1===== 2Usage 3===== 4 5This module supports the SMB3 family of advanced network protocols (as well 6as older dialects, originally called "CIFS" or SMB1). 7 8The CIFS VFS module for Linux supports many advanced network filesystem 9features such as hierarchical DFS like namespace, hardlinks, locking and more. 10It was designed to comply with the SNIA CIFS Technical Reference (which 11supersedes the 1992 X/Open SMB Standard) as well as to perform best practice 12practical interoperability with Windows 2000, Windows XP, Samba and equivalent 13servers. This code was developed in participation with the Protocol Freedom 14Information Foundation. CIFS and now SMB3 has now become a defacto 15standard for interoperating between Macs and Windows and major NAS appliances. 16 17Please see 18MS-SMB2 (for detailed SMB2/SMB3/SMB3.1.1 protocol specification) 19http://protocolfreedom.org/ and 20http://samba.org/samba/PFIF/ 21for more details. 22 23 24For questions or bug reports please contact: 25 26 smfrench@gmail.com 27 28See the project page at: https://wiki.samba.org/index.php/LinuxCIFS_utils 29 30Build instructions 31================== 32 33For Linux: 34 351) Download the kernel (e.g. from http://www.kernel.org) 36 and change directory into the top of the kernel directory tree 37 (e.g. /usr/src/linux-2.5.73) 382) make menuconfig (or make xconfig) 393) select cifs from within the network filesystem choices 404) save and exit 415) make 42 43 44Installation instructions 45========================= 46 47If you have built the CIFS vfs as module (successfully) simply 48type ``make modules_install`` (or if you prefer, manually copy the file to 49the modules directory e.g. /lib/modules/2.4.10-4GB/kernel/fs/cifs/cifs.ko). 50 51If you have built the CIFS vfs into the kernel itself, follow the instructions 52for your distribution on how to install a new kernel (usually you 53would simply type ``make install``). 54 55If you do not have the utility mount.cifs (in the Samba 4.x source tree and on 56the CIFS VFS web site) copy it to the same directory in which mount helpers 57reside (usually /sbin). Although the helper software is not 58required, mount.cifs is recommended. Most distros include a ``cifs-utils`` 59package that includes this utility so it is recommended to install this. 60 61Note that running the Winbind pam/nss module (logon service) on all of your 62Linux clients is useful in mapping Uids and Gids consistently across the 63domain to the proper network user. The mount.cifs mount helper can be 64found at cifs-utils.git on git.samba.org 65 66If cifs is built as a module, then the size and number of network buffers 67and maximum number of simultaneous requests to one server can be configured. 68Changing these from their defaults is not recommended. By executing modinfo:: 69 70 modinfo kernel/fs/cifs/cifs.ko 71 72on kernel/fs/cifs/cifs.ko the list of configuration changes that can be made 73at module initialization time (by running insmod cifs.ko) can be seen. 74 75Recommendations 76=============== 77 78To improve security the SMB2.1 dialect or later (usually will get SMB3) is now 79the new default. To use old dialects (e.g. to mount Windows XP) use "vers=1.0" 80on mount (or vers=2.0 for Windows Vista). Note that the CIFS (vers=1.0) is 81much older and less secure than the default dialect SMB3 which includes 82many advanced security features such as downgrade attack detection 83and encrypted shares and stronger signing and authentication algorithms. 84There are additional mount options that may be helpful for SMB3 to get 85improved POSIX behavior (NB: can use vers=3.0 to force only SMB3, never 2.1): 86 87 ``mfsymlinks`` and ``cifsacl`` and ``idsfromsid`` 88 89Allowing User Mounts 90==================== 91 92To permit users to mount and unmount over directories they own is possible 93with the cifs vfs. A way to enable such mounting is to mark the mount.cifs 94utility as suid (e.g. ``chmod +s /sbin/mount.cifs``). To enable users to 95umount shares they mount requires 96 971) mount.cifs version 1.4 or later 982) an entry for the share in /etc/fstab indicating that a user may 99 unmount it e.g.:: 100 101 //server/usersharename /mnt/username cifs user 0 0 102 103Note that when the mount.cifs utility is run suid (allowing user mounts), 104in order to reduce risks, the ``nosuid`` mount flag is passed in on mount to 105disallow execution of an suid program mounted on the remote target. 106When mount is executed as root, nosuid is not passed in by default, 107and execution of suid programs on the remote target would be enabled 108by default. This can be changed, as with nfs and other filesystems, 109by simply specifying ``nosuid`` among the mount options. For user mounts 110though to be able to pass the suid flag to mount requires rebuilding 111mount.cifs with the following flag: CIFS_ALLOW_USR_SUID 112 113There is a corresponding manual page for cifs mounting in the Samba 3.0 and 114later source tree in docs/manpages/mount.cifs.8 115 116Allowing User Unmounts 117====================== 118 119To permit users to ummount directories that they have user mounted (see above), 120the utility umount.cifs may be used. It may be invoked directly, or if 121umount.cifs is placed in /sbin, umount can invoke the cifs umount helper 122(at least for most versions of the umount utility) for umount of cifs 123mounts, unless umount is invoked with -i (which will avoid invoking a umount 124helper). As with mount.cifs, to enable user unmounts umount.cifs must be marked 125as suid (e.g. ``chmod +s /sbin/umount.cifs``) or equivalent (some distributions 126allow adding entries to a file to the /etc/permissions file to achieve the 127equivalent suid effect). For this utility to succeed the target path 128must be a cifs mount, and the uid of the current user must match the uid 129of the user who mounted the resource. 