Module: Sidekiq::AssuredJobs
- Defined in:
- lib/sidekiq-assured-jobs.rb,
lib/sidekiq/assured_jobs/web.rb,
lib/sidekiq/assured_jobs/worker.rb,
lib/sidekiq/assured_jobs/version.rb,
lib/sidekiq/assured_jobs/middleware.rb
Defined Under Namespace
Modules: Web, Worker Classes: Error, Middleware
Constant Summary collapse
- VERSION =
"1.1.0"
Class Attribute Summary collapse
-
.auto_recovery_enabled ⇒ Object
Returns the value of attribute auto_recovery_enabled.
-
.delayed_recovery_count ⇒ Object
Returns the value of attribute delayed_recovery_count.
-
.delayed_recovery_interval ⇒ Object
Returns the value of attribute delayed_recovery_interval.
-
.heartbeat_interval ⇒ Object
Returns the value of attribute heartbeat_interval.
-
.heartbeat_ttl ⇒ Object
Returns the value of attribute heartbeat_ttl.
-
.instance_id ⇒ Object
Returns the value of attribute instance_id.
-
.logger ⇒ Object
Returns the value of attribute logger.
-
.namespace ⇒ Object
Returns the value of attribute namespace.
-
.recovery_lock_ttl ⇒ Object
Returns the value of attribute recovery_lock_ttl.
-
.redis_options ⇒ Object
Returns the value of attribute redis_options.
Class Method Summary collapse
-
.clear_unique_jobs_lock(job_data) ⇒ Object
Clear unique-jobs lock for orphaned jobs to allow immediate re-enqueuing.
- .configure {|_self| ... } ⇒ Object
- .get_instances_status ⇒ Object
- .get_orphaned_job_by_jid(jid) ⇒ Object
-
.get_orphaned_jobs_info ⇒ Object
Web interface support methods.
-
.namespaced_key(key) ⇒ Object
Helper method to add namespace prefix to Redis keys.
- .redis(&block) ⇒ Object
- .redis_sync(&block) ⇒ Object
- .reenqueue_orphans! ⇒ Object
- .setup_sidekiq_hooks ⇒ Object
Class Attribute Details
.auto_recovery_enabled ⇒ Object
Returns the value of attribute auto_recovery_enabled.
26 27 28 |
# File 'lib/sidekiq-assured-jobs.rb', line 26 def auto_recovery_enabled @auto_recovery_enabled end |
.delayed_recovery_count ⇒ Object
Returns the value of attribute delayed_recovery_count.
26 27 28 |
# File 'lib/sidekiq-assured-jobs.rb', line 26 def delayed_recovery_count @delayed_recovery_count end |
.delayed_recovery_interval ⇒ Object
Returns the value of attribute delayed_recovery_interval.
26 27 28 |
# File 'lib/sidekiq-assured-jobs.rb', line 26 def delayed_recovery_interval @delayed_recovery_interval end |
.heartbeat_interval ⇒ Object
Returns the value of attribute heartbeat_interval.
26 27 28 |
# File 'lib/sidekiq-assured-jobs.rb', line 26 def heartbeat_interval @heartbeat_interval end |
.heartbeat_ttl ⇒ Object
Returns the value of attribute heartbeat_ttl.
26 27 28 |
# File 'lib/sidekiq-assured-jobs.rb', line 26 def heartbeat_ttl @heartbeat_ttl end |
.instance_id ⇒ Object
Returns the value of attribute instance_id.
26 27 28 |
# File 'lib/sidekiq-assured-jobs.rb', line 26 def instance_id @instance_id end |
.logger ⇒ Object
Returns the value of attribute logger.
26 27 28 |
# File 'lib/sidekiq-assured-jobs.rb', line 26 def logger @logger end |
.namespace ⇒ Object
Returns the value of attribute namespace.
26 27 28 |
# File 'lib/sidekiq-assured-jobs.rb', line 26 def namespace @namespace end |
.recovery_lock_ttl ⇒ Object
Returns the value of attribute recovery_lock_ttl.
26 27 28 |
# File 'lib/sidekiq-assured-jobs.rb', line 26 def recovery_lock_ttl @recovery_lock_ttl end |
.redis_options ⇒ Object
Returns the value of attribute redis_options.
