samba-mirror/source3/modules/vfs_ceph.c

/*
   Unix SMB/CIFS implementation.
   Wrap disk only vfs functions to sidestep dodgy compilers.
   Copyright (C) Tim Potter 1998
   Copyright (C) Jeremy Allison 2007
   Copyright (C) Brian Chrisman 2011 <bchrisman@gmail.com>
   Copyright (C) Richard Sharpe 2011 <realrichardsharpe@gmail.com>

   This program is free software; you can redistribute it and/or modify
   it under the terms of the GNU General Public License as published by
   the Free Software Foundation; either version 3 of the License, or
   (at your option) any later version.

   This program is distributed in the hope that it will be useful,
   but WITHOUT ANY WARRANTY; without even the implied warranty of
   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
   GNU General Public License for more details.

   You should have received a copy of the GNU General Public License
   along with this program.  If not, see <http://www.gnu.org/licenses/>.
*/

/*
 * This VFS only works with the libceph.so user-space client. It is not needed
 * if you are using the kernel client or the FUSE client.
 *
 * Add the following smb.conf parameter to each share that will be hosted on
 * Ceph:
 *
 *   vfs objects = ceph [any others you need go here]
 */

#include "includes.h"
#include "smbd/smbd.h"
#include <dirent.h>
#include <sys/statvfs.h>
#include "cephfs/libcephfs.h"
#include "smbprofile.h"

#undef DBGC_CLASS
#define DBGC_CLASS DBGC_VFS

/*
 * Use %llu whenever we have a 64bit unsigned int, and cast to (long long unsigned)
 */
#define llu(_var) ((long long unsigned)_var)

/*
 * Note, libceph's return code model is to return -errno! So we have to convert
 * to what Samba expects, with is set errno to -return and return -1
 */
#define WRAP_RETURN(_res) \
	errno = 0; \
	if (_res < 0) { \
		errno = -_res; \
		return -1; \
	} \
	return _res \

/*
 * We mount only one file system and then all shares are assumed to be in that.
 * FIXME: If we want to support more than one FS, then we have to deal with
 * this differently.
 *
 * So, cmount tells us if we have been this way before and whether
 * we need to mount ceph and cmount_cnt tells us how many times we have
 * connected
 */
static struct ceph_mount_info * cmount = NULL;
static uint32_t cmount_cnt = 0;

/* Check for NULL pointer parameters in cephwrap_* functions */

/* We don't want to have NULL function pointers lying around.  Someone
   is sure to try and execute them.  These stubs are used to prevent
   this possibility. */

static int cephwrap_connect(struct vfs_handle_struct *handle,  const char *service, const char *user)
{
	int ret;
	char buf[256];

	const char * conf_file;

	if (cmount) {
		handle->data = cmount; /* We have been here before */
		cmount_cnt++;
		return 0;
	}

	conf_file = lp_parm_const_string(SNUM(handle->conn), "ceph", "config_file", NULL);

	DEBUG(2, ( "[CEPH] calling: ceph_create\n" ));
	ret = ceph_create(&cmount, NULL);
	if (ret)
		goto err_out;

	if (conf_file) {
		/* Override the config file */
		DEBUG(2, ( "[CEPH] calling: ceph_conf_read_file\n" ));
		ret = ceph_conf_read_file(cmount, conf_file);
	} else {

		DEBUG(2, ( "[CEPH] calling: ceph_conf_read_file with %s\n", conf_file));
		ret = ceph_conf_read_file(cmount, NULL);
	}

	if (ret)
		goto err_out;

	DEBUG(2, ( "[CEPH] calling: ceph_conf_get\n" ));
	ret = ceph_conf_get(cmount, "log file", buf, sizeof(buf));
	if (ret < 0)
		goto err_out;

	DEBUG(2, ("[CEPH] calling: ceph_mount\n"));
	ret = ceph_mount(cmount, NULL);
	if (ret < 0)
		goto err_out;


	/*
	 * encode mount context/state into our vfs/connection holding structure
	 * cmount is a ceph_mount_t*
	 */
	handle->data = cmount;
	cmount_cnt++;

	return 0;

err_out:
	/*
	 * Handle the error correctly. Ceph returns -errno.
	 */
	DEBUG(2, ("[CEPH] Error return: %s\n", strerror(-ret)));
	WRAP_RETURN(ret);
}

static void cephwrap_disconnect(struct vfs_handle_struct *handle)
{
	if (!cmount) {
		DEBUG(0, ("[CEPH] Error, ceph not mounted\n"));
		return;
	}

	/* Should we unmount/shutdown? Only if the last disconnect? */
	if (--cmount_cnt) {
		DEBUG(10, ("[CEPH] Not shuting down CEPH because still more connections\n"));
		return;
	}

	ceph_shutdown(cmount);

	cmount = NULL;  /* Make it safe */
}

/* Disk operations */

static uint64_t cephwrap_disk_free(struct vfs_handle_struct *handle,  const char *path, bool small_query, uint64_t *bsize,
			       uint64_t *dfree, uint64_t *dsize)
{
	struct statvfs statvfs_buf;
	int ret;

	if (!(ret = ceph_statfs(handle->data, path, &statvfs_buf))) {
		/*
		 * Provide all the correct values.
		 */
		*bsize = statvfs_buf.f_bsize;
		*dfree = statvfs_buf.f_bsize * statvfs_buf.f_bavail;
		*dsize = statvfs_buf.f_bsize * statvfs_buf.f_blocks;
		DEBUG(10, ("[CEPH] bsize: %llu, dfree: %llu, dsize: %llu\n",
			llu(*bsize), llu(*dfree), llu(*dsize)));
		return *dfree;
	} else {
		DEBUG(10, ("[CEPH] ceph_statfs returned %d\n", ret));
		WRAP_RETURN(ret);
	}
}

static int cephwrap_get_quota(struct vfs_handle_struct *handle,  enum SMB_QUOTA_TYPE qtype, unid_t id, SMB_DISK_QUOTA *qt)
{
	/* libceph: Ceph does not implement this */
#if 0
/* was ifdef HAVE_SYS_QUOTAS */
	int ret;