130 131Also note that the customary way of allowing user mounts and unmounts is 132(instead of using mount.cifs and unmount.cifs as suid) to add a line 133to the file /etc/fstab for each //server/share you wish to mount, but 134this can become unwieldy when potential mount targets include many 135or unpredictable UNC names. 136 137Samba Considerations 138==================== 139 140Most current servers support SMB2.1 and SMB3 which are more secure, 141but there are useful protocol extensions for the older less secure CIFS 142dialect, so to get the maximum benefit if mounting using the older dialect 143(CIFS/SMB1), we recommend using a server that supports the SNIA CIFS 144Unix Extensions standard (e.g. almost any version of Samba ie version 1452.2.5 or later) but the CIFS vfs works fine with a wide variety of CIFS servers. 146Note that uid, gid and file permissions will display default values if you do 147not have a server that supports the Unix extensions for CIFS (such as Samba 1482.2.5 or later). To enable the Unix CIFS Extensions in the Samba server, add 149the line:: 150 151 unix extensions = yes 152 153to your smb.conf file on the server. Note that the following smb.conf settings 154are also useful (on the Samba server) when the majority of clients are Unix or 155Linux:: 156 157 case sensitive = yes 158 delete readonly = yes 159 ea support = yes 160 161Note that server ea support is required for supporting xattrs from the Linux 162cifs client, and that EA support is present in later versions of Samba (e.g. 1633.0.6 and later (also EA support works in all versions of Windows, at least to 164shares on NTFS filesystems). Extended Attribute (xattr) support is an optional 165feature of most Linux filesystems which may require enabling via 166make menuconfig. Client support for extended attributes (user xattr) can be 167disabled on a per-mount basis by specifying ``nouser_xattr`` on mount. 168 169The CIFS client can get and set POSIX ACLs (getfacl, setfacl) to Samba servers 170version 3.10 and later. Setting POSIX ACLs requires enabling both XATTR and 171then POSIX support in the CIFS configuration options when building the cifs 172module. POSIX ACL support can be disabled on a per mount basic by specifying 173``noacl`` on mount. 174 175Some administrators may want to change Samba's smb.conf ``map archive`` and 176``create mask`` parameters from the default. Unless the create mask is changed 177newly created files can end up with an unnecessarily restrictive default mode, 178which may not be what you want, although if the CIFS Unix extensions are 179enabled on the server and client, subsequent setattr calls (e.g. chmod) can 180fix the mode. Note that creating special devices (mknod) remotely 181may require specifying a mkdev function to Samba if you are not using 182Samba 3.0.6 or later. For more information on these see the manual pages 183(``man smb.conf``) on the Samba server system. Note that the cifs vfs, 184unlike the smbfs vfs, does not read the smb.conf on the client system 185(the few optional settings are passed in on mount via -o parameters instead). 186Note that Samba 2.2.7 or later includes a fix that allows the CIFS VFS to delete 187open files (required for strict POSIX compliance). Windows Servers already 188supported this feature. Samba server does not allow symlinks that refer to files 189outside of the share, so in Samba versions prior to 3.0.6, most symlinks to 190files with absolute paths (ie beginning with slash) such as:: 191 192 ln -s /mnt/foo bar 193 194would be forbidden. Samba 3.0.6 server or later includes the ability to create 195such symlinks safely by converting unsafe symlinks (ie symlinks to server 196files that are outside of the share) to a samba specific format on the server 197that is ignored by local server applications and non-cifs clients and that will 198not be traversed by the Samba server). This is opaque to the Linux client 199application using the cifs vfs. Absolute symlinks will work to Samba 3.0.5 or 200later, but only for remote clients using the CIFS Unix extensions, and will 201be invisbile to Windows clients and typically will not affect local 202applications running on the same server as Samba. 203 204Use instructions 205================ 206 207Once the CIFS VFS support is built into the kernel or installed as a module 208(cifs.ko), you can use mount syntax like the following to access Samba or 209Mac or Windows servers:: 210 211 mount -t cifs //9.53.216.11/e$ /mnt -o username=myname,password=mypassword 212 213Before -o the option -v may be specified to make the mount.cifs 214mount helper display the mount steps more verbosely. 