rgw: bucket check --check-objects

A new option that goes through the indexed objects, verifies
their actual state and updates the index accordingly.

Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>

diff --git a/src/rgw/rgw_admin.cc b/src/rgw/rgw_admin.cc
index 665ac5516e86107f710093a9d56d261d73e99386..6a7d7915a692e259bb01099093218bf54c905e44 100644 (file)
--- a/src/rgw/rgw_admin.cc
+++ b/src/rgw/rgw_admin.cc
@@ -90,6 +90,8 @@ void _usage()
   cerr << "   --end-date=<date>\n";
   cerr << "   --bucket-id=<bucket-id>\n";
   cerr << "   --fix                     besides checking bucket index, will also fix it\n";
+  cerr << "   --check-objects           bucket check: rebuilds bucket index according to\n";
+  cerr << "                             actual objects state\n";
   cerr << "   --format=<format>         specify output format for certain operations: xml,\n";
   cerr << "                             json\n";
   cerr << "   --purge-data              when specified, user removal will also purge all the\n";
@@ -648,6 +650,13 @@ static int remove_bucket(RGWRados *store, rgw_bucket& bucket, bool delete_childr
   return ret;
 }
 
+static bool bucket_object_check_filter(const string& name)
+{
+  string ns;
+  string obj = name;
+  return rgw_obj::translate_raw_obj_to_obj_in_ns(obj, ns);
+}
+
 int main(int argc, char **argv) 
 {
   vector<const char*> args;
@@ -691,6 +700,7 @@ int main(int argc, char **argv)
   int max_buckets = -1;
   map<string, bool> categories;
   string caps;
+  int check_objects = false;
 
   std::string val;
   std::ostringstream errs;
@@ -786,6 +796,8 @@ int main(int argc, char **argv)
       // do nothing
     } else if (ceph_argparse_binary_flag(args, i, &fix, NULL, "--fix", (char*)NULL)) {
       // do nothing
+    } else if (ceph_argparse_binary_flag(args, i, &check_objects, NULL, "--check-objects", (char*)NULL)) {
+      // do nothing
     } else if (ceph_argparse_witharg(args, i, &val, "--caps", (char*)NULL)) {
       caps = val;
     } else {
@@ -1608,6 +1620,45 @@ next:
     map<RGWObjCategory, RGWBucketStats> existing_stats;
     map<RGWObjCategory, RGWBucketStats> calculated_stats;
 
+    if (check_objects) {
+      if (!fix) {
+       cerr << "--check-objects flag requires --fix" << std::endl;
+       return 1;
+      }
+#define BUCKET_TAG_TIMEOUT 30
+      cout << "Checking objects, decreasing bucket 2-phase commit timeout.\n"
+              "** Note that timeout will reset only when operation completes successfully **" << std::endl;
+
+      store->cls_obj_set_bucket_tag_timeout(bucket, BUCKET_TAG_TIMEOUT);
+
+      string prefix;
+      string marker;
+      bool is_truncated = true;
+
+      while (is_truncated) {
+       map<string, RGWObjEnt> result;
+       string ns;
+       int r = store->cls_bucket_list(bucket, marker, prefix, 1000, 
+                                   result, &is_truncated, &marker,
+                                    bucket_object_check_filter);
+
+       if (r < 0 && r != -ENOENT) {
+          cerr << "ERROR: failed operation r=" << r << std::endl;
+       }
+
+       if (r == -ENOENT)
+         break;
+
+       map<string, RGWObjEnt>::iterator iter;
+       for (iter = result.begin(); iter != result.end(); ++iter) {
+         cout << iter->first << std::endl;
+       }
+
+      }
+
+      store->cls_obj_set_bucket_tag_timeout(bucket, 0);
+
+    }
     int r = store->bucket_check_index(bucket, &existing_stats, &calculated_stats);
     if (r < 0) {
       cerr << "failed to check index err=" << cpp_strerror(-r) << std::endl;
@@ -1631,6 +1682,7 @@ next:
         return r;
       }
     }
+
   }
 
   if (opt_cmd == OPT_BUCKET_RM) {

diff --git a/src/rgw/rgw_rados.cc b/src/rgw/rgw_rados.cc
index f877883c7ebaa3fd7cf06a9ed3b3b823762103bc..70220b4176d0d14547f22f2fedb23e500864fa11 100644 (file)
--- a/src/rgw/rgw_rados.cc
+++ b/src/rgw/rgw_rados.cc
@@ -3076,9 +3076,27 @@ int RGWRados::cls_obj_complete_cancel(rgw_bucket& bucket, string& tag, string& n
   return cls_obj_complete_op(bucket, CLS_RGW_OP_ADD, tag, 0, ent, RGW_OBJ_CATEGORY_NONE, NULL);
 }
 
