blob: e1a0056a365ffd42c0e93aa1233d04beb40b4e72 [file] [log] [blame]
Mike Marshall74a552a2015-07-17 10:38:16 -04001ORANGEFS
2========
3
4OrangeFS is an LGPL userspace scale-out parallel storage system. It is ideal
5for large storage problems faced by HPC, BigData, Streaming Video,
6Genomics, Bioinformatics.
7
8Orangefs, originally called PVFS, was first developed in 1993 by
9Walt Ligon and Eric Blumer as a parallel file system for Parallel
10Virtual Machine (PVM) as part of a NASA grant to study the I/O patterns
11of parallel programs.
12
13Orangefs features include:
14
15 * Distributes file data among multiple file servers
16 * Supports simultaneous access by multiple clients
17 * Stores file data and metadata on servers using local file system
18 and access methods
19 * Userspace implementation is easy to install and maintain
20 * Direct MPI support
21 * Stateless
22
23
24MAILING LIST
25============
26
27http://beowulf-underground.org/mailman/listinfo/pvfs2-users
28
29
30DOCUMENTATION
31=============
32
33http://www.orangefs.org/documentation/
34
35
36USERSPACE FILESYSTEM SOURCE
37===========================
38
39http://www.orangefs.org/download
40
41Orangefs versions prior to 2.9.3 would not be compatible with the
42upstream version of the kernel client.
43
44
45BUILDING THE USERSPACE FILESYSTEM ON A SINGLE SERVER
46====================================================
47
48When Orangefs is upstream, "--with-kernel" shouldn't be needed, but
49until then the path to where the kernel with the Orangefs kernel client
50patch was built is needed to ensure that pvfs2-client-core (the bridge
51between kernel space and user space) will build properly. You can omit
52--prefix if you don't care that things are sprinkled around in
53/usr/local.
54
55./configure --prefix=/opt/ofs --with-kernel=/path/to/orangefs/kernel
56
57make
58
59make install
60
61Create an orangefs config file:
62/opt/ofs/bin/pvfs2-genconfig /etc/pvfs2.conf
63
64 for "Enter hostnames", use the hostname, don't let it default to
65 localhost.
66
67create a pvfs2tab file in /etc:
68cat /etc/pvfs2tab
69tcp://myhostname:3334/orangefs /mymountpoint pvfs2 defaults,noauto 0 0
70
71create the mount point you specified in the tab file if needed:
72mkdir /mymountpoint
73
74bootstrap the server:
75/opt/ofs/sbin/pvfs2-server /etc/pvfs2.conf -f
76
77start the server:
78/opt/osf/sbin/pvfs2-server /etc/pvfs2.conf
79
80Now the server is running. At this point you might like to
81prove things are working with:
82
83/opt/osf/bin/pvfs2-ls /mymountpoint
84
85You might not want to enforce selinux, it doesn't seem to matter by
86linux 3.11...
87
88If stuff seems to be working, turn on the client core:
89/opt/osf/sbin/pvfs2-client -p /opt/osf/sbin/pvfs2-client-core
90
91Mount your filesystem.
92mount -t pvfs2 tcp://myhostname:3334/orangefs /mymountpoint
93
94
95OPTIONS
96=======
97
98The following mount options are accepted:
99
100 acl
101 Allow the use of Access Control Lists on files and directories.
102
103 intr
104 Some operations between the kernel client and the user space
105 filesystem can be interruptible, such as changes in debug levels
106 and the setting of tunable parameters.
107
108 local_lock
109 Enable posix locking from the perspective of "this" kernel. The
110 default file_operations lock action is to return ENOSYS. Posix
111 locking kicks in if the filesystem is mounted with -o local_lock.
112 Distributed locking is being worked on for the future.
113
114
115DEBUGGING
116=========
117
Mike Marshallfcac9d52016-01-13 14:28:13 -0500118If you want the debug (GOSSIP) statements in a particular
Mike Marshall74a552a2015-07-17 10:38:16 -0400119source file (inode.c for example) go to syslog:
120
121 echo inode > /sys/kernel/debug/orangefs/kernel-debug
122
123No debugging (the default):
124
125 echo none > /sys/kernel/debug/orangefs/kernel-debug
126
127Debugging from several source files:
128
129 echo inode,dir > /sys/kernel/debug/orangefs/kernel-debug
130
131All debugging:
132
133 echo all > /sys/kernel/debug/orangefs/kernel-debug
134
135Get a list of all debugging keywords:
136
137 cat /sys/kernel/debug/orangefs/debug-help
Mike Marshallfcac9d52016-01-13 14:28:13 -0500138
139
140PROTOCOL BETWEEN KERNEL MODULE AND USERSPACE
141============================================
142
143Orangefs is a user space filesystem and an associated kernel module.