	ret = ceph_get_quota(handle->conn->connectpath, qtype, id, qt);

	if (ret) {
		errno = -ret;
		ret = -1;
	}

	return ret;
#else
	errno = ENOSYS;
	return -1;
#endif
}

static int cephwrap_set_quota(struct vfs_handle_struct *handle,  enum SMB_QUOTA_TYPE qtype, unid_t id, SMB_DISK_QUOTA *qt)
{
	/* libceph: Ceph does not implement this */
#if 0
/* was ifdef HAVE_SYS_QUOTAS */
	int ret;

	ret = ceph_set_quota(handle->conn->connectpath, qtype, id, qt);
	if (ret) {
		errno = -ret;
		ret = -1;
	}

	return ret;
#else
	WRAP_RETURN(-ENOSYS);
#endif
}

static int cephwrap_statvfs(struct vfs_handle_struct *handle,  const char *path, vfs_statvfs_struct *statbuf)
{
	struct statvfs statvfs_buf;
	int ret;

	ret = ceph_statfs(handle->data, path, &statvfs_buf);
	if (ret < 0) {
		WRAP_RETURN(ret);
	} else {
		statbuf->OptimalTransferSize = statvfs_buf.f_frsize;
		statbuf->BlockSize = statvfs_buf.f_bsize;
		statbuf->TotalBlocks = statvfs_buf.f_blocks;
		statbuf->BlocksAvail = statvfs_buf.f_bfree;
		statbuf->UserBlocksAvail = statvfs_buf.f_bavail;
		statbuf->TotalFileNodes = statvfs_buf.f_files;
		statbuf->FreeFileNodes = statvfs_buf.f_ffree;
		statbuf->FsIdentifier = statvfs_buf.f_fsid;
		DEBUG(10, ("[CEPH] f_bsize: %ld, f_blocks: %ld, f_bfree: %ld, f_bavail: %ld\n",
			(long int)statvfs_buf.f_bsize, (long int)statvfs_buf.f_blocks,
			(long int)statvfs_buf.f_bfree, (long int)statvfs_buf.f_bavail));
	}
	return ret;
}

/* Directory operations */

static DIR *cephwrap_opendir(struct vfs_handle_struct *handle,  const char *fname, const char *mask, uint32 attr)
{
	int ret = 0;
	struct ceph_dir_result *result;
	DEBUG(10, ("[CEPH] opendir(%p, %s)\n", handle, fname));

	/* Returns NULL if it does not exist or there are problems ? */
	ret = ceph_opendir(handle->data, fname, &result);
	if (ret < 0) {
		result = NULL;
		errno = -ret; /* We return result which is NULL in this case */
	}

	DEBUG(10, ("[CEPH] opendir(...) = %d\n", ret));
	return (DIR *) result;
}

static DIR *cephwrap_fdopendir(struct vfs_handle_struct *handle,
			       struct files_struct *fsp,
			       const char *mask,
			       uint32 attributes)
{
	int ret = 0;
	struct ceph_dir_result *result;
	DEBUG(10, ("[CEPH] fdopendir(%p, %p)\n", handle, fsp));

	ret = ceph_opendir(handle->data, fsp->fsp_name->base_name, &result);
	if (ret < 0) {
		result = NULL;
		errno = -ret; /* We return result which is NULL in this case */
	}

	DEBUG(10, ("[CEPH] fdopendir(...) = %d\n", ret));
	return (DIR *) result;
}

static struct dirent *cephwrap_readdir(struct vfs_handle_struct *handle,
				       DIR *dirp,
				       SMB_STRUCT_STAT *sbuf)
{
	struct dirent *result;

	DEBUG(10, ("[CEPH] readdir(%p, %p)\n", handle, dirp));
	result = ceph_readdir(handle->data, (struct ceph_dir_result *) dirp);
	DEBUG(10, ("[CEPH] readdir(...) = %p\n", result));

	/* Default Posix readdir() does not give us stat info.
	 * Set to invalid to indicate we didn't return this info. */
	if (sbuf)
		SET_STAT_INVALID(*sbuf);
	return result;
}

static void cephwrap_seekdir(struct vfs_handle_struct *handle, DIR *dirp, long offset)
{
	DEBUG(10, ("[CEPH] seekdir(%p, %p, %ld)\n", handle, dirp, offset));
	ceph_seekdir(handle->data, (struct ceph_dir_result *) dirp, offset);
}

static long cephwrap_telldir(struct vfs_handle_struct *handle, DIR *dirp)
{
	long ret;
	DEBUG(10, ("[CEPH] telldir(%p, %p)\n", handle, dirp));
	ret = ceph_telldir(handle->data, (struct ceph_dir_result *) dirp);
	DEBUG(10, ("[CEPH] telldir(...) = %ld\n", ret));
	WRAP_RETURN(ret);
}

static void cephwrap_rewinddir(struct vfs_handle_struct *handle, DIR *dirp)
{
	DEBUG(10, ("[CEPH] rewinddir(%p, %p)\n", handle, dirp));
	ceph_rewinddir(handle->data, (struct ceph_dir_result *) dirp);
}

static int cephwrap_mkdir(struct vfs_handle_struct *handle,  const char *path, mode_t mode)
{
	int result;
	bool has_dacl = False;
	char *parent = NULL;

	DEBUG(10, ("[CEPH] mkdir(%p, %s)\n", handle, path));

	if (lp_inherit_acls(SNUM(handle->conn))
	    && parent_dirname(talloc_tos(), path, &parent, NULL)
	    && (has_dacl = directory_has_default_acl(handle->conn, parent)))
		mode = 0777;

	TALLOC_FREE(parent);

	result = ceph_mkdir(handle->data, path, mode);

