summaryrefslogtreecommitdiff
path: root/Documentation/filesystems/orangefs.txt
blob: e1a0056a365ffd42c0e93aa1233d04beb40b4e72 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
ORANGEFS
========

OrangeFS is an LGPL userspace scale-out parallel storage system. It is ideal
for large storage problems faced by HPC, BigData, Streaming Video,
Genomics, Bioinformatics.

Orangefs, originally called PVFS, was first developed in 1993 by
Walt Ligon and Eric Blumer as a parallel file system for Parallel
Virtual Machine (PVM) as part of a NASA grant to study the I/O patterns
of parallel programs.

Orangefs features include:

  * Distributes file data among multiple file servers
  * Supports simultaneous access by multiple clients
  * Stores file data and metadata on servers using local file system
    and access methods
  * Userspace implementation is easy to install and maintain
  * Direct MPI support
  * Stateless


MAILING LIST
============

http://beowulf-underground.org/mailman/listinfo/pvfs2-users


DOCUMENTATION
=============

http://www.orangefs.org/documentation/


USERSPACE FILESYSTEM SOURCE
===========================

http://www.orangefs.org/download

Orangefs versions prior to 2.9.3 would not be compatible with the
upstream version of the kernel client.


BUILDING THE USERSPACE FILESYSTEM ON A SINGLE SERVER
====================================================

When Orangefs is upstream, "--with-kernel" shouldn't be needed, but
until then the path to where the kernel with the Orangefs kernel client
patch was built is needed to ensure that pvfs2-client-core (the bridge
between kernel space and user space) will build properly. You can omit
--prefix if you don't care that things are sprinkled around in
/usr/local.

./configure --prefix=/opt/ofs --with-kernel=/path/to/orangefs/kernel

make

make install

Create an orangefs config file:
/opt/ofs/bin/pvfs2-genconfig /etc/pvfs2.conf

  for "Enter hostnames", use the hostname, don't let it default to
  localhost.

create a pvfs2tab file in /etc:
cat /etc/pvfs2tab
tcp://myhostname:3334/orangefs /mymountpoint pvfs2 defaults,noauto 0 0

create the mount point you specified in the tab file if needed:
mkdir /mymountpoint

bootstrap the server:
/opt/ofs/sbin/pvfs2-server /etc/pvfs2.conf -f

start the server:
/opt/osf/sbin/pvfs2-server /etc/pvfs2.conf

Now the server is running. At this point you might like to
prove things are working with:

/opt/osf/bin/pvfs2-ls /mymountpoint

You might not want to enforce selinux, it doesn't seem to matter by
linux 3.11...

If stuff seems to be working, turn on the client core:
/opt/osf/sbin/pvfs2-client -p /opt/osf/sbin/pvfs2-client-core

Mount your filesystem.
mount -t pvfs2 tcp://myhostname:3334/orangefs /mymountpoint


OPTIONS
=======

The following mount options are accepted:

  acl
    Allow the use of Access Control Lists on files and directories.

  intr
    Some operations between the kernel client and the user space
    filesystem can be interruptible, such as changes in debug levels
    and the setting of tunable parameters.

  local_lock
    Enable posix locking from the perspective of "this" kernel. The
    default file_operations lock action is to return ENOSYS. Posix
    locking kicks in if the filesystem is mounted with -o local_lock.
    Distributed locking is being worked on for the future.


DEBUGGING
=========

If you want the debug (GOSSIP) statements in a particular
source file (inode.c for example) go to syslog:

  echo inode > /sys/kernel/debug/orangefs/kernel-debug

No debugging (the default):

  echo none > /sys/kernel/debug/orangefs/kernel-debug

Debugging from several source files:

  echo inode,dir > /sys/kernel/debug/orangefs/kernel-debug

All debugging:

  echo all > /sys/kernel/debug/orangefs/kernel-debug

Get a list of all debugging keywords:

  cat /sys/kernel/debug/orangefs/debug-help


PROTOCOL BETWEEN KERNEL MODULE AND USERSPACE
============================================

Orangefs is a user space filesystem and an associated kernel module.
We'll just refer to the user space part of Orangefs as "userspace"
from here on out. Orangefs descends from PVFS, and userspace code
still uses PVFS for function and variable names. Userspace typedefs
many of the important structures. Function and variable names in
the kernel module have been transitioned to "orangefs", and The Linux
Coding Style avoids typedefs, so kernel module structures that
correspond to userspace structures are not typedefed.

The kernel module implements a pseudo device that userspace
can read from and write to. Userspace can also manipulate the
kernel module through the pseudo device with ioctl.

THE BUFMAP:

At startup userspace allocates two page-size-aligned (posix_memalign)
mlocked memory buffers, one is used for IO and one is used for readdir
operations. The IO buffer is 41943040 bytes and the readdir buffer is
4194304 bytes. Each buffer contains logical chunks, or partitions, and
a pointer to each buffer is added to its own PVFS_dev_map_desc structure
which also describes its total size, as well as the size and number of
the partitions.