144We'll just refer to the user space part of Orangefs as "userspace"
145from here on out. Orangefs descends from PVFS, and userspace code
146still uses PVFS for function and variable names. Userspace typedefs
147many of the important structures. Function and variable names in
148the kernel module have been transitioned to "orangefs", and The Linux
149Coding Style avoids typedefs, so kernel module structures that
150correspond to userspace structures are not typedefed.
151
152The kernel module implements a pseudo device that userspace
153can read from and write to. Userspace can also manipulate the
154kernel module through the pseudo device with ioctl.
155
156THE BUFMAP:
157
158At startup userspace allocates two page-size-aligned (posix_memalign)
159mlocked memory buffers, one is used for IO and one is used for readdir
160operations. The IO buffer is 41943040 bytes and the readdir buffer is
1614194304 bytes. Each buffer contains logical chunks, or partitions, and
162a pointer to each buffer is added to its own PVFS_dev_map_desc structure
163which also describes its total size, as well as the size and number of
164the partitions.
165
166A pointer to the IO buffer's PVFS_dev_map_desc structure is sent to a
167mapping routine in the kernel module with an ioctl. The structure is
168copied from user space to kernel space with copy_from_user and is used
169to initialize the kernel module's "bufmap" (struct orangefs_bufmap), which
170then contains:
171
172 * refcnt - a reference counter
173 * desc_size - PVFS2_BUFMAP_DEFAULT_DESC_SIZE (4194304) - the IO buffer's
174 partition size, which represents the filesystem's block size and
175 is used for s_blocksize in super blocks.
176 * desc_count - PVFS2_BUFMAP_DEFAULT_DESC_COUNT (10) - the number of
177 partitions in the IO buffer.
178 * desc_shift - log2(desc_size), used for s_blocksize_bits in super blocks.
179 * total_size - the total size of the IO buffer.
180 * page_count - the number of 4096 byte pages in the IO buffer.
181 * page_array - a pointer to page_count * (sizeof(struct page*)) bytes
182 of kcalloced memory. This memory is used as an array of pointers
183 to each of the pages in the IO buffer through a call to get_user_pages.
184 * desc_array - a pointer to desc_count * (sizeof(struct orangefs_bufmap_desc))
185 bytes of kcalloced memory. This memory is further intialized:
186
187 user_desc is the kernel's copy of the IO buffer's ORANGEFS_dev_map_desc
188 structure. user_desc->ptr points to the IO buffer.
189
190 pages_per_desc = bufmap->desc_size / PAGE_SIZE
191 offset = 0
192
193 bufmap->desc_array[0].page_array = &bufmap->page_array[offset]
194 bufmap->desc_array[0].array_count = pages_per_desc = 1024
195 bufmap->desc_array[0].uaddr = (user_desc->ptr) + (0 * 1024 * 4096)
196 offset += 1024
197 .
198 .
199 .
200 bufmap->desc_array[9].page_array = &bufmap->page_array[offset]
201 bufmap->desc_array[9].array_count = pages_per_desc = 1024
202 bufmap->desc_array[9].uaddr = (user_desc->ptr) +
203 (9 * 1024 * 4096)
204 offset += 1024
205
206 * buffer_index_array - a desc_count sized array of ints, used to
207 indicate which of the IO buffer's partitions are available to use.
208 * buffer_index_lock - a spinlock to protect buffer_index_array during update.
209 * readdir_index_array - a five (ORANGEFS_READDIR_DEFAULT_DESC_COUNT) element
210 int array used to indicate which of the readdir buffer's partitions are
211 available to use.
212 * readdir_index_lock - a spinlock to protect readdir_index_array during
213 update.
214
215OPERATIONS:
216
217The kernel module builds an "op" (struct orangefs_kernel_op_s) when it
218needs to communicate with userspace. Part of the op contains the "upcall"
219which expresses the request to userspace. Part of the op eventually
220contains the "downcall" which expresses the results of the request.