	/*
	 * Note. This order is important
	 */
	if (result) {
		WRAP_RETURN(result);
	} else if (result == 0 && !has_dacl) {
		/*
		 * We need to do this as the default behavior of POSIX ACLs
		 * is to set the mask to be the requested group permission
		 * bits, not the group permission bits to be the requested
		 * group permission bits. This is not what we want, as it will
		 * mess up any inherited ACL bits that were set. JRA.
		 */
		int saved_errno = errno; /* We may get ENOSYS */
		if ((SMB_VFS_CHMOD_ACL(handle->conn, path, mode) == -1) && (errno == ENOSYS))
			errno = saved_errno;
	}

	return result;
}

static int cephwrap_rmdir(struct vfs_handle_struct *handle,  const char *path)
{
	int result;

	DEBUG(10, ("[CEPH] rmdir(%p, %s)\n", handle, path));
	result = ceph_rmdir(handle->data, path);
	DEBUG(10, ("[CEPH] rmdir(...) = %d\n", result));
	WRAP_RETURN(result);
}

static int cephwrap_closedir(struct vfs_handle_struct *handle, DIR *dirp)
{
	int result;

	DEBUG(10, ("[CEPH] closedir(%p, %p)\n", handle, dirp));
	result = ceph_closedir(handle->data, (struct ceph_dir_result *) dirp);
	DEBUG(10, ("[CEPH] closedir(...) = %d\n", result));
	WRAP_RETURN(result);
}

/* File operations */

static int cephwrap_open(struct vfs_handle_struct *handle,
			struct smb_filename *smb_fname,
			files_struct *fsp, int flags, mode_t mode)
{
	int result = -ENOENT;
	DEBUG(10, ("[CEPH] open(%p, %s, %p, %d, %d)\n", handle, smb_fname_str_dbg(smb_fname), fsp, flags, mode));

	if (smb_fname->stream_name) {
		goto out;
	}

	result = ceph_open(handle->data, smb_fname->base_name, flags, mode);
out:
	DEBUG(10, ("[CEPH] open(...) = %d\n", result));
	WRAP_RETURN(result);
}

static int cephwrap_close(struct vfs_handle_struct *handle, files_struct *fsp)
{
	int result;

	DEBUG(10, ("[CEPH] close(%p, %p)\n", handle, fsp));
	result = ceph_close(handle->data, fsp->fh->fd);
	DEBUG(10, ("[CEPH] close(...) = %d\n", result));

	WRAP_RETURN(result);
}

static ssize_t cephwrap_read(struct vfs_handle_struct *handle, files_struct *fsp, void *data, size_t n)
{
	ssize_t result;

	DEBUG(10, ("[CEPH] read(%p, %p, %p, %llu)\n", handle, fsp, data, llu(n)));

	/* Using -1 for the offset means read/write rather than pread/pwrite */
	result = ceph_read(handle->data, fsp->fh->fd, data, n, -1);
	DEBUG(10, ("[CEPH] read(...) = %llu\n", llu(result)));
	WRAP_RETURN(result);
}

static ssize_t cephwrap_pread(struct vfs_handle_struct *handle, files_struct *fsp, void *data,
			size_t n, off_t offset)
{
	ssize_t result;

	DEBUG(10, ("[CEPH] pread(%p, %p, %p, %llu, %llu)\n", handle, fsp, data, llu(n), llu(offset)));

	result = ceph_read(handle->data, fsp->fh->fd, data, n, offset);
	DEBUG(10, ("[CEPH] pread(...) = %llu\n", llu(result)));
	WRAP_RETURN(result);
}


static ssize_t cephwrap_write(struct vfs_handle_struct *handle, files_struct *fsp, const void *data, size_t n)
{
	ssize_t result;

	DEBUG(10, ("[CEPH] write(%p, %p, %p, %llu)\n", handle, fsp, data, llu(n)));

	result = ceph_write(handle->data, fsp->fh->fd, data, n, -1);

	DEBUG(10, ("[CEPH] write(...) = %llu\n", llu(result)));
	if (result < 0) {
		WRAP_RETURN(result);
	}
	fsp->fh->pos += result;
	return result;
}

static ssize_t cephwrap_pwrite(struct vfs_handle_struct *handle, files_struct *fsp, const void *data,
			size_t n, off_t offset)
{
	ssize_t result;

	DEBUG(10, ("[CEPH] pwrite(%p, %p, %p, %llu, %llu)\n", handle, fsp, data, llu(n), llu(offset)));
	result = ceph_write(handle->data, fsp->fh->fd, data, n, offset);
	DEBUG(10, ("[CEPH] pwrite(...) = %llu\n", llu(result)));
	WRAP_RETURN(result);
}

static off_t cephwrap_lseek(struct vfs_handle_struct *handle, files_struct *fsp, off_t offset, int whence)
{
	off_t result = 0;

	DEBUG(10, ("[CEPH] cephwrap_lseek\n"));
	/* Cope with 'stat' file opens. */
	if (fsp->fh->fd != -1) {
		result = ceph_lseek(handle->data, fsp->fh->fd, offset, whence);
	}
	WRAP_RETURN(result);
}

static ssize_t cephwrap_sendfile(struct vfs_handle_struct *handle, int tofd, files_struct *fromfsp, const DATA_BLOB *hdr,
			off_t offset, size_t n)
{
	/*
	 * We cannot support sendfile because libceph is in user space.
	 */
	DEBUG(10, ("[CEPH] cephwrap_sendfile\n"));
	errno = ENOTSUP;
	return -1;
}

static ssize_t cephwrap_recvfile(struct vfs_handle_struct *handle,
			int fromfd,
			files_struct *tofsp,
			off_t offset,
			size_t n)
{
	/*
	 * We cannot support recvfile because libceph is in user space.
	 */
	DEBUG(10, ("[CEPH] cephwrap_recvfile\n"));
	errno=ENOTSUP;
	return -1;
}

static int cephwrap_rename(struct vfs_handle_struct *handle,
			  const struct smb_filename *smb_fname_src,
			  const struct smb_filename *smb_fname_dst)
{
	int result = -1;
	DEBUG(10, ("[CEPH] cephwrap_rename\n"));
	if (smb_fname_src->stream_name || smb_fname_dst->stream_name) {
		errno = ENOENT;
		return result;
	}