26 27 28 |
# File 'lib/sidekiq-assured-jobs.rb', line 26 def end |
Class Method Details
.clear_unique_jobs_lock(job_data) ⇒ Object
Clear unique-jobs lock for orphaned jobs to allow immediate re-enqueuing
62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 |
# File 'lib/sidekiq-assured-jobs.rb', line 62 def clear_unique_jobs_lock(job_data) return unless job_data['unique_digest'] begin # Check if SidekiqUniqueJobs is available if defined?(SidekiqUniqueJobs::Digests) SidekiqUniqueJobs::Digests.del(digest: job_data['unique_digest']) logger.info "AssuredJobs cleared unique-jobs lock for job #{job_data['jid']} with digest #{job_data['unique_digest']}" else logger.debug "AssuredJobs: SidekiqUniqueJobs not available, skipping lock cleanup for job #{job_data['jid']}" end rescue => e logger.warn "AssuredJobs failed to clear unique-jobs lock for job #{job_data['jid']}: #{e.message}" end end |
.configure {|_self| ... } ⇒ Object
28 29 30 31 32 |
# File 'lib/sidekiq-assured-jobs.rb', line 28 def configure yield self if block_given? setup_defaults setup_sidekiq_hooks end |
.get_instances_status ⇒ Object
171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 |
# File 'lib/sidekiq-assured-jobs.rb', line 171 def get_instances_status instances = {} redis_sync do |conn| # Get live instances instance_keys = conn.keys(namespaced_key("instance:*")) instance_keys.each do |key| instance_id = key.split(":").last heartbeat = conn.get(key) instances[instance_id] = { status: 'alive', last_heartbeat: heartbeat ? Time.at(heartbeat.to_f) : nil } end # Get dead instances with orphaned jobs job_keys = conn.keys(namespaced_key("jobs:*")) job_keys.each do |job_key| instance_id = job_key.split(":").last unless instances[instance_id] instances[instance_id] = { status: 'dead', last_heartbeat: get_instance_last_heartbeat(instance_id, conn), orphaned_job_count: conn.scard(job_key) } end end end instances end |
.get_orphaned_job_by_jid(jid) ⇒ Object
203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 |
# File 'lib/sidekiq-assured-jobs.rb', line 203 def get_orphaned_job_by_jid(jid) redis_sync do |conn| job_data_key = namespaced_key("job:#{jid}") job_payload = conn.get(job_data_key) if job_payload job_data = JSON.parse(job_payload) # Find which instance this job belongs to job_keys = conn.keys(namespaced_key("jobs:*")) job_keys.each do |job_key| if conn.sismember(job_key, jid) instance_id = job_key.split(":").last job_data['instance_id'] = instance_id job_data['orphaned_at'] = get_instance_last_heartbeat(instance_id, conn) job_data['orphaned_duration'] = calculate_orphaned_duration(job_data['orphaned_at']) break end end job_data end end end |
.get_orphaned_jobs_info ⇒ Object
Web interface support methods
135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 |
# File 'lib/sidekiq-assured-jobs.rb', line 135 def get_orphaned_jobs_info orphaned_jobs = [] redis_sync do |conn| # Get all job keys and instance keys job_keys = conn.keys(namespaced_key("jobs:*")) instance_keys = conn.keys(namespaced_key("instance:*")) # Extract live instance IDs live_instances = instance_keys.map { |key| key.split(":").last }.to_set job_keys.each do |job_key| instance_id = job_key.split(":").last unless live_instances.include?(instance_id) # Get all job IDs for this dead instance job_ids = conn.smembers(job_key) job_ids.each do |jid| job_data_key = namespaced_key("job:#{jid}") job_payload = conn.get(job_data_key) if job_payload job_data = JSON.parse(job_payload) job_data['instance_id'] = instance_id job_data['orphaned_at'] = get_instance_last_heartbeat(instance_id, conn) job_data['orphaned_duration'] = calculate_orphaned_duration(job_data['orphaned_at']) orphaned_jobs << job_data end end end end end orphaned_jobs.sort_by { |job| job['orphaned_at'] || 0 }.reverse end |
.namespaced_key(key) ⇒ Object
Helper method to add namespace prefix to Redis keys
57 58 59 |
# File 'lib/sidekiq-assured-jobs.rb', line 57 def namespaced_key(key) "#{namespace}:#{key}" end |
.redis(&block) ⇒ Object
34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 |