221
222The slab allocator is used to keep a cache of op structures handy.
223
Mike Marshall9f08cfe92016-02-26 14:39:08 -0500224At init time the kernel module defines and initializes a request list
225and an in_progress hash table to keep track of all the ops that are
226in flight at any given time.
Mike Marshallfcac9d52016-01-13 14:28:13 -0500227
Mike Marshall9f08cfe92016-02-26 14:39:08 -0500228Ops are stateful:
Mike Marshallfcac9d52016-01-13 14:28:13 -0500229
Mike Marshall9f08cfe92016-02-26 14:39:08 -0500230 * unknown - op was just initialized
231 * waiting - op is on request_list (upward bound)
232 * inprogr - op is in progress (waiting for downcall)
233 * serviced - op has matching downcall; ok
234 * purged - op has to start a timer since client-core
235 exited uncleanly before servicing op
236 * given up - submitter has given up waiting for it
Mike Marshallfcac9d52016-01-13 14:28:13 -0500237
Mike Marshall9f08cfe92016-02-26 14:39:08 -0500238When some arbitrary userspace program needs to perform a
239filesystem operation on Orangefs (readdir, I/O, create, whatever)
240an op structure is initialized and tagged with a distinguishing ID
241number. The upcall part of the op is filled out, and the op is
242passed to the "service_operation" function.
Mike Marshallfcac9d52016-01-13 14:28:13 -0500243
Mike Marshall9f08cfe92016-02-26 14:39:08 -0500244Service_operation changes the op's state to "waiting", puts
245it on the request list, and signals the Orangefs file_operations.poll
246function through a wait queue. Userspace is polling the pseudo-device
247and thus becomes aware of the upcall request that needs to be read.
Mike Marshallfcac9d52016-01-13 14:28:13 -0500248
Mike Marshall9f08cfe92016-02-26 14:39:08 -0500249When the Orangefs file_operations.read function is triggered, the
250request list is searched for an op that seems ready-to-process.
251The op is removed from the request list. The tag from the op and
252the filled-out upcall struct are copy_to_user'ed back to userspace.
253
254If any of these (and some additional protocol) copy_to_users fail,
255the op's state is set to "waiting" and the op is added back to
256the request list. Otherwise, the op's state is changed to "in progress",
257and the op is hashed on its tag and put onto the end of a list in the
258in_progress hash table at the index the tag hashed to.
259
260When userspace has assembled the response to the upcall, it
261writes the response, which includes the distinguishing tag, back to
262the pseudo device in a series of io_vecs. This triggers the Orangefs
263file_operations.write_iter function to find the op with the associated
264tag and remove it from the in_progress hash table. As long as the op's
265state is not "canceled" or "given up", its state is set to "serviced".
266The file_operations.write_iter function returns to the waiting vfs,
267and back to service_operation through wait_for_matching_downcall.
268
269Service operation returns to its caller with the op's downcall
270part (the response to the upcall) filled out.
271
272The "client-core" is the bridge between the kernel module and
273userspace. The client-core is a daemon. The client-core has an
274associated watchdog daemon. If the client-core is ever signaled
275to die, the watchdog daemon restarts the client-core. Even though
276the client-core is restarted "right away", there is a period of
277time during such an event that the client-core is dead. A dead client-core
278can't be triggered by the Orangefs file_operations.poll function.
279Ops that pass through service_operation during a "dead spell" can timeout
280on the wait queue and one attempt is made to recycle them. Obviously,
281if the client-core stays dead too long, the arbitrary userspace processes
282trying to use Orangefs will be negatively affected. Waiting ops
283that can't be serviced will be removed from the request list and
284have their states set to "given up". In-progress ops that can't
285be serviced will be removed from the in_progress hash table and
286have their states set to "given up".
287
288Readdir and I/O ops are atypical with respect to their payloads.
Mike Marshallfcac9d52016-01-13 14:28:13 -0500289
290 - readdir ops use the smaller of the two pre-allocated pre-partitioned
291 memory buffers. The readdir buffer is only available to userspace.
292 The kernel module obtains an index to a free partition before launching
293 a readdir op. Userspace deposits the results into the indexed partition
294 and then writes them to back to the pvfs device.
