From: Sage Weil <sage@newdream.net>
Date: Tue, 24 May 2011 16:42:39 +0000 (-0700)
Subject: osd: make automatically marking of unfound as lost optional
X-Git-Tag: v0.28.2~6
X-Git-Url: http://git-server-git.apps.pok.os.sepia.ceph.com/?a=commitdiff_plain;h=87309e946bed4175433048276b13e94e844ef958;p=ceph.git

osd: make automatically marking of unfound as lost optional

We may not want to do this automatically until we have more confidense in
the recovery code.  Even then, possible not.  In particular, the OSDs may
believe they have contact all possible homes for the data even though there
is some long-lost OSD that has the data on disk that if offline.

For now, we make the marking process explicit so that the administrator can
make the call.

Signed-off-by: Sage Weil <sage@newdream.net>
---

diff --git a/src/common/config.cc b/src/common/config.cc
index 0794dcaa73fa..3e7431578070 100644
--- a/src/common/config.cc
+++ b/src/common/config.cc
@@ -361,6 +361,7 @@ struct config_option config_optionsp[] = {
   OPTION(osd_min_down_reports, OPT_INT, 3),     // number of times a down OSD must be reported for it to count
   OPTION(osd_replay_window, OPT_INT, 45),
   OPTION(osd_preserve_trimmed_log, OPT_BOOL, true),
+  OPTION(osd_auto_mark_unfound_lost, OPT_BOOL, false),
   OPTION(osd_recovery_delay_start, OPT_FLOAT, 15),
   OPTION(osd_recovery_max_active, OPT_INT, 5),
   OPTION(osd_recovery_max_chunk, OPT_U64, 1<<20),  // max size of push chunk
diff --git a/src/common/config.h b/src/common/config.h
index 76a0674d33cc..5e325af7222a 100644
--- a/src/common/config.h
+++ b/src/common/config.h
@@ -433,6 +433,7 @@ public:
   int   osd_min_down_reports;
   int   osd_replay_window;
   bool  osd_preserve_trimmed_log;
+  bool  osd_auto_mark_unfound_lost;
 
   float osd_recovery_delay_start;
   int osd_recovery_max_active;
diff --git a/src/osd/PG.cc b/src/osd/PG.cc
index 27e85f82227b..8e58d884dd78 100644
--- a/src/osd/PG.cc
+++ b/src/osd/PG.cc
@@ -4115,10 +4115,15 @@ PG::RecoveryState::Active::react(const ActMap&) {
   if (g_conf.osd_check_for_log_corruption)
     pg->check_log_for_corruption(pg->osd->store);
 
-  if (pg->missing.num_missing() > pg->missing_loc.size()) {
-    if (pg->all_unfound_are_lost(pg->osd->osdmap)) {
+  int unfound = pg->missing.num_missing() - pg->missing_loc.size();
+  if (unfound > 0 &&
+      pg->all_unfound_are_lost(pg->osd->osdmap)) {
+    if (g_conf.osd_auto_mark_unfound_lost) {
+      pg->osd->clog.error() << pg->info.pgid << " has " << unfound
+			    << " objects unfound and apparently lost, automatically marking lost\n";
       pg->mark_all_unfound_as_lost(*context< RecoveryMachine >().get_cur_transaction());
-    }
+    } else
+      pg->osd->clog.error() << pg->info.pgid << " has " << unfound << " objects unfound and apparently lost\n";
   }
 
   if (!pg->snap_trimq.empty() &&