Move GC into PHP and simplify it

Summary:
  - Move GC options into PHP.
  - Remove the "run at" and "run for" options. The GC daemon doesn't actually do any table scans, is very gentle, and runs for like 3 seconds per day in any normal install. Just limit it to running once every 4 hours when it's caught up and call it a day.

Test Plan: Edited GC options.

Reviewers: btrahan, codeblock

Reviewed By: codeblock

CC: aran

Maniphest Tasks: T2255

Differential Revision: https://secure.phabricator.com/D4321

Author: epriestley

conf/default.conf.php

@@ -1144,20 +1144,6 @@
   // manageable. To run garbage collection, launch a
   // PhabricatorGarbageCollector daemon.
 
-  // Since the GC daemon can issue large writes and table scans, you may want to
-  // run it only during off hours or make sure it is scheduled so it doesn't
-  // overlap with backups. This determines when the daemon can start running
-  // each day.
-  'gcdaemon.run-at'    => '12 AM',
-
-  // How many seconds after 'gcdaemon.run-at' the daemon may collect garbage
-  // for. By default it runs continuously, but you can set it to run for a
-  // limited period of time. For instance, if you do backups at 3 AM, you might
-  // run garbage collection for an hour beforehand. This is not a high-precision
-  // limit so you may want to leave some room for the GC to actually stop, and
-  // if you set it to something like 3 seconds you're on your own.
-  'gcdaemon.run-for'   => 24 * 60 * 60,
-
   // These 'ttl' keys configure how much old data the GC daemon keeps around.
   // Objects older than the ttl will be collected. Set any value to 0 to store
   // data indefinitely.

src/__phutil_library_map__.php

@@ -843,6 +843,7 @@
     'PhabricatorFlagQuery' => 'applications/flag/query/PhabricatorFlagQuery.php',
     'PhabricatorFlagsUIEventListener' => 'applications/flag/events/PhabricatorFlagsUIEventListener.php',
     'PhabricatorFormExample' => 'applications/uiexample/examples/PhabricatorFormExample.php',
+    'PhabricatorGarbageCollectorConfigOptions' => 'applications/config/option/PhabricatorGarbageCollectorConfigOptions.php',
     'PhabricatorGarbageCollectorDaemon' => 'infrastructure/daemon/PhabricatorGarbageCollectorDaemon.php',
     'PhabricatorGitGraphStream' => 'applications/repository/daemon/PhabricatorGitGraphStream.php',
     'PhabricatorGitHubConfigOptions' => 'applications/config/option/PhabricatorGitHubConfigOptions.php',
@@ -2180,6 +2181,7 @@
     'PhabricatorFlagListView' => 'AphrontView',
     'PhabricatorFlagsUIEventListener' => 'PhutilEventListener',
     'PhabricatorFormExample' => 'PhabricatorUIExample',
+    'PhabricatorGarbageCollectorConfigOptions' => 'PhabricatorApplicationConfigOptions',
     'PhabricatorGarbageCollectorDaemon' => 'PhabricatorDaemon',
     'PhabricatorGitHubConfigOptions' => 'PhabricatorApplicationConfigOptions',
     'PhabricatorGlobalLock' => 'PhutilLock',

src/applications/config/option/PhabricatorGarbageCollectorConfigOptions.php

@@ -0,0 +1,52 @@
+<?php
+
+final class PhabricatorGarbageCollectorConfigOptions
+  extends PhabricatorApplicationConfigOptions {
+
+  public function getName() {
+    return pht("Garbage Collector");
+  }
+
+  public function getDescription() {
+    return pht("Configure the GC for old logs, caches, etc.");
+  }
+
+  public function getOptions() {
+
+    $options = array(
+      'gcdaemon.ttl.herald-transcripts' => array(
+        30,
+        pht('Number of seconds to retain Herald transcripts for.')),
+      'gcdaemon.ttl.daemon-logs' => array(
+        14,
+        pht('Number of seconds to retain Daemon logs for.')),
+      'gcdaemon.ttl.differential-parse-cache' => array(
+        14,
+        pht('Number of seconds to retain Differential parse caches for.')),
+      'gcdaemon.ttl.markup-cache' => array(
+        30,
+        pht('Number of seconds to retain Markup cache entries for.')),
+      'gcdaemon.ttl.task-archive' => array(
+        14,
+        pht('Number of seconds to retain archived background tasks for.')),
+      'gcdaemon.ttl.general-cache' => array(
+        30,
+        pht('Number of seconds to retain general cache entries for.')),
+    );
+
+    $result = array();
+    foreach ($options as $key => $spec) {
+      list($default_days, $description) = $spec;
+      $result[] = $this
+        ->newOption($key, 'int', $default_days * (24 * 60 * 60))
+        ->setDescription($description)
+        ->addExample((7 * 24 * 60 * 60), pht('Retain for 1 week'))
+        ->addExample((14 * 24 * 60 * 60), pht('Retain for 2 weeks'))
+        ->addExample((30 * 24 * 60 * 60), pht('Retain for 30 days'))
+        ->addExample((60 * 24 * 60 * 60), pht('Retain for 60 days'))
+        ->addExample(0, pht('Retain indefinitely'));
+    }
+    return $result;
+  }
+
+}

src/infrastructure/daemon/PhabricatorGarbageCollectorDaemon.php

@@ -9,41 +9,7 @@
 final class PhabricatorGarbageCollectorDaemon extends PhabricatorDaemon {
 
   public function run() {
-
-    // Keep track of when we start and stop the GC so we can emit useful log
-    // messages.
-    $just_ran = false;
-
     do {
-      $run_at   = PhabricatorEnv::getEnvConfig('gcdaemon.run-at');
-      $run_for  = PhabricatorEnv::getEnvConfig('gcdaemon.run-for');
-
-      // Just use the default timezone, we don't need to get fancy and try
-      // to localize this.
-      $start = strtotime($run_at);
-      if ($start === false) {
-        throw new Exception(
-          "Configuration 'gcdaemon.run-at' could not be parsed: '{$run_at}'.");
-      }
-
-      $now = time();
-
-      if ($now < $start || $now > ($start + $run_for)) {
-        if ($just_ran) {
-          $this->log("Stopped garbage collector.");
-          $just_ran = false;
-        }
-        // The configuration says we can't collect garbage right now, so
-        // just sleep until we can.
-        $this->sleep(300);
-        continue;
-      }
-
-      if (!$just_ran) {
-        $this->log("Started garbage collector.");
-        $just_ran = true;
-      }
-
       $n_herald = $this->collectHeraldTranscripts();
       $n_daemon = $this->collectDaemonLogs();
       $n_parse  = $this->collectParseCaches();
@@ -70,8 +36,8 @@ public function run() {
       if ($total < 100) {
         // We didn't max out any of the GCs so we're basically caught up. Ease
         // off the GC loop so we don't keep doing table scans just to delete
-        // a handful of rows.
-        $this->sleep(300);
+        // a handful of rows; wake up in a few hours.
+        $this->sleep(4 * (60 * 60));
       } else {
         $this->stillWorking();
       }