295
296 - io (read and write) ops use the larger of the two pre-allocated
297 pre-partitioned memory buffers. The IO buffer is accessible from
298 both userspace and the kernel module. The kernel module obtains an
299 index to a free partition before launching an io op. The kernel module
300 deposits write data into the indexed partition, to be consumed
301 directly by userspace. Userspace deposits the results of read
302 requests into the indexed partition, to be consumed directly
303 by the kernel module.
304
305Responses to kernel requests are all packaged in pvfs2_downcall_t
306structs. Besides a few other members, pvfs2_downcall_t contains a
307union of structs, each of which is associated with a particular
308response type.
309
310The several members outside of the union are:
311 - int32_t type - type of operation.
312 - int32_t status - return code for the operation.
313 - int64_t trailer_size - 0 unless readdir operation.
314 - char *trailer_buf - initialized to NULL, used during readdir operations.
315
316The appropriate member inside the union is filled out for any
317particular response.
318
319 PVFS2_VFS_OP_FILE_IO
320 fill a pvfs2_io_response_t
321
322 PVFS2_VFS_OP_LOOKUP
323 fill a PVFS_object_kref
324
325 PVFS2_VFS_OP_CREATE
326 fill a PVFS_object_kref
327
328 PVFS2_VFS_OP_SYMLINK
329 fill a PVFS_object_kref
330
331 PVFS2_VFS_OP_GETATTR
332 fill in a PVFS_sys_attr_s (tons of stuff the kernel doesn't need)
333 fill in a string with the link target when the object is a symlink.
334
335 PVFS2_VFS_OP_MKDIR
336 fill a PVFS_object_kref
337
338 PVFS2_VFS_OP_STATFS
339 fill a pvfs2_statfs_response_t with useless info <g>. It is hard for
340 us to know, in a timely fashion, these statistics about our
341 distributed network filesystem.
342
343 PVFS2_VFS_OP_FS_MOUNT
344 fill a pvfs2_fs_mount_response_t which is just like a PVFS_object_kref
345 except its members are in a different order and "__pad1" is replaced
346 with "id".
347
348 PVFS2_VFS_OP_GETXATTR
349 fill a pvfs2_getxattr_response_t
350
351 PVFS2_VFS_OP_LISTXATTR
352 fill a pvfs2_listxattr_response_t
353
354 PVFS2_VFS_OP_PARAM
355 fill a pvfs2_param_response_t
356
357 PVFS2_VFS_OP_PERF_COUNT
358 fill a pvfs2_perf_count_response_t
359
360 PVFS2_VFS_OP_FSKEY
361 file a pvfs2_fs_key_response_t
362
363 PVFS2_VFS_OP_READDIR
364 jamb everything needed to represent a pvfs2_readdir_response_t into
365 the readdir buffer descriptor specified in the upcall.
366
Mike Marshall9f08cfe92016-02-26 14:39:08 -0500367Userspace uses writev() on /dev/pvfs2-req to pass responses to the requests
Mike Marshallfcac9d52016-01-13 14:28:13 -0500368made by the kernel side.
369
370A buffer_list containing:
371 - a pointer to the prepared response to the request from the
372 kernel (struct pvfs2_downcall_t).
373 - and also, in the case of a readdir request, a pointer to a
374 buffer containing descriptors for the objects in the target
375 directory.
376... is sent to the function (PINT_dev_write_list) which performs
377the writev.
378
379PINT_dev_write_list has a local iovec array: struct iovec io_array[10];
380
381The first four elements of io_array are initialized like this for all
382responses:
383
384 io_array[0].iov_base = address of local variable "proto_ver" (int32_t)
385 io_array[0].iov_len = sizeof(int32_t)
386
387 io_array[1].iov_base = address of global variable "pdev_magic" (int32_t)
388 io_array[1].iov_len = sizeof(int32_t)
389
390 io_array[2].iov_base = address of parameter "tag" (PVFS_id_gen_t)
391 io_array[2].iov_len = sizeof(int64_t)
392
393 io_array[3].iov_base = address of out_downcall member (pvfs2_downcall_t)
394 of global variable vfs_request (vfs_request_t)
395 io_array[3].iov_len = sizeof(pvfs2_downcall_t)
396
397Readdir responses initialize the fifth element io_array like this:
398
399 io_array[4].iov_base = contents of member trailer_buf (char *)
400 from out_downcall member of global variable
401 vfs_request
402 io_array[4].iov_len = contents of member trailer_size (PVFS_size)
403 from out_downcall member of global variable
404 vfs_request
405
406