	result = ceph_rename(handle->data, smb_fname_src->base_name, smb_fname_dst->base_name);
	WRAP_RETURN(result);
}

static int cephwrap_fsync(struct vfs_handle_struct *handle, files_struct *fsp)
{
	int result;
	DEBUG(10, ("[CEPH] cephwrap_fsync\n"));
	result = ceph_fsync(handle->data, fsp->fh->fd, false);
	WRAP_RETURN(result);
}

static void cephwrap_init_stat_ex_from_stat(struct stat_ex *dst, const struct stat *src)
{
	ZERO_STRUCT(*dst);

	dst->st_ex_dev = src->st_dev;
	dst->st_ex_ino = src->st_ino;
	dst->st_ex_mode = src->st_mode;
	dst->st_ex_nlink = src->st_nlink;
	dst->st_ex_uid = src->st_uid;
	dst->st_ex_gid = src->st_gid;
	dst->st_ex_rdev = src->st_rdev;
	dst->st_ex_size = src->st_size;
	dst->st_ex_atime.tv_sec = src->st_atime;
	dst->st_ex_mtime.tv_sec = src->st_mtime;
	dst->st_ex_ctime.tv_sec = src->st_ctime;
	dst->st_ex_btime.tv_sec = src->st_mtime;
	dst->st_ex_blksize = src->st_blksize;
	dst->st_ex_blocks = src->st_blocks;
}

static int cephwrap_stat(struct vfs_handle_struct *handle,
			struct smb_filename *smb_fname)
{
	int result = -1;
	struct stat stbuf;

	DEBUG(10, ("[CEPH] stat(%p, %s)\n", handle, smb_fname_str_dbg(smb_fname)));

	if (smb_fname->stream_name) {
		errno = ENOENT;
		return result;
	}

	result = ceph_stat(handle->data, smb_fname->base_name, (struct stat *) &stbuf);
	DEBUG(10, ("[CEPH] stat(...) = %d\n", result));
	if (result < 0) {
		WRAP_RETURN(result);
	} else {
		DEBUG(10, ("[CEPH]\tstbuf = {dev = %llu, ino = %llu, mode = 0x%x, nlink = %llu, "
			   "uid = %d, gid = %d, rdev = %llu, size = %llu, blksize = %llu, "
			   "blocks = %llu, atime = %llu, mtime = %llu, ctime = %llu}\n",
			   llu(stbuf.st_dev), llu(stbuf.st_ino), stbuf.st_mode, llu(stbuf.st_nlink),
			   stbuf.st_uid, stbuf.st_gid, llu(stbuf.st_rdev), llu(stbuf.st_size), llu(stbuf.st_blksize),
			   llu(stbuf.st_blocks), llu(stbuf.st_atime), llu(stbuf.st_mtime), llu(stbuf.st_ctime)));
	}
	cephwrap_init_stat_ex_from_stat(&(smb_fname->st), &stbuf);
	DEBUG(10, ("[CEPH] mode = 0x%x\n", smb_fname->st.st_ex_mode));
	return result;
}

static int cephwrap_fstat(struct vfs_handle_struct *handle, files_struct *fsp, SMB_STRUCT_STAT *sbuf)
{
	int result = -1;
	struct stat stbuf;

	DEBUG(10, ("[CEPH] fstat(%p, %d)\n", handle, fsp->fh->fd));
	result = ceph_fstat(handle->data, fsp->fh->fd, (struct stat *) &stbuf);
	DEBUG(10, ("[CEPH] fstat(...) = %d\n", result));
	if (result < 0) {
		WRAP_RETURN(result);
	} else {
		DEBUG(10, ("[CEPH]\tstbuf = {dev = %llu, ino = %llu, mode = 0x%x, nlink = %llu, "
			   "uid = %d, gid = %d, rdev = %llu, size = %llu, blksize = %llu, "
			   "blocks = %llu, atime = %llu, mtime = %llu, ctime = %llu}\n",
			   llu(stbuf.st_dev), llu(stbuf.st_ino), stbuf.st_mode, llu(stbuf.st_nlink),
			   stbuf.st_uid, stbuf.st_gid, llu(stbuf.st_rdev), llu(stbuf.st_size), llu(stbuf.st_blksize),
			   llu(stbuf.st_blocks), llu(stbuf.st_atime), llu(stbuf.st_mtime), llu(stbuf.st_ctime)));
	}

	cephwrap_init_stat_ex_from_stat(sbuf, &stbuf);
	DEBUG(10, ("[CEPH] mode = 0x%x\n", sbuf->st_ex_mode));
	return result;
}

static int cephwrap_lstat(struct vfs_handle_struct *handle,
			 struct smb_filename *smb_fname)
{
	int result = -1;
	struct stat stbuf;

	DEBUG(10, ("[CEPH] lstat(%p, %s)\n", handle, smb_fname_str_dbg(smb_fname)));

	if (smb_fname->stream_name) {
		errno = ENOENT;
		return result;
	}

	result = ceph_lstat(handle->data, smb_fname->base_name, &stbuf);
	DEBUG(10, ("[CEPH] lstat(...) = %d\n", result));
	if (result < 0) {
		WRAP_RETURN(result);
	}
	cephwrap_init_stat_ex_from_stat(&(smb_fname->st), &stbuf);
	return result;
}

static int cephwrap_unlink(struct vfs_handle_struct *handle,
			  const struct smb_filename *smb_fname)
{
	int result = -1;

	DEBUG(10, ("[CEPH] unlink(%p, %s)\n", handle, smb_fname_str_dbg(smb_fname)));
	if (smb_fname->stream_name) {
		errno = ENOENT;
		return result;
	}
	result = ceph_unlink(handle->data, smb_fname->base_name);
	DEBUG(10, ("[CEPH] unlink(...) = %d\n", result));
	WRAP_RETURN(result);
}

static int cephwrap_chmod(struct vfs_handle_struct *handle,  const char *path, mode_t mode)
{
	int result;