215After -o the following commonly used cifs vfs specific options 216are supported:: 217 218 username=<username> 219 password=<password> 220 domain=<domain name> 221 222Other cifs mount options are described below. Use of TCP names (in addition to 223ip addresses) is available if the mount helper (mount.cifs) is installed. If 224you do not trust the server to which are mounted, or if you do not have 225cifs signing enabled (and the physical network is insecure), consider use 226of the standard mount options ``noexec`` and ``nosuid`` to reduce the risk of 227running an altered binary on your local system (downloaded from a hostile server 228or altered by a hostile router). 229 230Although mounting using format corresponding to the CIFS URL specification is 231not possible in mount.cifs yet, it is possible to use an alternate format 232for the server and sharename (which is somewhat similar to NFS style mount 233syntax) instead of the more widely used UNC format (i.e. \\server\share):: 234 235 mount -t cifs tcp_name_of_server:share_name /mnt -o user=myname,pass=mypasswd 236 237When using the mount helper mount.cifs, passwords may be specified via alternate 238mechanisms, instead of specifying it after -o using the normal ``pass=`` syntax 239on the command line: 2401) By including it in a credential file. Specify credentials=filename as one 241of the mount options. Credential files contain two lines:: 242 243 username=someuser 244 password=your_password 245 2462) By specifying the password in the PASSWD environment variable (similarly 247 the user name can be taken from the USER environment variable). 2483) By specifying the password in a file by name via PASSWD_FILE 2494) By specifying the password in a file by file descriptor via PASSWD_FD 250 251If no password is provided, mount.cifs will prompt for password entry 252 253Restrictions 254============ 255 256Servers must support either "pure-TCP" (port 445 TCP/IP CIFS connections) or RFC 2571001/1002 support for "Netbios-Over-TCP/IP." This is not likely to be a 258problem as most servers support this. 259 260Valid filenames differ between Windows and Linux. Windows typically restricts 261filenames which contain certain reserved characters (e.g.the character : 262which is used to delimit the beginning of a stream name by Windows), while 263Linux allows a slightly wider set of valid characters in filenames. Windows 264servers can remap such characters when an explicit mapping is specified in 265the Server's registry. Samba starting with version 3.10 will allow such 266filenames (ie those which contain valid Linux characters, which normally 267would be forbidden for Windows/CIFS semantics) as long as the server is 268configured for Unix Extensions (and the client has not disabled 269/proc/fs/cifs/LinuxExtensionsEnabled). In addition the mount option 270``mapposix`` can be used on CIFS (vers=1.0) to force the mapping of 271illegal Windows/NTFS/SMB characters to a remap range (this mount parm 272is the default for SMB3). This remap (``mapposix``) range is also 273compatible with Mac (and "Services for Mac" on some older Windows). 274 275CIFS VFS Mount Options 276====================== 277A partial list of the supported mount options follows: 278 279 username 280 The user name to use when trying to establish 281 the CIFS session. 282 password 283 The user password. If the mount helper is 284 installed, the user will be prompted for password 285 if not supplied. 286 ip 287 The ip address of the target server 288 unc 289 The target server Universal Network Name (export) to 290 mount. 291 domain 292 Set the SMB/CIFS workgroup name prepended to the 293 username during CIFS session establishment 294 forceuid 295 Set the default uid for inodes to the uid 296 passed in on mount. For mounts to servers 297 which do support the CIFS Unix extensions, such as a 298 properly configured Samba server, the server provides 299 the uid, gid and mode so this parameter should not be 300 specified unless the server and clients uid and gid 301 numbering differ. If the server and client are in the 302 same domain (e.g. running winbind or nss_ldap) and 303 the server supports the Unix Extensions then the uid 304 and gid can be retrieved from the server (and uid 305 and gid would not have to be specified on the mount. 306 For servers which do not support the CIFS Unix 307 extensions, the default uid (and gid) returned on lookup 308 of existing files will be the uid (gid) of the person 309 who executed the mount (root, except when mount.cifs 310 is configured setuid for user mounts) unless the ``uid=`` 311 (gid) mount option is specified. Also note that permission 312 checks (authorization checks) on accesses to a file occur 313 at the server, but there are cases in which an administrator 314 may want to restrict at the client as well. For those 315 servers which do not report a uid/gid owner 316 (such as Windows), permissions can also be checked at the 317 client, and a crude form of client side permission checking 318 can be enabled by specifying file_mode and dir_mode on 319 the client. (default) 320 forcegid 321 (similar to above but for the groupid instead of uid) (default) 322 noforceuid 323 Fill in file owner information (uid) by requesting it from 324 the server if possible. With this option, the value given in 325 the uid= option (on mount) will only be used if the server 326 can not support returning uids on inodes. 