rgw: add configurable AWS-compat invalid range get behavior

If rgw_ignore_get_invalid_range is set, treat invalid range
restrictions as a request for the full object.  By default, retain
the RGW behavior to fail with ERANGE.

Fixes: http://tracker.ceph.com/issues/24317
Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
(cherry picked from commit b8a3baffddb0f0082a9b250693d26d934eaf2650)
Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>

diff --git a/src/common/legacy_config_opts.h b/src/common/legacy_config_opts.h
index f664c96e4f9ae5f350753d97b995af6d2add58e6..51c91b0bd8dbbed7c05e62628c2a965161f4f7fb 100644 (file)
--- a/src/common/legacy_config_opts.h
+++ b/src/common/legacy_config_opts.h
@@ -1342,6 +1342,7 @@ OPTION(rgw_lc_max_objs, OPT_INT)
 OPTION(rgw_lc_debug_interval, OPT_INT)  // Debug run interval, in seconds
 OPTION(rgw_script_uri, OPT_STR) // alternative value for SCRIPT_URI if not set in request
 OPTION(rgw_request_uri, OPT_STR) // alternative value for REQUEST_URI if not set in request
+OPTION(rgw_ignore_get_invalid_range, OPT_BOOL) // treat invalid (e.g., negative) range requests as full
 OPTION(rgw_swift_url, OPT_STR)             // the swift url, being published by the internal swift auth
 OPTION(rgw_swift_url_prefix, OPT_STR) // entry point for which a url is considered a swift url
 OPTION(rgw_swift_auth_url, OPT_STR)        // default URL to go and verify tokens for v1 auth (if not using internal swift auth)

diff --git a/src/common/options.cc b/src/common/options.cc
index 5a81dda59ecdc4fb9897f02122954e5490b5278e..32c85f5308180837b76562150fb4f9c08e63198d 100644 (file)
--- a/src/common/options.cc
+++ b/src/common/options.cc
@@ -4539,6 +4539,12 @@ std::vector<Option> get_rgw_options() {
     .set_default("")
     .set_description(""),
 
+    Option("rgw_ignore_get_invalid_range", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+    .set_default(false)
+    .set_description("Treat invalid (e.g., negative) range request as full")
+    .set_long_description("Treat invalid (e.g., negative) range request "
+                         "as request for the full object (AWS compatibility)"),
+
     Option("rgw_swift_url", Option::TYPE_STR, Option::LEVEL_ADVANCED)
     .set_default("")
     .set_description("Swift-auth storage URL")

diff --git a/src/rgw/rgw_op.cc b/src/rgw/rgw_op.cc
index bfa76c4446089bad6bc60239804568ca6b49f297..ddd0f74844f54b5f3e3605b9edd4992e731141fb 100644 (file)
--- a/src/rgw/rgw_op.cc
+++ b/src/rgw/rgw_op.cc
@@ -76,41 +76,42 @@ static int forward_request_to_master(struct req_state *s, obj_version *objv, RGW
 
 static MultipartMetaFilter mp_filter;
 
-static int parse_range(const char *range, off_t& ofs, off_t& end, bool *partial_content)
+int RGWGetObj::parse_range(void)
 {
   int r = -ERANGE;
-  string s(range);
+  string rs(range_str);
   string ofs_str;
   string end_str;
 
-  *partial_content = false;
+  ignore_invalid_range = s->cct->_conf->rgw_ignore_get_invalid_range;
+  partial_content = false;
 
