Merge pull request ceph#59725 from mchangir/mds-fallocate-return-EOPN…

…OTSUPP-for-mode-0

client: return EOPNOTSUPP for fallocate with mode 0

Reviewed-by: Venky Shankar <[email protected]>

PendingReleaseNotes

@@ -43,6 +43,10 @@
   there are alternative monitoring solutions, like `prometheus`, which is the most
   widely adopted among the Ceph user community.
 
+* CephFS: EOPNOTSUPP (Operation not supported ) is now returned by the CephFS
+  fuse client for `fallocate` for the default case (i.e. mode == 0) since
+  CephFS does not support disk space reservation. The only flags supported are
+  `FALLOC_FL_KEEP_SIZE` and `FALLOC_FL_PUNCH_HOLE`.
 
 >=19.0.0
 

qa/workunits/fs/misc/fallocate.sh

@@ -0,0 +1,17 @@
+#!/bin/sh -x
+
+# fallocate with mode 0 should fail with EOPNOTSUPP
+set -e
+mkdir -p testdir
+cd testdir
+
+expect_failure() {
+	if "$@"; then return 1; else return 0; fi
+}
+
+expect_failure fallocate -l 1M preallocated.txt
+rm -f preallocated.txt
+
+cd ..
+rmdir testdir
+echo OK

src/client/Client.cc

@@ -16234,7 +16234,7 @@ int Client::_fallocate(Fh *fh, int mode, int64_t offset, int64_t length)
   if (offset < 0 || length <= 0)
     return -CEPHFS_EINVAL;
 
-  if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE))
+  if (mode == 0 || (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE)))
     return -CEPHFS_EOPNOTSUPP;
 
   if ((mode & FALLOC_FL_PUNCH_HOLE) && !(mode & FALLOC_FL_KEEP_SIZE))

src/pybind/cephfs/cephfs.pyx

@@ -923,12 +923,12 @@ cdef class LibCephFS(object):
 
         :param fd: the file descriptor of the file to fallocate.
         :param mode: the flags determines the operation to be performed on the given
-                     range. default operation (0) allocate and initialize to zero
-                     the file in the byte range, and the file size will be changed
-                     if offset + length is greater than the file size. if the
-                     FALLOC_FL_KEEP_SIZE flag is specified in the mode, the file size
-                     will not be changed. if the FALLOC_FL_PUNCH_HOLE flag is specified
-                     in the mode, the operation is deallocate space and zero the byte range.
+                     range. default operation (0) is to return -EOPNOTSUPP since
+                     cephfs does not allocate disk blocks to provide write guarantees.
+                     if the FALLOC_FL_KEEP_SIZE flag is specified in the mode,
+                     the file size will not be changed.  if the FALLOC_FL_PUNCH_HOLE
+                     flag is specified in the mode, the operation is deallocate
+                     space and zero the byte range.
         :param offset: the byte range starting.
         :param length: the length of the range.
         """

src/test/pybind/test_cephfs.py

@@ -610,10 +610,10 @@ def test_ftruncate(testdir):
 def test_fallocate(testdir):
     fd = cephfs.open(b'/file-fallocate', 'w', 0o755)
     assert_raises(TypeError, cephfs.fallocate, b'/file-fallocate', 0, 10)
-    cephfs.fallocate(fd, 0, 10)
+    assert_raises(libcephfs.OperationNotSupported, cephfs.fallocate, fd, 0, 10)
     stat = cephfs.fsync(fd, 0)
     st = cephfs.fstat(fd)
-    assert_equal(st.st_size, 10)
+    assert_equal(st.st_size, 0)
     cephfs.close(fd)
     cephfs.unlink(b'/file-fallocate')