327 noforcegid 328 (similar to above but for the group owner, gid, instead of uid) 329 uid 330 Set the default uid for inodes, and indicate to the 331 cifs kernel driver which local user mounted. If the server 332 supports the unix extensions the default uid is 333 not used to fill in the owner fields of inodes (files) 334 unless the ``forceuid`` parameter is specified. 335 gid 336 Set the default gid for inodes (similar to above). 337 file_mode 338 If CIFS Unix extensions are not supported by the server 339 this overrides the default mode for file inodes. 340 fsc 341 Enable local disk caching using FS-Cache (off by default). This 342 option could be useful to improve performance on a slow link, 343 heavily loaded server and/or network where reading from the 344 disk is faster than reading from the server (over the network). 345 This could also impact scalability positively as the 346 number of calls to the server are reduced. However, local 347 caching is not suitable for all workloads for e.g. read-once 348 type workloads. So, you need to consider carefully your 349 workload/scenario before using this option. Currently, local 350 disk caching is functional for CIFS files opened as read-only. 351 dir_mode 352 If CIFS Unix extensions are not supported by the server 353 this overrides the default mode for directory inodes. 354 port 355 attempt to contact the server on this tcp port, before 356 trying the usual ports (port 445, then 139). 357 iocharset 358 Codepage used to convert local path names to and from 359 Unicode. Unicode is used by default for network path 360 names if the server supports it. If iocharset is 361 not specified then the nls_default specified 362 during the local client kernel build will be used. 363 If server does not support Unicode, this parameter is 364 unused. 365 rsize 366 default read size (usually 16K). The client currently 367 can not use rsize larger than CIFSMaxBufSize. CIFSMaxBufSize 368 defaults to 16K and may be changed (from 8K to the maximum 369 kmalloc size allowed by your kernel) at module install time 370 for cifs.ko. Setting CIFSMaxBufSize to a very large value 371 will cause cifs to use more memory and may reduce performance 372 in some cases. To use rsize greater than 127K (the original 373 cifs protocol maximum) also requires that the server support 374 a new Unix Capability flag (for very large read) which some 375 newer servers (e.g. Samba 3.0.26 or later) do. rsize can be 376 set from a minimum of 2048 to a maximum of 130048 (127K or 377 CIFSMaxBufSize, whichever is smaller) 378 wsize 379 default write size (default 57344) 380 maximum wsize currently allowed by CIFS is 57344 (fourteen 381 4096 byte pages) 382 actimeo=n 383 attribute cache timeout in seconds (default 1 second). 384 After this timeout, the cifs client requests fresh attribute 385 information from the server. This option allows to tune the 386 attribute cache timeout to suit the workload needs. Shorter 387 timeouts mean better the cache coherency, but increased number 388 of calls to the server. Longer timeouts mean reduced number 389 of calls to the server at the expense of less stricter cache 390 coherency checks (i.e. incorrect attribute cache for a short 391 period of time). 392 rw 393 mount the network share read-write (note that the 394 server may still consider the share read-only) 395 ro 396 mount network share read-only 397 version 398 used to distinguish different versions of the 399 mount helper utility (not typically needed) 400 sep 401 if first mount option (after the -o), overrides 402 the comma as the separator between the mount 403 parms. e.g.:: 404 405 -o user=myname,password=mypassword,domain=mydom 406 407 could be passed instead with period as the separator by:: 408 409 -o sep=.user=myname.password=mypassword.domain=mydom 410 411 this might be useful when comma is contained within username 412 or password or domain. This option is less important 413 when the cifs mount helper cifs.mount (version 1.1 or later) 414 is used. 415 nosuid 416 Do not allow remote executables with the suid bit 417 program to be executed. This is only meaningful for mounts 418 to servers such as Samba which support the CIFS Unix Extensions. 419 If you do not trust the servers in your network (your mount 420 targets) it is recommended that you specify this option for 421 greater security. 422 exec 423 Permit execution of binaries on the mount. 424 noexec 425 Do not permit execution of binaries on the mount. 426 dev 427 Recognize block devices on the remote mount. 428 nodev 429 Do not recognize devices on the remote mount. 430 suid 431 Allow remote files on this mountpoint with suid enabled to 432 be executed (default for mounts when executed as root, 433 nosuid is default for user mounts). 