A pointer to the IO buffer's PVFS_dev_map_desc structure is sent to a
mapping routine in the kernel module with an ioctl. The structure is
copied from user space to kernel space with copy_from_user and is used
to initialize the kernel module's "bufmap" (struct orangefs_bufmap), which
then contains:

  * refcnt - a reference counter
  * desc_size - PVFS2_BUFMAP_DEFAULT_DESC_SIZE (4194304) - the IO buffer's
    partition size, which represents the filesystem's block size and
    is used for s_blocksize in super blocks.
  * desc_count - PVFS2_BUFMAP_DEFAULT_DESC_COUNT (10) - the number of
    partitions in the IO buffer.
  * desc_shift - log2(desc_size), used for s_blocksize_bits in super blocks.
  * total_size - the total size of the IO buffer.
  * page_count - the number of 4096 byte pages in the IO buffer.
  * page_array - a pointer to page_count * (sizeof(struct page*)) bytes
    of kcalloced memory. This memory is used as an array of pointers
    to each of the pages in the IO buffer through a call to get_user_pages.
  * desc_array - a pointer to desc_count * (sizeof(struct orangefs_bufmap_desc))
    bytes of kcalloced memory. This memory is further intialized:

      user_desc is the kernel's copy of the IO buffer's ORANGEFS_dev_map_desc
      structure. user_desc->ptr points to the IO buffer.

      pages_per_desc = bufmap->desc_size / PAGE_SIZE
      offset = 0

        bufmap->desc_array[0].page_array = &bufmap->page_array[offset]
        bufmap->desc_array[0].array_count = pages_per_desc = 1024
        bufmap->desc_array[0].uaddr = (user_desc->ptr) + (0 * 1024 * 4096)
        offset += 1024
                           .
                           .
                           .
        bufmap->desc_array[9].page_array = &bufmap->page_array[offset]
        bufmap->desc_array[9].array_count = pages_per_desc = 1024
        bufmap->desc_array[9].uaddr = (user_desc->ptr) +
                                               (9 * 1024 * 4096)
        offset += 1024

  * buffer_index_array - a desc_count sized array of ints, used to
    indicate which of the IO buffer's partitions are available to use.
  * buffer_index_lock - a spinlock to protect buffer_index_array during update.
  * readdir_index_array - a five (ORANGEFS_READDIR_DEFAULT_DESC_COUNT) element
    int array used to indicate which of the readdir buffer's partitions are
    available to use.
  * readdir_index_lock - a spinlock to protect readdir_index_array during
    update.

OPERATIONS:

The kernel module builds an "op" (struct orangefs_kernel_op_s) when it
needs to communicate with userspace. Part of the op contains the "upcall"
which expresses the request to userspace. Part of the op eventually
contains the "downcall" which expresses the results of the request.

The slab allocator is used to keep a cache of op structures handy.

At init time the kernel module defines and initializes a request list
and an in_progress hash table to keep track of all the ops that are
in flight at any given time.

Ops are stateful:

 * unknown  - op was just initialized
 * waiting  - op is on request_list (upward bound)
 * inprogr  - op is in progress (waiting for downcall)
 * serviced - op has matching downcall; ok
 * purged   - op has to start a timer since client-core
              exited uncleanly before servicing op
 * given up - submitter has given up waiting for it

When some arbitrary userspace program needs to perform a
filesystem operation on Orangefs (readdir, I/O, create, whatever)
an op structure is initialized and tagged with a distinguishing ID
number. The upcall part of the op is filled out, and the op is
passed to the "service_operation" function.

Service_operation changes the op's state to "waiting", puts
it on the request list, and signals the Orangefs file_operations.poll
function through a wait queue. Userspace is polling the pseudo-device
and thus becomes aware of the upcall request that needs to be read.

When the Orangefs file_operations.read function is triggered, the
request list is searched for an op that seems ready-to-process.
The op is removed from the request list. The tag from the op and
the filled-out upcall struct are copy_to_user'ed back to userspace.

If any of these (and some additional protocol) copy_to_users fail,
the op's state is set to "waiting" and the op is added back to
the request list. Otherwise, the op's state is changed to "in progress",
and the op is hashed on its tag and put onto the end of a list in the
in_progress hash table at the index the tag hashed to.

When userspace has assembled the response to the upcall, it
writes the response, which includes the distinguishing tag, back to
the pseudo device in a series of io_vecs. This triggers the Orangefs
file_operations.write_iter function to find the op with the associated
tag and remove it from the in_progress hash table. As long as the op's
state is not "canceled" or "given up", its state is set to "serviced".
The file_operations.write_iter function returns to the waiting vfs,
and back to service_operation through wait_for_matching_downcall.

Service operation returns to its caller with the op's downcall
part (the response to the upcall) filled out.

