSupport::TestCase
+  setup do
+    WebMock.disable_net_connect!(allow_localhost: true)
+    @user = users(:user_with_testimonial)
+    @user.update_columns(username: "octocat")
+  end
+
+  teardown do
+    WebMock.reset!
+    WebMock.allow_net_connect!
+  end
+
+  # Skip the exponential backoff so tests run fast.
+  setup { User.singleton_class.send(:define_method, :sleep) { |_| } }
+  teardown { User.singleton_class.send(:remove_method, :sleep) }
+
+  test "batch_sync_github_data! retries on EOFError and succeeds on second attempt" do
+    stub_request(:post, User::GithubSyncable::GITHUB_GRAPHQL_ENDPOINT)
+      .to_raise(EOFError.new("end of file reached"))
+      .then.to_return(
+        status: 200,
+        body: {
+          data: {
+            user_0: { login: "octocat", name: "The Octocat", email: nil, bio: nil, company: nil, websiteUrl: nil, twitterUsername: nil, location: nil, avatarUrl: "https://example.com/a.png" },
+            repos_0: { nodes: [] }
+          }
+        }.to_json
+      )
+
+    result = User.batch_sync_github_data!([ @user ], api_token: "token")
+
+    assert_equal 1, result[:updated], result[:errors].inspect
+    assert_equal 0, result[:failed]
+  end
+
+  test "batch_sync_github_data! returns a network error after exhausting retries" do
+    stub_request(:post, User::GithubSyncable::GITHUB_GRAPHQL_ENDPOINT)
+      .to_raise(EOFError.new("end of file reached"))
+
+    result = User.batch_sync_github_data!([ @user ], api_token: "token")
+
+    assert_equal 0, result[:updated]
+    assert_equal 1, result[:failed]
+    assert_match(/EOFError/, result[:errors].first.to_s)
+  end
+end