434 credentials 435 Although ignored by the cifs kernel component, it is used by 436 the mount helper, mount.cifs. When mount.cifs is installed it 437 opens and reads the credential file specified in order 438 to obtain the userid and password arguments which are passed to 439 the cifs vfs. 440 guest 441 Although ignored by the kernel component, the mount.cifs 442 mount helper will not prompt the user for a password 443 if guest is specified on the mount options. If no 444 password is specified a null password will be used. 445 perm 446 Client does permission checks (vfs_permission check of uid 447 and gid of the file against the mode and desired operation), 448 Note that this is in addition to the normal ACL check on the 449 target machine done by the server software. 450 Client permission checking is enabled by default. 451 noperm 452 Client does not do permission checks. This can expose 453 files on this mount to access by other users on the local 454 client system. It is typically only needed when the server 455 supports the CIFS Unix Extensions but the UIDs/GIDs on the 456 client and server system do not match closely enough to allow 457 access by the user doing the mount, but it may be useful with 458 non CIFS Unix Extension mounts for cases in which the default 459 mode is specified on the mount but is not to be enforced on the 460 client (e.g. perhaps when MultiUserMount is enabled) 461 Note that this does not affect the normal ACL check on the 462 target machine done by the server software (of the server 463 ACL against the user name provided at mount time). 464 serverino 465 Use server's inode numbers instead of generating automatically 466 incrementing inode numbers on the client. Although this will 467 make it easier to spot hardlinked files (as they will have 468 the same inode numbers) and inode numbers may be persistent, 469 note that the server does not guarantee that the inode numbers 470 are unique if multiple server side mounts are exported under a 471 single share (since inode numbers on the servers might not 472 be unique if multiple filesystems are mounted under the same 473 shared higher level directory). Note that some older 474 (e.g. pre-Windows 2000) do not support returning UniqueIDs 475 or the CIFS Unix Extensions equivalent and for those 476 this mount option will have no effect. Exporting cifs mounts 477 under nfsd requires this mount option on the cifs mount. 478 This is now the default if server supports the 479 required network operation. 480 noserverino 481 Client generates inode numbers (rather than using the actual one 482 from the server). These inode numbers will vary after 483 unmount or reboot which can confuse some applications, 484 but not all server filesystems support unique inode 485 numbers. 486 setuids 487 If the CIFS Unix extensions are negotiated with the server 488 the client will attempt to set the effective uid and gid of 489 the local process on newly created files, directories, and 490 devices (create, mkdir, mknod). If the CIFS Unix Extensions 491 are not negotiated, for newly created files and directories 492 instead of using the default uid and gid specified on 493 the mount, cache the new file's uid and gid locally which means 494 that the uid for the file can change when the inode is 495 reloaded (or the user remounts the share). 496 nosetuids 497 The client will not attempt to set the uid and gid on 498 on newly created files, directories, and devices (create, 499 mkdir, mknod) which will result in the server setting the 500 uid and gid to the default (usually the server uid of the 501 user who mounted the share). Letting the server (rather than 502 the client) set the uid and gid is the default. If the CIFS 503 Unix Extensions are not negotiated then the uid and gid for 504 new files will appear to be the uid (gid) of the mounter or the 505 uid (gid) parameter specified on the mount. 506 netbiosname 507 When mounting to servers via port 139, specifies the RFC1001 508 source name to use to represent the client netbios machine 509 name when doing the RFC1001 netbios session initialize. 510 direct 511 Do not do inode data caching on files opened on this mount. 512 This precludes mmapping files on this mount. In some cases 513 with fast networks and little or no caching benefits on the 514 client (e.g. when the application is doing large sequential 515 reads bigger than page size without rereading the same data) 516 this can provide better performance than the default 517 behavior which caches reads (readahead) and writes 518 (writebehind) through the local Linux client pagecache 519 if oplock (caching token) is granted and held. Note that 520 direct allows write operations larger than page size 521 to be sent to the server. 522 strictcache 523 Use for switching on strict cache mode. In this mode the 524 client read from the cache all the time it has Oplock Level II, 525 otherwise - read from the server. All written data are stored 526 in the cache, but if the client doesn't have Exclusive Oplock, 527 it writes the data to the server. 528 rwpidforward 529 Forward pid of a process who opened a file to any read or write 530 operation on that file. This prevent applications like WINE 531 from failing on read and write if we use mandatory brlock style. 