	DEBUG(10, ("[CEPH] chmod(%p, %s, %d)\n", handle, path, mode));

	/*
	 * We need to do this due to the fact that the default POSIX ACL
	 * chmod modifies the ACL *mask* for the group owner, not the
	 * group owner bits directly. JRA.
	 */


	{
		int saved_errno = errno; /* We might get ENOSYS */
		if ((result = SMB_VFS_CHMOD_ACL(handle->conn, path, mode)) == 0) {
			return result;
		}
		/* Error - return the old errno. */
		errno = saved_errno;
	}

	result = ceph_chmod(handle->data, path, mode);
	DEBUG(10, ("[CEPH] chmod(...) = %d\n", result));
	WRAP_RETURN(result);
}

static int cephwrap_fchmod(struct vfs_handle_struct *handle, files_struct *fsp, mode_t mode)
{
	int result;

	DEBUG(10, ("[CEPH] fchmod(%p, %p, %d)\n", handle, fsp, mode));

	/*
	 * We need to do this due to the fact that the default POSIX ACL
	 * chmod modifies the ACL *mask* for the group owner, not the
	 * group owner bits directly. JRA.
	 */

	{
		int saved_errno = errno; /* We might get ENOSYS */
		if ((result = SMB_VFS_FCHMOD_ACL(fsp, mode)) == 0) {
			return result;
		}
		/* Error - return the old errno. */
		errno = saved_errno;
	}

#if defined(HAVE_FCHMOD)
	result = ceph_fchmod(handle->data, fsp->fh->fd, mode);
	DEBUG(10, ("[CEPH] fchmod(...) = %d\n", result));
	WRAP_RETURN(result);
#else
	errno = ENOSYS;
#endif
	return -1;
}

static int cephwrap_chown(struct vfs_handle_struct *handle, const char *path, uid_t uid, gid_t gid)
{
	int result;
	DEBUG(10, ("[CEPH] chown(%p, %s, %d, %d)\n", handle, path, uid, gid));
	result = ceph_chown(handle->data, path, uid, gid);
	DEBUG(10, ("[CEPH] chown(...) = %d\n", result));
	WRAP_RETURN(result);
}

static int cephwrap_fchown(struct vfs_handle_struct *handle, files_struct *fsp, uid_t uid, gid_t gid)
{
	int result;
#ifdef HAVE_FCHOWN

	DEBUG(10, ("[CEPH] fchown(%p, %p, %d, %d)\n", handle, fsp, uid, gid));
	result = ceph_fchown(handle->data, fsp->fh->fd, uid, gid);
	DEBUG(10, ("[CEPH] fchown(...) = %d\n", result));
	WRAP_RETURN(result);
#else
	errno = ENOSYS;
	result = -1;
#endif
	return result;
}

static int cephwrap_lchown(struct vfs_handle_struct *handle, const char *path, uid_t uid, gid_t gid)
{
	int result;

	DEBUG(10, ("[CEPH] lchown(%p, %s, %d, %d)\n", handle, path, uid, gid));
	result = ceph_lchown(handle->data, path, uid, gid);
	DEBUG(10, ("[CEPH] lchown(...) = %d\n", result));
	WRAP_RETURN(result);
}

static int cephwrap_chdir(struct vfs_handle_struct *handle,  const char *path)
{
	int result = -1;
	DEBUG(10, ("[CEPH] chdir(%p, %s)\n", handle, path));
	/*
	 * If the path is just / use chdir because Ceph is below / and
	 * cannot deal with changing directory above its mount point
	 */
	if (path && !strcmp(path, "/"))
		return chdir(path);

	result = ceph_chdir(handle->data, path);
	DEBUG(10, ("[CEPH] chdir(...) = %d\n", result));
	WRAP_RETURN(result);
}

static char *cephwrap_getwd(struct vfs_handle_struct *handle)
{
	const char *cwd = ceph_getcwd(handle->data);
	DEBUG(10, ("[CEPH] getwd(%p) = %s\n", handle, cwd));
	return SMB_STRDUP(cwd);
}

static int cephwrap_ntimes(struct vfs_handle_struct *handle,
			 const struct smb_filename *smb_fname,
			 struct smb_file_time *ft)
{
	struct utimbuf buf;
	buf.actime = ft->atime.tv_sec;
	buf.modtime = ft->mtime.tv_sec;
	int result = ceph_utime(handle->data, smb_fname->base_name, &buf);
	DEBUG(10, ("[CEPH] ntimes(%p, %s, {%ld, %ld, %ld, %ld}) = %d\n", handle, smb_fname_str_dbg(smb_fname),
				ft->mtime.tv_sec, ft->atime.tv_sec, ft->ctime.tv_sec,
				ft->create_time.tv_sec, result));
	return result;
}

static int strict_allocate_ftruncate(struct vfs_handle_struct *handle, files_struct *fsp, off_t len)
{
	off_t space_to_write;
	uint64_t space_avail;
	uint64_t bsize,dfree,dsize;
	int ret;
	NTSTATUS status;
	SMB_STRUCT_STAT *pst;

	status = vfs_stat_fsp(fsp);
	if (!NT_STATUS_IS_OK(status)) {
		return -1;
	}
	pst = &fsp->fsp_name->st;

#ifdef S_ISFIFO
	if (S_ISFIFO(pst->st_ex_mode))
		return 0;
#endif

	if (pst->st_ex_size == len)
		return 0;

	/* Shrink - just ftruncate. */
	if (pst->st_ex_size > len)
		return ftruncate(fsp->fh->fd, len);

	space_to_write = len - pst->st_ex_size;

