Merge branch 'fc/remote-bzr'
[git/git.git] / Documentation / technical / pack-protocol.txt
1 Packfile transfer protocols
2 ===========================
3
4 Git supports transferring data in packfiles over the ssh://, git:// and
5 file:// transports. There exist two sets of protocols, one for pushing
6 data from a client to a server and another for fetching data from a
7 server to a client. All three transports (ssh, git, file) use the same
8 protocol to transfer data.
9
10 The processes invoked in the canonical Git implementation are 'upload-pack'
11 on the server side and 'fetch-pack' on the client side for fetching data;
12 then 'receive-pack' on the server and 'send-pack' on the client for pushing
13 data. The protocol functions to have a server tell a client what is
14 currently on the server, then for the two to negotiate the smallest amount
15 of data to send in order to fully update one or the other.
16
17 Transports
18 ----------
19 There are three transports over which the packfile protocol is
20 initiated. The Git transport is a simple, unauthenticated server that
21 takes the command (almost always 'upload-pack', though Git
22 servers can be configured to be globally writable, in which 'receive-
23 pack' initiation is also allowed) with which the client wishes to
24 communicate and executes it and connects it to the requesting
25 process.
26
27 In the SSH transport, the client just runs the 'upload-pack'
28 or 'receive-pack' process on the server over the SSH protocol and then
29 communicates with that invoked process over the SSH connection.
30
31 The file:// transport runs the 'upload-pack' or 'receive-pack'
32 process locally and communicates with it over a pipe.
33
34 Git Transport
35 -------------
36
37 The Git transport starts off by sending the command and repository
38 on the wire using the pkt-line format, followed by a NUL byte and a
39 hostname parameter, terminated by a NUL byte.
40
41 0032git-upload-pack /project.git\0host=myserver.com\0
42
43 --
44 git-proto-request = request-command SP pathname NUL [ host-parameter NUL ]
45 request-command = "git-upload-pack" / "git-receive-pack" /
46 "git-upload-archive" ; case sensitive
47 pathname = *( %x01-ff ) ; exclude NUL
48 host-parameter = "host=" hostname [ ":" port ]
49 --
50
51 Only host-parameter is allowed in the git-proto-request. Clients
52 MUST NOT attempt to send additional parameters. It is used for the
53 git-daemon name based virtual hosting. See --interpolated-path
54 option to git daemon, with the %H/%CH format characters.
55
56 Basically what the Git client is doing to connect to an 'upload-pack'
57 process on the server side over the Git protocol is this:
58
59 $ echo -e -n \
60 "0039git-upload-pack /schacon/gitbook.git\0host=example.com\0" |
61 nc -v example.com 9418
62
63 If the server refuses the request for some reasons, it could abort
64 gracefully with an error message.
65
66 ----
67 error-line = PKT-LINE("ERR" SP explanation-text)
68 ----
69
70
71 SSH Transport
72 -------------
73
74 Initiating the upload-pack or receive-pack processes over SSH is
75 executing the binary on the server via SSH remote execution.
76 It is basically equivalent to running this:
77
78 $ ssh git.example.com "git-upload-pack '/project.git'"
79
80 For a server to support Git pushing and pulling for a given user over
81 SSH, that user needs to be able to execute one or both of those
82 commands via the SSH shell that they are provided on login. On some
83 systems, that shell access is limited to only being able to run those
84 two commands, or even just one of them.
85
86 In an ssh:// format URI, it's absolute in the URI, so the '/' after
87 the host name (or port number) is sent as an argument, which is then
88 read by the remote git-upload-pack exactly as is, so it's effectively
89 an absolute path in the remote filesystem.
90
91 git clone ssh://user@example.com/project.git
92 |
93 v
94 ssh user@example.com "git-upload-pack '/project.git'"
95
96 In a "user@host:path" format URI, its relative to the user's home
97 directory, because the Git client will run:
98
99 git clone user@example.com:project.git
100 |
101 v
102 ssh user@example.com "git-upload-pack 'project.git'"
103
104 The exception is if a '~' is used, in which case
105 we execute it without the leading '/'.
106
107 ssh://user@example.com/~alice/project.git,
108 |
109 v
110 ssh user@example.com "git-upload-pack '~alice/project.git'"
111
112 A few things to remember here:
113
114 - The "command name" is spelled with dash (e.g. git-upload-pack), but
115 this can be overridden by the client;
116
117 - The repository path is always quoted with single quotes.
118
119 Fetching Data From a Server
120 ---------------------------
121
122 When one Git repository wants to get data that a second repository
123 has, the first can 'fetch' from the second. This operation determines
124 what data the server has that the client does not then streams that
125 data down to the client in packfile format.
126
127
128 Reference Discovery
129 -------------------
130
131 When the client initially connects the server will immediately respond
132 with a listing of each reference it has (all branches and tags) along
133 with the object name that each reference currently points to.
134
135 $ echo -e -n "0039git-upload-pack /schacon/gitbook.git\0host=example.com\0" |
136 nc -v example.com 9418
137 00887217a7c7e582c46cec22a130adf4b9d7d950fba0 HEAD\0multi_ack thin-pack
138 side-band side-band-64k ofs-delta shallow no-progress include-tag
139 00441d3fcd5ced445d1abc402225c0b8a1299641f497 refs/heads/integration
140 003f7217a7c7e582c46cec22a130adf4b9d7d950fba0 refs/heads/master
141 003cb88d2441cac0977faf98efc80305012112238d9d refs/tags/v0.9
142 003c525128480b96c89e6418b1e40909bf6c5b2d580f refs/tags/v1.0
143 003fe92df48743b7bc7d26bcaabfddde0a1e20cae47c refs/tags/v1.0^{}
144 0000
145
146 Server SHOULD terminate each non-flush line using LF ("\n") terminator;
147 client MUST NOT complain if there is no terminator.