532 acl 533 Allow setfacl and getfacl to manage posix ACLs if server 534 supports them. (default) 535 noacl 536 Do not allow setfacl and getfacl calls on this mount 537 user_xattr 538 Allow getting and setting user xattrs (those attributes whose 539 name begins with ``user.`` or ``os2.``) as OS/2 EAs (extended 540 attributes) to the server. This allows support of the 541 setfattr and getfattr utilities. (default) 542 nouser_xattr 543 Do not allow getfattr/setfattr to get/set/list xattrs 544 mapchars 545 Translate six of the seven reserved characters (not backslash):: 546 547 *?<>|: 548 549 to the remap range (above 0xF000), which also 550 allows the CIFS client to recognize files created with 551 such characters by Windows's POSIX emulation. This can 552 also be useful when mounting to most versions of Samba 553 (which also forbids creating and opening files 554 whose names contain any of these seven characters). 555 This has no effect if the server does not support 556 Unicode on the wire. 557 nomapchars 558 Do not translate any of these seven characters (default). 559 nocase 560 Request case insensitive path name matching (case 561 sensitive is the default if the server supports it). 562 (mount option ``ignorecase`` is identical to ``nocase``) 563 posixpaths 564 If CIFS Unix extensions are supported, attempt to 565 negotiate posix path name support which allows certain 566 characters forbidden in typical CIFS filenames, without 567 requiring remapping. (default) 568 noposixpaths 569 If CIFS Unix extensions are supported, do not request 570 posix path name support (this may cause servers to 571 reject creatingfile with certain reserved characters). 572 nounix 573 Disable the CIFS Unix Extensions for this mount (tree 574 connection). This is rarely needed, but it may be useful 575 in order to turn off multiple settings all at once (ie 576 posix acls, posix locks, posix paths, symlink support 577 and retrieving uids/gids/mode from the server) or to 578 work around a bug in server which implement the Unix 579 Extensions. 580 nobrl 581 Do not send byte range lock requests to the server. 582 This is necessary for certain applications that break 583 with cifs style mandatory byte range locks (and most 584 cifs servers do not yet support requesting advisory 585 byte range locks). 586 forcemandatorylock 587 Even if the server supports posix (advisory) byte range 588 locking, send only mandatory lock requests. For some 589 (presumably rare) applications, originally coded for 590 DOS/Windows, which require Windows style mandatory byte range 591 locking, they may be able to take advantage of this option, 592 forcing the cifs client to only send mandatory locks 593 even if the cifs server would support posix advisory locks. 594 ``forcemand`` is accepted as a shorter form of this mount 595 option. 596 nostrictsync 597 If this mount option is set, when an application does an 598 fsync call then the cifs client does not send an SMB Flush 599 to the server (to force the server to write all dirty data 600 for this file immediately to disk), although cifs still sends 601 all dirty (cached) file data to the server and waits for the 602 server to respond to the write. Since SMB Flush can be 603 very slow, and some servers may be reliable enough (to risk 604 delaying slightly flushing the data to disk on the server), 605 turning on this option may be useful to improve performance for 606 applications that fsync too much, at a small risk of server 607 crash. If this mount option is not set, by default cifs will 608 send an SMB flush request (and wait for a response) on every 609 fsync call. 610 nodfs 611 Disable DFS (global name space support) even if the 612 server claims to support it. This can help work around 613 a problem with parsing of DFS paths with Samba server 614 versions 3.0.24 and 3.0.25. 615 remount 616 remount the share (often used to change from ro to rw mounts 617 or vice versa) 618 cifsacl 619 Report mode bits (e.g. on stat) based on the Windows ACL for 620 the file. (EXPERIMENTAL) 621 servern 622 Specify the server 's netbios name (RFC1001 name) to use 623 when attempting to setup a session to the server. 624 This is needed for mounting to some older servers (such 625 as OS/2 or Windows 98 and Windows ME) since they do not 626 support a default server name. A server name can be up 627 to 15 characters long and is usually uppercased. 628 sfu 629 When the CIFS Unix Extensions are not negotiated, attempt to 630 create device files and fifos in a format compatible with 631 Services for Unix (SFU). In addition retrieve bits 10-12 632 of the mode via the SETFILEBITS extended attribute (as 633 SFU does). In the future the bottom 9 bits of the 634 mode also will be emulated using queries of the security 635 descriptor (ACL). 636 mfsymlinks 637 Enable support for Minshall+French symlinks 638 (see http://wiki.samba.org/index.php/UNIX_Extensions#Minshall.2BFrench_symlinks) 639 This option is ignored when specified together with the 640 'sfu' option. Minshall+French symlinks are used even if 641 the server supports the CIFS Unix Extensions. 642 sign 643 Must use packet signing (helps avoid unwanted data modification 644 by intermediate systems in the route). Note that signing 645 does not work with lanman or plaintext authentication. 