# File 'lib/sidekiq-assured-jobs.rb', line 34 def redis(&block) if # Use custom Redis configuration if provided redis_client = Redis.new() if block_given? result = yield redis_client redis_client.close result else redis_client end else # Use Sidekiq's Redis connection pool Sidekiq.redis(&block) end end |
.redis_sync(&block) ⇒ Object
51 52 53 54 |
# File 'lib/sidekiq-assured-jobs.rb', line 51 def redis_sync(&block) # Synchronous Redis operations using Sidekiq's pool or custom config redis(&block) end |
.reenqueue_orphans! ⇒ Object
78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 |
# File 'lib/sidekiq-assured-jobs.rb', line 78 def reenqueue_orphans! with_recovery_lock do logger.info "AssuredJobs starting orphan job recovery" redis_sync do |conn| # Get all job keys and instance keys (using custom namespacing) job_keys = conn.keys(namespaced_key("jobs:*")) instance_keys = conn.keys(namespaced_key("instance:*")) # Extract instance IDs from keys live_instances = instance_keys.map { |key| key.split(":").last }.to_set orphaned_jobs = [] job_keys.each do |job_key| instance_id = job_key.split(":").last unless live_instances.include?(instance_id) # Get all job IDs for this dead instance job_ids = conn.smembers(job_key) job_ids.each do |jid| # Get the job payload job_data_key = namespaced_key("job:#{jid}") job_payload = conn.get(job_data_key) if job_payload orphaned_jobs << JSON.parse(job_payload) # Clean up the job data key conn.del(job_data_key) end end # Clean up the job tracking key conn.del(job_key) end end if orphaned_jobs.any? logger.info "AssuredJobs found #{orphaned_jobs.size} orphaned jobs, re-enqueuing" orphaned_jobs.each do |job_data| # Clear unique-jobs lock before re-enqueuing to avoid lock conflicts clear_unique_jobs_lock(job_data) Sidekiq::Client.push(job_data) logger.info "AssuredJobs re-enqueued job #{job_data['jid']} (#{job_data['class']})" end else logger.info "AssuredJobs found no orphaned jobs" end end end rescue => e logger.error "AssuredJobs orphan recovery failed: #{e.message}" logger.error e.backtrace.join("\n") end |
.setup_sidekiq_hooks ⇒ Object
228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 |
# File 'lib/sidekiq-assured-jobs.rb', line 228 def setup_sidekiq_hooks return unless defined?(Sidekiq::VERSION) Sidekiq.configure_server do |config| config.server_middleware do |chain| chain.add AssuredJobs::Middleware end # Add startup hook for heartbeat and orphan recovery config.on(:startup) do # Ensure configuration is set up setup_defaults unless @instance_id logger.info "AssuredJobs starting up on instance #{instance_id}" # Start heartbeat system setup_heartbeat # Run orphan recovery on startup only (if enabled) if auto_recovery_enabled Thread.new do sleep 5 # Give the server a moment to fully start begin reenqueue_orphans! spinup_delayed_recovery_thread rescue => e logger.error "AssuredJobs startup orphan recovery failed: #{e.message}" logger.error e.backtrace.join("\n") end end else logger.info "AssuredJobs auto-recovery is disabled" end end # Add shutdown hook to clean up config.on(:shutdown) do logger.info "AssuredJobs shutting down instance #{instance_id}" begin # Stop heartbeat thread if @heartbeat_thread&.alive? @heartbeat_thread.kill @heartbeat_thread = nil end redis_sync do |conn| # Only clean up instance heartbeat - let orphan recovery handle job cleanup # This ensures that if there are running jobs during shutdown, they will be # detected as orphaned and recovered by the next instance conn.del(namespaced_key("instance:#{instance_id}")) # Log tracked jobs but don't clean them up - they should be recovered as orphans job_tracking_key = namespaced_key("jobs:#{instance_id}") tracked_jobs = conn.smembers(job_tracking_key) if tracked_jobs.any? logger.warn "AssuredJobs leaving #{tracked_jobs.size} tracked jobs for orphan recovery: #{tracked_jobs.join(', ')}" logger.info "AssuredJobs: These jobs will be recovered by the next instance startup" else logger.info "AssuredJobs: No tracked jobs to leave for recovery" end end rescue => e logger.error "AssuredJobs shutdown cleanup failed: #{e.message}" end end end end |