	/* for allocation try fallocate first. This can fail on some
	   platforms e.g. when the filesystem doesn't support it and no
	   emulation is being done by the libc (like on AIX with JFS1). In that
	   case we do our own emulation. fallocate implementations can
	   return ENOTSUP or EINVAL in cases like that. */
	ret = SMB_VFS_FALLOCATE(fsp, VFS_FALLOCATE_EXTEND_SIZE,
				pst->st_ex_size, space_to_write);
	if (ret == ENOSPC) {
		errno = ENOSPC;
		return -1;
	}
	if (ret == 0) {
		return 0;
	}
	DEBUG(10,("[CEPH] strict_allocate_ftruncate: SMB_VFS_FALLOCATE failed with "
		"error %d. Falling back to slow manual allocation\n", ret));

	/* available disk space is enough or not? */
	space_avail = get_dfree_info(fsp->conn,
				     fsp->fsp_name->base_name, false,
				     &bsize,&dfree,&dsize);
	/* space_avail is 1k blocks */
	if (space_avail == (uint64_t)-1 ||
			((uint64_t)space_to_write/1024 > space_avail) ) {
		errno = ENOSPC;
		return -1;
	}

	/* Write out the real space on disk. */
	ret = vfs_slow_fallocate(fsp, pst->st_ex_size, space_to_write);
	if (ret != 0) {
		errno = ret;
		ret = -1;
	}

	return 0;
}

static int cephwrap_ftruncate(struct vfs_handle_struct *handle, files_struct *fsp, off_t len)
{
	int result = -1;
	SMB_STRUCT_STAT st;
	char c = 0;
	off_t currpos;

	DEBUG(10, ("[CEPH] ftruncate(%p, %p, %llu\n", handle, fsp, llu(len)));

	if (lp_strict_allocate(SNUM(fsp->conn))) {
		result = strict_allocate_ftruncate(handle, fsp, len);
		return result;
	}

	/* we used to just check HAVE_FTRUNCATE_EXTEND and only use
	   sys_ftruncate if the system supports it. Then I discovered that
	   you can have some filesystems that support ftruncate
	   expansion and some that don't! On Linux fat can't do
	   ftruncate extend but ext2 can. */

	result = ceph_ftruncate(handle->data, fsp->fh->fd, len);
	if (result == 0)
		goto done;

	/* According to W. R. Stevens advanced UNIX prog. Pure 4.3 BSD cannot
	   extend a file with ftruncate. Provide alternate implementation
	   for this */
	currpos = SMB_VFS_LSEEK(fsp, 0, SEEK_CUR);
	if (currpos == -1) {
		goto done;
	}

	/* Do an fstat to see if the file is longer than the requested
	   size in which case the ftruncate above should have
	   succeeded or shorter, in which case seek to len - 1 and
	   write 1 byte of zero */
	if (SMB_VFS_FSTAT(fsp, &st) == -1) {
		goto done;
	}

#ifdef S_ISFIFO
	if (S_ISFIFO(st.st_ex_mode)) {
		result = 0;
		goto done;
	}
#endif

	if (st.st_ex_size == len) {
		result = 0;
		goto done;
	}

	if (st.st_ex_size > len) {
		/* the sys_ftruncate should have worked */
		goto done;
	}

	if (SMB_VFS_LSEEK(fsp, len-1, SEEK_SET) != len -1)
		goto done;

	if (SMB_VFS_WRITE(fsp, &c, 1)!=1)
		goto done;

	/* Seek to where we were */
	if (SMB_VFS_LSEEK(fsp, currpos, SEEK_SET) != currpos)
		goto done;
	result = 0;

  done:

	return result;
}

static bool cephwrap_lock(struct vfs_handle_struct *handle, files_struct *fsp, int op, off_t offset, off_t count, int type)
{
	DEBUG(10, ("[CEPH] lock\n"));
	return true;
}

static int cephwrap_kernel_flock(struct vfs_handle_struct *handle, files_struct *fsp,
				uint32 share_mode, uint32 access_mask)
{
	DEBUG(10, ("[CEPH] kernel_flock\n"));
	/*
	 * We must return zero here and pretend all is good.
	 * One day we might have this in CEPH.
	 */
	return 0;
}

static bool cephwrap_getlock(struct vfs_handle_struct *handle, files_struct *fsp, off_t *poffset, off_t *pcount, int *ptype, pid_t *ppid)
{
	DEBUG(10, ("[CEPH] getlock returning false and errno=0\n"));

	errno = 0;
	return false;
}

/*
 * We cannot let this fall through to the default, because the file might only
 * be accessible from libceph (which is a user-space client) but the fd might
 * be for some file the kernel knows about.
 */
static int cephwrap_linux_setlease(struct vfs_handle_struct *handle, files_struct *fsp,
				int leasetype)
{
	int result = -1;

	DEBUG(10, ("[CEPH] linux_setlease\n"));
	errno = ENOSYS;
	return result;
}

static int cephwrap_symlink(struct vfs_handle_struct *handle,  const char *oldpath, const char *newpath)
{
	int result = -1;
	DEBUG(10, ("[CEPH] symlink(%p, %s, %s)\n", handle, oldpath, newpath));
	result = ceph_symlink(handle->data, oldpath, newpath);
	DEBUG(10, ("[CEPH] symlink(...) = %d\n", result));
	WRAP_RETURN(result);
}

static int cephwrap_readlink(struct vfs_handle_struct *handle,  const char *path, char *buf, size_t bufsiz)
{
	int result = -1;
	DEBUG(10, ("[CEPH] readlink(%p, %s, %p, %llu)\n", handle, path, buf, llu(bufsiz)));
	result = ceph_readlink(handle->data, path, buf, bufsiz);
	DEBUG(10, ("[CEPH] readlink(...) = %d\n", result));
	WRAP_RETURN(result);
}

static int cephwrap_link(struct vfs_handle_struct *handle,  const char *oldpath, const char *newpath)
{
	int result = -1;
	DEBUG(10, ("[CEPH] link(%p, %s, %s)\n", handle, oldpath, newpath));
	result = ceph_link(handle->data, oldpath, newpath);
	DEBUG(10, ("[CEPH] link(...) = %d\n", result));
	WRAP_RETURN(result);
}

static int cephwrap_mknod(struct vfs_handle_struct *handle,  const char *pathname, mode_t mode, SMB_DEV_T dev)
{
	int result = -1;
	DEBUG(10, ("[CEPH] mknod(%p, %s)\n", handle, pathname));
	result = ceph_mknod(handle->data, pathname, mode, dev);
	DEBUG(10, ("[CEPH] mknod(...) = %d\n", result));
	WRAP_RETURN(result);
}