The "client-core" is the bridge between the kernel module and
userspace. The client-core is a daemon. The client-core has an
associated watchdog daemon. If the client-core is ever signaled
to die, the watchdog daemon restarts the client-core. Even though
the client-core is restarted "right away", there is a period of
time during such an event that the client-core is dead. A dead client-core
can't be triggered by the Orangefs file_operations.poll function.
Ops that pass through service_operation during a "dead spell" can timeout
on the wait queue and one attempt is made to recycle them. Obviously,
if the client-core stays dead too long, the arbitrary userspace processes
trying to use Orangefs will be negatively affected. Waiting ops
that can't be serviced will be removed from the request list and
have their states set to "given up". In-progress ops that can't 
be serviced will be removed from the in_progress hash table and
have their states set to "given up".

Readdir and I/O ops are atypical with respect to their payloads.

  - readdir ops use the smaller of the two pre-allocated pre-partitioned
    memory buffers. The readdir buffer is only available to userspace.
    The kernel module obtains an index to a free partition before launching
    a readdir op. Userspace deposits the results into the indexed partition
    and then writes them to back to the pvfs device.

  - io (read and write) ops use the larger of the two pre-allocated
    pre-partitioned memory buffers. The IO buffer is accessible from
    both userspace and the kernel module. The kernel module obtains an
    index to a free partition before launching an io op. The kernel module
    deposits write data into the indexed partition, to be consumed
    directly by userspace. Userspace deposits the results of read
    requests into the indexed partition, to be consumed directly
    by the kernel module.

Responses to kernel requests are all packaged in pvfs2_downcall_t
structs. Besides a few other members, pvfs2_downcall_t contains a
union of structs, each of which is associated with a particular
response type.

The several members outside of the union are:
 - int32_t type - type of operation.
 - int32_t status - return code for the operation.
 - int64_t trailer_size - 0 unless readdir operation.
 - char *trailer_buf - initialized to NULL, used during readdir operations.

The appropriate member inside the union is filled out for any
particular response.

  PVFS2_VFS_OP_FILE_IO
    fill a pvfs2_io_response_t

  PVFS2_VFS_OP_LOOKUP
    fill a PVFS_object_kref

  PVFS2_VFS_OP_CREATE
    fill a PVFS_object_kref

  PVFS2_VFS_OP_SYMLINK
    fill a PVFS_object_kref

  PVFS2_VFS_OP_GETATTR
    fill in a PVFS_sys_attr_s (tons of stuff the kernel doesn't need)
    fill in a string with the link target when the object is a symlink.

  PVFS2_VFS_OP_MKDIR
    fill a PVFS_object_kref

  PVFS2_VFS_OP_STATFS
    fill a pvfs2_statfs_response_t with useless info <g>. It is hard for
    us to know, in a timely fashion, these statistics about our
    distributed network filesystem. 

  PVFS2_VFS_OP_FS_MOUNT
    fill a pvfs2_fs_mount_response_t which is just like a PVFS_object_kref
    except its members are in a different order and "__pad1" is replaced
    with "id".

  PVFS2_VFS_OP_GETXATTR
    fill a pvfs2_getxattr_response_t

  PVFS2_VFS_OP_LISTXATTR
    fill a pvfs2_listxattr_response_t

  PVFS2_VFS_OP_PARAM
    fill a pvfs2_param_response_t

  PVFS2_VFS_OP_PERF_COUNT
    fill a pvfs2_perf_count_response_t

  PVFS2_VFS_OP_FSKEY
    file a pvfs2_fs_key_response_t

  PVFS2_VFS_OP_READDIR
    jamb everything needed to represent a pvfs2_readdir_response_t into
    the readdir buffer descriptor specified in the upcall.

Userspace uses writev() on /dev/pvfs2-req to pass responses to the requests
made by the kernel side.

A buffer_list containing:
  - a pointer to the prepared response to the request from the
    kernel (struct pvfs2_downcall_t).
  - and also, in the case of a readdir request, a pointer to a
    buffer containing descriptors for the objects in the target
    directory.
... is sent to the function (PINT_dev_write_list) which performs
the writev.

PINT_dev_write_list has a local iovec array: struct iovec io_array[10];

The first four elements of io_array are initialized like this for all
responses:

  io_array[0].iov_base = address of local variable "proto_ver" (int32_t)
  io_array[0].iov_len = sizeof(int32_t)

  io_array[1].iov_base = address of global variable "pdev_magic" (int32_t)
  io_array[1].iov_len = sizeof(int32_t)
  
  io_array[2].iov_base = address of parameter "tag" (PVFS_id_gen_t)
  io_array[2].iov_len = sizeof(int64_t)

  io_array[3].iov_base = address of out_downcall member (pvfs2_downcall_t)
                         of global variable vfs_request (vfs_request_t)
  io_array[3].iov_len = sizeof(pvfs2_downcall_t)

Readdir responses initialize the fifth element io_array like this:

  io_array[4].iov_base = contents of member trailer_buf (char *)
                         from out_downcall member of global variable
                         vfs_request
  io_array[4].iov_len = contents of member trailer_size (PVFS_size)
                        from out_downcall member of global variable
                        vfs_request