+int RGWRados::cls_obj_set_bucket_tag_timeout(rgw_bucket& bucket, uint64_t timeout)
+{
+  librados::IoCtx io_ctx;
+  string oid;
+
+  int r = open_bucket(bucket, io_ctx, oid);
+  if (r < 0)
+    return r;
+
+  ObjectWriteOperation o;
+  cls_rgw_bucket_set_tag_timeout(o, timeout);
+
+  r = io_ctx.operate(oid, &o);
+
+  return r;
+}
+
 int RGWRados::cls_bucket_list(rgw_bucket& bucket, string start, string prefix,
                              uint32_t num, map<string, RGWObjEnt>& m,
-                             bool *is_truncated, string *last_entry)
+                             bool *is_truncated, string *last_entry,
+                             bool (*force_check_filter)(const string&  name))
 {
   ldout(cct, 10) << "cls_bucket_list " << bucket << " start " << start << " num " << num << dendl;
 
@@ -3114,7 +3132,9 @@ int RGWRados::cls_bucket_list(rgw_bucket& bucket, string start, string prefix,
       continue;
     }
 
-    if (!dirent.exists || !dirent.pending_map.empty()) {
+    bool force_check = force_check_filter && force_check_filter(dirent.name);
+
+    if (!dirent.exists || !dirent.pending_map.empty() || force_check) {
       /* there are uncommitted ops. We need to check the current state,
        * and if the tags are old we need to do cleanup as well. */
       librados::IoCtx sub_ctx;

diff --git a/src/rgw/rgw_rados.h b/src/rgw/rgw_rados.h
index 50d8572f0819e66788102149069331e8b67664ed..e2ab4e244994fce53311c84418cca4bf8e6edfef 100644 (file)
--- a/src/rgw/rgw_rados.h
+++ b/src/rgw/rgw_rados.h
@@ -683,9 +683,10 @@ public:
   int cls_obj_complete_add(rgw_bucket& bucket, string& tag, uint64_t epoch, RGWObjEnt& ent, RGWObjCategory category, list<string> *remove_objs);
   int cls_obj_complete_del(rgw_bucket& bucket, string& tag, uint64_t epoch, string& name);
   int cls_obj_complete_cancel(rgw_bucket& bucket, string& tag, string& name);
+  int cls_obj_set_bucket_tag_timeout(rgw_bucket& bucket, uint64_t timeout);
   int cls_bucket_list(rgw_bucket& bucket, string start, string prefix, uint32_t num,
                       map<string, RGWObjEnt>& m, bool *is_truncated,
-                      string *last_entry = NULL);
+                      string *last_entry, bool (*force_check_filter)(const string&  name) = NULL);
   int cls_bucket_head(rgw_bucket& bucket, struct rgw_bucket_dir_header& header);
   int prepare_update_index(RGWObjState *state, rgw_bucket& bucket,
                            rgw_obj& oid, string& tag);

diff --git a/src/test/cli/radosgw-admin/help.t b/src/test/cli/radosgw-admin/help.t
index 4dd8634c28cbae75adc1c6a6a127347a6cae4df9..a9269ecb69142dc6126a89445e2c5c31f81ceb5c 100644 (file)
--- a/src/test/cli/radosgw-admin/help.t
+++ b/src/test/cli/radosgw-admin/help.t
@@ -56,6 +56,8 @@
      --end-date=<date>
      --bucket-id=<bucket-id>
      --fix                     besides checking bucket index, will also fix it
+     --check-objects           bucket check: rebuilds bucket index according to
+                               actual objects state
      --format=<format>         specify output format for certain operations: xml,
                                json
      --purge-data              when specified, user removal will also purge all the