-  size_t pos = s.find("bytes=");
+  size_t pos = rs.find("bytes=");
   if (pos == string::npos) {
     pos = 0;
-    while (isspace(s[pos]))
+    while (isspace(rs[pos]))
       pos++;
     int end = pos;
-    while (isalpha(s[end]))
+    while (isalpha(rs[end]))
       end++;
-    if (strncasecmp(s.c_str(), "bytes", end - pos) != 0)
+    if (strncasecmp(rs.c_str(), "bytes", end - pos) != 0)
       return 0;
-    while (isspace(s[end]))
+    while (isspace(rs[end]))
       end++;
-    if (s[end] != '=')
+    if (rs[end] != '=')
       return 0;
-    s = s.substr(end + 1);
+    rs = rs.substr(end + 1);
   } else {
-    s = s.substr(pos + 6); /* size of("bytes=")  */
+    rs = rs.substr(pos + 6); /* size of("bytes=")  */
   }
-  pos = s.find('-');
+  pos = rs.find('-');
   if (pos == string::npos)
     goto done;
 
-  *partial_content = true;
+  partial_content = true;
 
-  ofs_str = s.substr(0, pos);
-  end_str = s.substr(pos + 1);
+  ofs_str = rs.substr(0, pos);
+  end_str = rs.substr(pos + 1);
   if (end_str.length()) {
     end = atoll(end_str.c_str());
     if (end < 0)
@@ -127,8 +128,18 @@ static int parse_range(const char *range, off_t& ofs, off_t& end, bool *partial_
   if (end >= 0 && end < ofs)
     goto done;
 
-  r = 0;
+  range_parsed = true;
+  return 0;
+
 done:
+  if (ignore_invalid_range) {
+    partial_content = false;
+    ofs = 0;
+    end = -1;
+    range_parsed = false; // allow retry
+    r = 0;
+  }
+
   return r;
 }
 
@@ -1588,16 +1599,13 @@ bool RGWGetObj::prefetch_data()
   bool prefetch_first_chunk = true;
   range_str = s->info.env->get("HTTP_RANGE");
 
-  if(range_str) {
-    int r = parse_range(range_str, ofs, end, &partial_content);
-    /* error on parsing the range, stop prefetch and will fail in execte() */
+  if (range_str) {
+    int r = parse_range();
+    /* error on parsing the range, stop prefetch and will fail in execute() */
     if (r < 0) {
-      range_parsed = false;
-      return false;
-    } else {
-      range_parsed = true;
+      return false; /* range_parsed==false */
     }
-    /* range get goes to shadown objects, stop prefetch */
+    /* range get goes to shadow objects, stop prefetch */
     if (ofs >= s->cct->_conf->rgw_max_chunk_size) {
       prefetch_first_chunk = false;
     }
@@ -1605,6 +1613,7 @@ bool RGWGetObj::prefetch_data()
 
   return get_data && prefetch_first_chunk;
 }
+
 void RGWGetObj::pre_exec()
 {
   rgw_bucket_object_pre_exec(s);
@@ -1812,9 +1821,9 @@ done_err:
 int RGWGetObj::init_common()
 {
   if (range_str) {
-    /* range parsed error when prefetch*/
+    /* range parsed error when prefetch */
     if (!range_parsed) {
-      int r = parse_range(range_str, ofs, end, &partial_content);
+      int r = parse_range();
       if (r < 0)
         return r;
     }

diff --git a/src/rgw/rgw_op.h b/src/rgw/rgw_op.h
index 0a127d2a9a46c2491a311d5c5c8e25f987a4426f..e24ca1ab31491b0d1e3883099c4a30d397bfdcfe 100644 (file)
--- a/src/rgw/rgw_op.h
+++ b/src/rgw/rgw_op.h
@@ -204,6 +204,7 @@ protected:
   map<string, bufferlist> attrs;
   bool get_data;
   bool partial_content;
+  bool ignore_invalid_range;
   bool range_parsed;
   bool skip_manifest;
   bool skip_decrypt{false};
@@ -257,9 +258,11 @@ public:
   void set_get_data(bool get_data) {
     this->get_data = get_data;
   }
+
   int verify_permission() override;
   void pre_exec() override;
   void execute() override;
+  int parse_range();
   int read_user_manifest_part(
     rgw_bucket& bucket,
     const rgw_bucket_dir_entry& ent,