/*
 * This is a simple version of real-path ... a better version is needed to
 * ask libceph about symbolic links.
 */
static char *cephwrap_realpath(struct vfs_handle_struct *handle,  const char *path)
{
	char *result;
	size_t len = strlen(path);

	result = SMB_MALLOC_ARRAY(char, PATH_MAX+1);
	if (len && (path[0] == '/')) {
		int r = asprintf(&result, "%s", path);
		if (r < 0) return NULL;
	} else if ((len >= 2) && (path[0] == '.') && (path[1] == '/')) {
		if (len == 2) {
			int r = asprintf(&result, "%s",
					handle->conn->connectpath);
			if (r < 0) return NULL;
		} else {
			int r = asprintf(&result, "%s/%s",
					handle->conn->connectpath, &path[2]);
			if (r < 0) return NULL;
		}
	} else {
		int r = asprintf(&result, "%s/%s",
				handle->conn->connectpath, path);
		if (r < 0) return NULL;
	}
	DEBUG(10, ("[CEPH] realpath(%p, %s) = %s\n", handle, path, result));
	return result;
}

static NTSTATUS cephwrap_notify_watch(struct vfs_handle_struct *vfs_handle,
				     struct sys_notify_context *ctx,
				     const char *path,
				     uint32_t *filter,
				     uint32_t *subdir_filter,
				     void (*callback)(struct sys_notify_context *ctx,
						      void *private_data,
						      struct notify_event *ev),
				     void *private_data,
				     void *handle_p)
{
	/*
	 * We cannot call inotify on files the kernel does not know about
	 */
	return NT_STATUS_OK;
}

static int cephwrap_chflags(struct vfs_handle_struct *handle, const char *path,
			   unsigned int flags)
{
	errno = ENOSYS;
	return -1;
}

static int cephwrap_get_real_filename(struct vfs_handle_struct *handle,
				     const char *path,
				     const char *name,
				     TALLOC_CTX *mem_ctx,
				     char **found_name)
{
	/*
	 * Don't fall back to get_real_filename so callers can differentiate
	 * between a full directory scan and an actual case-insensitive stat.
	 */
	errno = EOPNOTSUPP;
	return -1;
}

static const char *cephwrap_connectpath(struct vfs_handle_struct *handle,
				       const char *fname)
{
	return handle->conn->connectpath;
}

/****************************************************************
 Extended attribute operations.
*****************************************************************/

static ssize_t cephwrap_getxattr(struct vfs_handle_struct *handle,const char *path, const char *name, void *value, size_t size)
{
	DEBUG(10, ("[CEPH] getxattr(%p, %s, %s, %p, %llu)\n", handle, path, name, value, llu(size)));
	int ret = ceph_getxattr(handle->data, path, name, value, size);
	DEBUG(10, ("[CEPH] getxattr(...) = %d\n", ret));
	if (ret < 0) {
		WRAP_RETURN(ret);
	} else {
		return (ssize_t)ret;
	}
}

static ssize_t cephwrap_fgetxattr(struct vfs_handle_struct *handle, struct files_struct *fsp, const char *name, void *value, size_t size)
{
	DEBUG(10, ("[CEPH] fgetxattr(%p, %p, %s, %p, %llu)\n", handle, fsp, name, value, llu(size)));
	int ret = ceph_getxattr(handle->data, fsp->fsp_name->base_name, name, value, size);
	DEBUG(10, ("[CEPH] fgetxattr(...) = %d\n", ret));
	if (ret < 0) {
		WRAP_RETURN(ret);
	} else {
		return (ssize_t)ret;
	}
}

static ssize_t cephwrap_listxattr(struct vfs_handle_struct *handle, const char *path, char *list, size_t size)
{
	DEBUG(10, ("[CEPH] listxattr(%p, %s, %p, %llu)\n", handle, path, list, llu(size)));
	int ret = ceph_listxattr(handle->data, path, list, size);
	DEBUG(10, ("[CEPH] listxattr(...) = %d\n", ret));
	if (ret < 0) {
		WRAP_RETURN(ret);
	} else {
		return (ssize_t)ret;
	}
}

#if 0
static ssize_t cephwrap_llistxattr(struct vfs_handle_struct *handle, const char *path, char *list, size_t size)
{
	DEBUG(10, ("[CEPH] llistxattr(%p, %s, %p, %llu)\n", handle, path, list, llu(size)));
	int ret = ceph_llistxattr(handle->data, path, list, size);
	DEBUG(10, ("[CEPH] listxattr(...) = %d\n", ret));
	if (ret < 0) {
		WRAP_RETURN(ret);
	} else {
		return (ssize_t)ret;
	}
}
#endif

static ssize_t cephwrap_flistxattr(struct vfs_handle_struct *handle, struct files_struct *fsp, char *list, size_t size)
{
	DEBUG(10, ("[CEPH] flistxattr(%p, %p, %s, %llu)\n", handle, fsp, list, llu(size)));
	int ret = ceph_listxattr(handle->data, fsp->fsp_name->base_name, list, size);
	DEBUG(10, ("[CEPH] flistxattr(...) = %d\n", ret));
	if (ret < 0) {
		WRAP_RETURN(ret);
	} else {
		return (ssize_t)ret;
	}
}

static int cephwrap_removexattr(struct vfs_handle_struct *handle, const char *path, const char *name)
{
	DEBUG(10, ("[CEPH] removexattr(%p, %s, %s)\n", handle, path, name));
	int ret = ceph_removexattr(handle->data, path, name);
	DEBUG(10, ("[CEPH] removexattr(...) = %d\n", ret));
	WRAP_RETURN(ret);
}

