From: Casey Bodley <cbodley@redhat.com>
Date: Wed, 28 Jun 2023 21:14:16 +0000 (-0400)
Subject: rgw: fetch_remote_obj() preserves original part lengths for BlockDecrypt
X-Git-Tag: v17.2.7~116^2~5
X-Git-Url: http://git-server-git.apps.pok.os.sepia.ceph.com/?a=commitdiff_plain;h=26ce89e098b1dd76c80223dd9f9cc4d9944aa950;p=ceph.git

rgw: fetch_remote_obj() preserves original part lengths for BlockDecrypt

because multisite replicates multipart objects as a single part, we lose
information about the part sizes from the original manifest that is
necessary to correctly decrypt across those part boundaries

on replication, parse the part lengths out of the source object's
manifest, and store them in a separate RGW_ATTR_CRYPT_PARTS for use on
decryption

Fixes: https://tracker.ceph.com/issues/46062

Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit e7d98fb36994559b08bf0e2273d333d8fc125886)
---

diff --git a/src/rgw/rgw_common.h b/src/rgw/rgw_common.h
index 25e5b24ad817..afa82ddeb238 100644
--- a/src/rgw/rgw_common.h
+++ b/src/rgw/rgw_common.h
@@ -145,6 +145,7 @@ using ceph::crypto::MD5;
 #define RGW_ATTR_CRYPT_KEYSEL   RGW_ATTR_CRYPT_PREFIX "keysel"
 #define RGW_ATTR_CRYPT_CONTEXT  RGW_ATTR_CRYPT_PREFIX "context"
 #define RGW_ATTR_CRYPT_DATAKEY  RGW_ATTR_CRYPT_PREFIX "datakey"
+#define RGW_ATTR_CRYPT_PARTS    RGW_ATTR_CRYPT_PREFIX "part-lengths"
 
 /* SSE-S3 Encryption Attributes */
 #define RGW_ATTR_BUCKET_ENCRYPTION_PREFIX RGW_ATTR_PREFIX "sse-s3."
diff --git a/src/rgw/rgw_rados.cc b/src/rgw/rgw_rados.cc
index 6f5484cd21ae..66b2d8019877 100644
--- a/src/rgw/rgw_rados.cc
+++ b/src/rgw/rgw_rados.cc
@@ -44,6 +44,7 @@
 #include "rgw_tools.h"
 #include "rgw_coroutine.h"
 #include "rgw_compression.h"
+#include "rgw_crypt.h"
 #include "rgw_etag_verifier.h"
 #include "rgw_worker.h"
 #include "rgw_notify.h"
@@ -3485,11 +3486,28 @@ public:
           }
         }
       }
+
       /* We need the manifest to recompute the ETag for verification */
       iter = src_attrs.find(RGW_ATTR_MANIFEST);
       if (iter != src_attrs.end()) {
         manifest_bl = std::move(iter->second);
         src_attrs.erase(iter);
+
+        // if the source object was encrypted, preserve the original object's
+        // part lengths
+        if (src_attrs.count(RGW_ATTR_CRYPT_MODE)) {
+          std::vector<size_t> parts_len;
+          int r = RGWGetObj_BlockDecrypt::read_manifest_parts(dpp, manifest_bl,
+                                                              parts_len);
+          if (r < 0) {
+            ldpp_dout(dpp, 4) << "failed to read part lengths from the manifest" << dendl;
+          } else {
+            // store the encoded part lenghts in RGW_ATTR_CRYPT_PARTS
+            bufferlist parts_bl;
+            encode(parts_len, parts_bl);
+            src_attrs[RGW_ATTR_CRYPT_PARTS] = std::move(parts_bl);
+          }
+        }
       }
 
       // filter out olh attributes
diff --git a/src/rgw/rgw_rest_s3.cc b/src/rgw/rgw_rest_s3.cc
index 347deac7bf73..1667b4e3d739 100644
--- a/src/rgw/rgw_rest_s3.cc
+++ b/src/rgw/rgw_rest_s3.cc
@@ -585,10 +585,24 @@ int RGWGetObj_ObjStore_S3::get_decrypt_filter(std::unique_ptr<RGWGetObj_Filter>
   // in case of a multipart upload, we need to know the part lengths to
   // correctly decrypt across part boundaries
   std::vector<size_t> parts_len;
-  res = RGWGetObj_BlockDecrypt::read_manifest_parts(this, *manifest_bl,
-                                                    parts_len);
-  if (res < 0) {
-    return res;
+
+  // for replicated objects, the original part lengths are preserved in an xattr
+  if (auto i = attrs.find(RGW_ATTR_CRYPT_PARTS); i != attrs.end()) {
+    try {
+      auto p = i->second.cbegin();
+      using ceph::decode;
+      decode(parts_len, p);
+    } catch (const buffer::error&) {
+      ldpp_dout(this, 1) << "failed to decode RGW_ATTR_CRYPT_PARTS" << dendl;
+      return -EIO;
+    }
+  } else {
+    // otherwise, we read the part lengths from the manifest
+    res = RGWGetObj_BlockDecrypt::read_manifest_parts(this, *manifest_bl,
+                                                      parts_len);
+    if (res < 0) {
+      return res;
+    }
   }
 
   *filter = std::make_unique<RGWGetObj_BlockDecrypt>(
@@ -2705,10 +2719,24 @@ int RGWPutObj_ObjStore_S3::get_decrypt_filter(
   // in case of a multipart upload, we need to know the part lengths to
   // correctly decrypt across part boundaries
   std::vector<size_t> parts_len;
-  res = RGWGetObj_BlockDecrypt::read_manifest_parts(this, *manifest_bl,
-                                                    parts_len);
-  if (res < 0) {
-    return res;
+
+  // for replicated objects, the original part lengths are preserved in an xattr
+  if (auto i = attrs.find(RGW_ATTR_CRYPT_PARTS); i != attrs.end()) {
+    try {
+      auto p = i->second.cbegin();
+      using ceph::decode;
+      decode(parts_len, p);
+    } catch (const buffer::error&) {
+      ldpp_dout(this, 1) << "failed to decode RGW_ATTR_CRYPT_PARTS" << dendl;
+      return -EIO;
+    }
+  } else {
+    // otherwise, we read the part lengths from the manifest
+    res = RGWGetObj_BlockDecrypt::read_manifest_parts(this, *manifest_bl,
+                                                      parts_len);
+    if (res < 0) {
+      return res;
+    }
   }
 
   *filter = std::make_unique<RGWGetObj_BlockDecrypt>(