646 seal 647 Must seal (encrypt) all data on this mounted share before 648 sending on the network. Requires support for Unix Extensions. 649 Note that this differs from the sign mount option in that it 650 causes encryption of data sent over this mounted share but other 651 shares mounted to the same server are unaffected. 652 locallease 653 This option is rarely needed. Fcntl F_SETLEASE is 654 used by some applications such as Samba and NFSv4 server to 655 check to see whether a file is cacheable. CIFS has no way 656 to explicitly request a lease, but can check whether a file 657 is cacheable (oplocked). Unfortunately, even if a file 658 is not oplocked, it could still be cacheable (ie cifs client 659 could grant fcntl leases if no other local processes are using 660 the file) for cases for example such as when the server does not 661 support oplocks and the user is sure that the only updates to 662 the file will be from this client. Specifying this mount option 663 will allow the cifs client to check for leases (only) locally 664 for files which are not oplocked instead of denying leases 665 in that case. (EXPERIMENTAL) 666 sec 667 Security mode. Allowed values are: 668 669 none 670 attempt to connection as a null user (no name) 671 krb5 672 Use Kerberos version 5 authentication 673 krb5i 674 Use Kerberos authentication and packet signing 675 ntlm 676 Use NTLM password hashing (default) 677 ntlmi 678 Use NTLM password hashing with signing (if 679 /proc/fs/cifs/PacketSigningEnabled on or if 680 server requires signing also can be the default) 681 ntlmv2 682 Use NTLMv2 password hashing 683 ntlmv2i 684 Use NTLMv2 password hashing with packet signing 685 lanman 686 (if configured in kernel config) use older 687 lanman hash 688 hard 689 Retry file operations if server is not responding 690 soft 691 Limit retries to unresponsive servers (usually only 692 one retry) before returning an error. (default) 693 694The mount.cifs mount helper also accepts a few mount options before -o 695including: 696 697=============== =============================================================== 698 -S take password from stdin (equivalent to setting the environment 699 variable ``PASSWD_FD=0`` 700 -V print mount.cifs version 701 -? display simple usage information 702=============== =============================================================== 703 704With most 2.6 kernel versions of modutils, the version of the cifs kernel 705module can be displayed via modinfo. 706 707Misc /proc/fs/cifs Flags and Debug Info 708======================================= 709 710Informational pseudo-files: 711 712======================= ======================================================= 713DebugData Displays information about active CIFS sessions and 714 shares, features enabled as well as the cifs.ko 715 version. 716Stats Lists summary resource usage information as well as per 717 share statistics. 718======================= ======================================================= 719 720Configuration pseudo-files: 721 722======================= ======================================================= 723SecurityFlags Flags which control security negotiation and 724 also packet signing. Authentication (may/must) 725 flags (e.g. for NTLM and/or NTLMv2) may be combined with 726 the signing flags. Specifying two different password 727 hashing mechanisms (as "must use") on the other hand 728 does not make much sense. Default flags are:: 729 730 0x07007 731 732 (NTLM, NTLMv2 and packet signing allowed). The maximum 733 allowable flags if you want to allow mounts to servers 734 using weaker password hashes is 0x37037 (lanman, 735 plaintext, ntlm, ntlmv2, signing allowed). Some 736 SecurityFlags require the corresponding menuconfig 737 options to be enabled (lanman and plaintext require 738 CONFIG_CIFS_WEAK_PW_HASH for example). Enabling 739 plaintext authentication currently requires also 740 enabling lanman authentication in the security flags 741 because the cifs module only supports sending 742 laintext passwords using the older lanman dialect 743 form of the session setup SMB. (e.g. for authentication 744 using plain text passwords, set the SecurityFlags 745 to 0x30030):: 746 747 may use packet signing 0x00001 748 must use packet signing 0x01001 749 may use NTLM (most common password hash) 0x00002 750 must use NTLM 0x02002 751 may use NTLMv2 0x00004 752 must use NTLMv2 0x04004 753 may use Kerberos security 0x00008 754 must use Kerberos 0x08008 755 may use lanman (weak) password hash 0x00010 756 must use lanman password hash 0x10010 757 may use plaintext passwords 0x00020 758 must use plaintext passwords 0x20020 759 (reserved for future packet encryption) 0x00040 760 761cifsFYI If set to non-zero value, additional debug information 762 will be logged to the system error log. This field 763 contains three flags controlling different classes of 764 debugging entries. The maximum value it can be set 765 to is 7 which enables all debugging points (default 0). 