static int cephwrap_fremovexattr(struct vfs_handle_struct *handle, struct files_struct *fsp, const char *name)
{
	DEBUG(10, ("[CEPH] fremovexattr(%p, %p, %s)\n", handle, fsp, name));
	int ret = ceph_removexattr(handle->data, fsp->fsp_name->base_name, name);
	DEBUG(10, ("[CEPH] fremovexattr(...) = %d\n", ret));
	WRAP_RETURN(ret);
}

static int cephwrap_setxattr(struct vfs_handle_struct *handle, const char *path, const char *name, const void *value, size_t size, int flags)
{
	DEBUG(10, ("[CEPH] setxattr(%p, %s, %s, %p, %llu, %d)\n", handle, path, name, value, llu(size), flags));
	int ret = ceph_setxattr(handle->data, path, name, value, size, flags);
	DEBUG(10, ("[CEPH] setxattr(...) = %d\n", ret));
	WRAP_RETURN(ret);
}

static int cephwrap_fsetxattr(struct vfs_handle_struct *handle, struct files_struct *fsp, const char *name, const void *value, size_t size, int flags)
{
	DEBUG(10, ("[CEPH] fsetxattr(%p, %p, %s, %p, %llu, %d)\n", handle, fsp, name, value, llu(size), flags));
	int ret = ceph_setxattr(handle->data, fsp->fsp_name->base_name, name, value, size, flags);
	DEBUG(10, ("[CEPH] fsetxattr(...) = %d\n", ret));
	WRAP_RETURN(ret);
}

static bool cephwrap_aio_force(struct vfs_handle_struct *handle, struct files_struct *fsp)
{

	/*
	 * We do not support AIO yet.
	 */

	DEBUG(10, ("[CEPH] cephwrap_aio_force(%p, %p) = false (errno = ENOTSUP)\n", handle, fsp));
	errno = ENOTSUP;
	return false;
}

static bool cephwrap_is_offline(struct vfs_handle_struct *handle,
				const struct smb_filename *fname,
				SMB_STRUCT_STAT *sbuf)
{
	return false;
}

static int cephwrap_set_offline(struct vfs_handle_struct *handle,
				const struct smb_filename *fname)
{
	errno = ENOTSUP;
	return -1;
}

static struct vfs_fn_pointers ceph_fns = {
	/* Disk operations */

	.connect_fn = cephwrap_connect,
	.disconnect_fn = cephwrap_disconnect,
	.disk_free_fn = cephwrap_disk_free,
	.get_quota_fn = cephwrap_get_quota,
	.set_quota_fn = cephwrap_set_quota,
	.statvfs_fn = cephwrap_statvfs,

	/* Directory operations */

	.opendir_fn = cephwrap_opendir,
	.fdopendir_fn = cephwrap_fdopendir,
	.readdir_fn = cephwrap_readdir,
	.seekdir_fn = cephwrap_seekdir,
	.telldir_fn = cephwrap_telldir,
	.rewind_dir_fn = cephwrap_rewinddir,
	.mkdir_fn = cephwrap_mkdir,
	.rmdir_fn = cephwrap_rmdir,
	.closedir_fn = cephwrap_closedir,

	/* File operations */

	.open_fn = cephwrap_open,
	.close_fn = cephwrap_close,
	.read_fn = cephwrap_read,
	.pread_fn = cephwrap_pread,
	.write_fn = cephwrap_write,
	.pwrite_fn = cephwrap_pwrite,
	.lseek_fn = cephwrap_lseek,
	.sendfile_fn = cephwrap_sendfile,
	.recvfile_fn = cephwrap_recvfile,
	.rename_fn = cephwrap_rename,
	.fsync_fn = cephwrap_fsync,
	.stat_fn = cephwrap_stat,
	.fstat_fn = cephwrap_fstat,
	.lstat_fn = cephwrap_lstat,
	.unlink_fn = cephwrap_unlink,
	.chmod_fn = cephwrap_chmod,
	.fchmod_fn = cephwrap_fchmod,
	.chown_fn = cephwrap_chown,
	.fchown_fn = cephwrap_fchown,
	.lchown_fn = cephwrap_lchown,
	.chdir_fn = cephwrap_chdir,
	.getwd_fn = cephwrap_getwd,
	.ntimes_fn = cephwrap_ntimes,
	.ftruncate_fn = cephwrap_ftruncate,
	.lock_fn = cephwrap_lock,
	.kernel_flock_fn = cephwrap_kernel_flock,
	.linux_setlease_fn = cephwrap_linux_setlease,
	.getlock_fn = cephwrap_getlock,
	.symlink_fn = cephwrap_symlink,
	.readlink_fn = cephwrap_readlink,
	.link_fn = cephwrap_link,
	.mknod_fn = cephwrap_mknod,
	.realpath_fn = cephwrap_realpath,
	.notify_watch_fn = cephwrap_notify_watch,
	.chflags_fn = cephwrap_chflags,
	.get_real_filename_fn = cephwrap_get_real_filename,
	.connectpath_fn = cephwrap_connectpath,

	/* EA operations. */
	.getxattr_fn = cephwrap_getxattr,
	.fgetxattr_fn = cephwrap_fgetxattr,
	.listxattr_fn = cephwrap_listxattr,
	.flistxattr_fn = cephwrap_flistxattr,
	.removexattr_fn = cephwrap_removexattr,
	.fremovexattr_fn = cephwrap_fremovexattr,
	.setxattr_fn = cephwrap_setxattr,
	.fsetxattr_fn = cephwrap_fsetxattr,

	/* aio operations */
	.aio_force_fn = cephwrap_aio_force,

	/* offline operations */
	.is_offline_fn = cephwrap_is_offline,
	.set_offline_fn = cephwrap_set_offline
};

NTSTATUS vfs_ceph_init(void);
NTSTATUS vfs_ceph_init(void)
{
	return smb_register_vfs(SMB_VFS_INTERFACE_VERSION,
				"ceph", &ceph_fns);
}