Merge pull request ceph#55979 from athanatos/sjust/wip-64546-max-creating-pgs

mon: always create pgs in the epoch in which the pool was created

Reviewed-by: Matan Breizman <[email protected]>

Author: athanatos

src/common/options/mgr.yaml.in

@@ -103,6 +103,13 @@ options:
   services:
   - mgr
   with_legacy: true
+- name: mgr_max_pg_creating
+  type: uint
+  level: advanced
+  desc: bound on max creating pgs when acting to create more pgs
+  default: 1024
+  services:
+  - mgr
 - name: mgr_module_path
   type: str
   level: advanced

src/common/options/mon.yaml.in

@@ -1258,14 +1258,6 @@ options:
   services:
   - mon
   with_legacy: true
-- name: mon_osd_max_creating_pgs
-  type: int
-  level: advanced
-  desc: maximum number of PGs the mon will create at once
-  default: 1024
-  services:
-  - mon
-  with_legacy: true
 - name: mon_osd_max_initial_pgs
   type: int
   level: advanced

src/mgr/DaemonServer.cc

@@ -2678,7 +2678,9 @@ void DaemonServer::send_report()
 void DaemonServer::adjust_pgs()
 {
   dout(20) << dendl;
-  unsigned max = std::max<int64_t>(1, g_conf()->mon_osd_max_creating_pgs);
+  uint64_t max = std::max<uint64_t>(
+    1,
+    g_conf().get_val<uint64_t>("mgr_max_pg_creating"));
   double max_misplaced = g_conf().get_val<double>("target_max_misplaced_ratio");
   bool aggro = g_conf().get_val<bool>("mgr_debug_aggressive_pg_num_changes");
 
@@ -2889,7 +2891,7 @@ void DaemonServer::adjust_pgs()
 		     << " pgp_num_target " << p.get_pgp_num_target()
 		     << " pgp_num " << p.get_pgp_num()
 		     << " - misplaced_ratio " << misplaced_ratio
-		     << " > max " << max_misplaced
+		     << " > max_misplaced " << max_misplaced
 		     << ", deferring pgp_num update" << dendl;
 	  } else {
 	    // NOTE: this calculation assumes objects are

src/mon/OSDMonitor.cc

@@ -1251,32 +1251,25 @@ OSDMonitor::update_pending_pgs(const OSDMap::Incremental& inc,
   }
 
   // process queue
-  unsigned max = std::max<int64_t>(1, g_conf()->mon_osd_max_creating_pgs);
   const auto total = pending_creatings.pgs.size();
-  while (pending_creatings.pgs.size() < max &&
-	 !pending_creatings.queue.empty()) {
+  while (!pending_creatings.queue.empty()) {
     auto p = pending_creatings.queue.begin();
     int64_t poolid = p->first;
     dout(10) << __func__ << " pool " << poolid
 	     << " created " << p->second.created
 	     << " modified " << p->second.modified
 	     << " [" << p->second.start << "-" << p->second.end << ")"
 	     << dendl;
-    int64_t n = std::min<int64_t>(max - pending_creatings.pgs.size(),
-				  p->second.end - p->second.start);
-    ps_t first = p->second.start;
-    ps_t end = first + n;
-    for (ps_t ps = first; ps < end; ++ps) {
+    for (ps_t ps = p->second.start; ps < p->second.end; ++ps) {
       const pg_t pgid{ps, static_cast<uint64_t>(poolid)};
-      // NOTE: use the *current* epoch as the PG creation epoch so that the
-      // OSD does not have to generate a long set of PastIntervals.
+      // The current epoch must be the pool creation epoch
       pending_creatings.pgs.emplace(
 	pgid,
-	creating_pgs_t::pg_create_info(inc.epoch,
+	creating_pgs_t::pg_create_info(p->second.created,
 				       p->second.modified));
       dout(10) << __func__ << " adding " << pgid << dendl;
     }
-    p->second.start = end;
+    p->second.start = p->second.end;
     if (p->second.done()) {
       dout(10) << __func__ << " done with queue for " << poolid << dendl;
       pending_creatings.queue.erase(p);