766 Some debugging statements are not compiled into the 767 cifs kernel unless CONFIG_CIFS_DEBUG2 is enabled in the 768 kernel configuration. cifsFYI may be set to one or 769 nore of the following flags (7 sets them all):: 770 771 +-----------------------------------------------+------+ 772 | log cifs informational messages | 0x01 | 773 +-----------------------------------------------+------+ 774 | log return codes from cifs entry points | 0x02 | 775 +-----------------------------------------------+------+ 776 | log slow responses | 0x04 | 777 | (ie which take longer than 1 second) | | 778 | | | 779 | CONFIG_CIFS_STATS2 must be enabled in .config | | 780 +-----------------------------------------------+------+ 781 782traceSMB If set to one, debug information is logged to the 783 system error log with the start of smb requests 784 and responses (default 0) 785LookupCacheEnable If set to one, inode information is kept cached 786 for one second improving performance of lookups 787 (default 1) 788LinuxExtensionsEnabled If set to one then the client will attempt to 789 use the CIFS "UNIX" extensions which are optional 790 protocol enhancements that allow CIFS servers 791 to return accurate UID/GID information as well 792 as support symbolic links. If you use servers 793 such as Samba that support the CIFS Unix 794 extensions but do not want to use symbolic link 795 support and want to map the uid and gid fields 796 to values supplied at mount (rather than the 797 actual values, then set this to zero. (default 1) 798======================= ======================================================= 799 800These experimental features and tracing can be enabled by changing flags in 801/proc/fs/cifs (after the cifs module has been installed or built into the 802kernel, e.g. insmod cifs). To enable a feature set it to 1 e.g. to enable 803tracing to the kernel message log type:: 804 805 echo 7 > /proc/fs/cifs/cifsFYI 806 807cifsFYI functions as a bit mask. Setting it to 1 enables additional kernel 808logging of various informational messages. 2 enables logging of non-zero 809SMB return codes while 4 enables logging of requests that take longer 810than one second to complete (except for byte range lock requests). 811Setting it to 4 requires CONFIG_CIFS_STATS2 to be set in kernel configuration 812(.config). Setting it to seven enables all three. Finally, tracing 813the start of smb requests and responses can be enabled via:: 814 815 echo 1 > /proc/fs/cifs/traceSMB 816 817Per share (per client mount) statistics are available in /proc/fs/cifs/Stats. 818Additional information is available if CONFIG_CIFS_STATS2 is enabled in the 819kernel configuration (.config). The statistics returned include counters which 820represent the number of attempted and failed (ie non-zero return code from the 821server) SMB3 (or cifs) requests grouped by request type (read, write, close etc.). 822Also recorded is the total bytes read and bytes written to the server for 823that share. Note that due to client caching effects this can be less than the 824number of bytes read and written by the application running on the client. 825Statistics can be reset to zero by ``echo 0 > /proc/fs/cifs/Stats`` which may be 826useful if comparing performance of two different scenarios. 827 828Also note that ``cat /proc/fs/cifs/DebugData`` will display information about 829the active sessions and the shares that are mounted. 830 831Enabling Kerberos (extended security) works but requires version 1.2 or later 832of the helper program cifs.upcall to be present and to be configured in the 833/etc/request-key.conf file. The cifs.upcall helper program is from the Samba 834project(http://www.samba.org). NTLM and NTLMv2 and LANMAN support do not 835require this helper. Note that NTLMv2 security (which does not require the 836cifs.upcall helper program), instead of using Kerberos, is sufficient for 837some use cases. 838 839DFS support allows transparent redirection to shares in an MS-DFS name space. 840In addition, DFS support for target shares which are specified as UNC 841names which begin with host names (rather than IP addresses) requires 842a user space helper (such as cifs.upcall) to be present in order to 843translate host names to ip address, and the user space helper must also 844be configured in the file /etc/request-key.conf. Samba, Windows servers and 845many NAS appliances support DFS as a way of constructing a global name 846space to ease network configuration and improve reliability. 847 848To use cifs Kerberos and DFS support, the Linux keyutils package should be 849installed and something like the following lines should be added to the 850/etc/request-key.conf file:: 851 852 create cifs.spnego * * /usr/local/sbin/cifs.upcall %k 853 create dns_resolver * * /usr/local/sbin/cifs.upcall %k 854 855CIFS kernel module parameters 856============================= 857These module parameters can be specified or modified either during the time of 858module loading or during the runtime by using the interface:: 859 860 /proc/module/cifs/parameters/<param> 861 862i.e.:: 863 864 echo "value" > /sys/module/cifs/parameters/<param> 865 866================= ========================================================== 8671. enable_oplocks Enable or disable oplocks. Oplocks are enabled by default. 868 [Y/y/1]. To disable use any of [N/n/0]. 869================= ========================================================== 870