148
149 The returned response is a pkt-line stream describing each ref and
150 its current value. The stream MUST be sorted by name according to
151 the C locale ordering.
152
153 If HEAD is a valid ref, HEAD MUST appear as the first advertised
154 ref. If HEAD is not a valid ref, HEAD MUST NOT appear in the
155 advertisement list at all, but other refs may still appear.
156
157 The stream MUST include capability declarations behind a NUL on the
158 first ref. The peeled value of a ref (that is "ref^{}") MUST be
159 immediately after the ref itself, if presented. A conforming server
160 MUST peel the ref if it's an annotated tag.
161
162 ----
163 advertised-refs = (no-refs / list-of-refs)
164 flush-pkt
165
166 no-refs = PKT-LINE(zero-id SP "capabilities^{}"
167 NUL capability-list LF)
168
169 list-of-refs = first-ref *other-ref
170 first-ref = PKT-LINE(obj-id SP refname
171 NUL capability-list LF)
172
173 other-ref = PKT-LINE(other-tip / other-peeled)
174 other-tip = obj-id SP refname LF
175 other-peeled = obj-id SP refname "^{}" LF
176
177 capability-list = capability *(SP capability)
178 capability = 1*(LC_ALPHA / DIGIT / "-" / "_")
179 LC_ALPHA = %x61-7A
180 ----
181
182 Server and client MUST use lowercase for obj-id, both MUST treat obj-id
183 as case-insensitive.
184
185 See protocol-capabilities.txt for a list of allowed server capabilities
186 and descriptions.
187
188 Packfile Negotiation
189 --------------------
190 After reference and capabilities discovery, the client can decide to
191 terminate the connection by sending a flush-pkt, telling the server it can
192 now gracefully terminate, and disconnect, when it does not need any pack
193 data. This can happen with the ls-remote command, and also can happen when
194 the client already is up-to-date.
195
196 Otherwise, it enters the negotiation phase, where the client and
197 server determine what the minimal packfile necessary for transport is,
198 by telling the server what objects it wants, its shallow objects
199 (if any), and the maximum commit depth it wants (if any). The client
200 will also send a list of the capabilities it wants to be in effect,
201 out of what the server said it could do with the first 'want' line.
202
203 ----
204 upload-request = want-list
205 *shallow-line
206 *1depth-request
207 flush-pkt
208
209 want-list = first-want
210 *additional-want
211
212 shallow-line = PKT_LINE("shallow" SP obj-id)
213
214 depth-request = PKT_LINE("deepen" SP depth)
215
216 first-want = PKT-LINE("want" SP obj-id SP capability-list LF)
217 additional-want = PKT-LINE("want" SP obj-id LF)
218
219 depth = 1*DIGIT
220 ----
221
222 Clients MUST send all the obj-ids it wants from the reference
223 discovery phase as 'want' lines. Clients MUST send at least one
224 'want' command in the request body. Clients MUST NOT mention an
225 obj-id in a 'want' command which did not appear in the response
226 obtained through ref discovery.
227
228 The client MUST write all obj-ids which it only has shallow copies
229 of (meaning that it does not have the parents of a commit) as
230 'shallow' lines so that the server is aware of the limitations of
231 the client's history. Clients MUST NOT mention an obj-id which
232 it does not know exists on the server.
233
234 The client now sends the maximum commit history depth it wants for
235 this transaction, which is the number of commits it wants from the
236 tip of the history, if any, as a 'deepen' line. A depth of 0 is the
237 same as not making a depth request. The client does not want to receive
238 any commits beyond this depth, nor objects needed only to complete
239 those commits. Commits whose parents are not received as a result are
240 defined as shallow and marked as such in the server. This information
241 is sent back to the client in the next step.
242
243 Once all the 'want's and 'shallow's (and optional 'deepen') are
244 transferred, clients MUST send a flush-pkt, to tell the server side
245 that it is done sending the list.
246
247 Otherwise, if the client sent a positive depth request, the server
248 will determine which commits will and will not be shallow and
249 send this information to the client. If the client did not request
250 a positive depth, this step is skipped.
251
252 ----
253 shallow-update = *shallow-line
254 *unshallow-line
255 flush-pkt
256
257 shallow-line = PKT-LINE("shallow" SP obj-id)
258
259 unshallow-line = PKT-LINE("unshallow" SP obj-id)
260 ----
261
262 If the client has requested a positive depth, the server will compute
263 the set of commits which are no deeper than the desired depth. The set
264 of commits start at the client's wants.
265
266 The server writes 'shallow' lines for each
267 commit whose parents will not be sent as a result. The server writes
268 an 'unshallow' line for each commit which the client has indicated is
269 shallow, but is no longer shallow at the currently requested depth
270 (that is, its parents will now be sent). The server MUST NOT mark
271 as unshallow anything which the client has not indicated was shallow.
272
273 Now the client will send a list of the obj-ids it has using 'have'
274 lines, so the server can make a packfile that only contains the objects
275 that the client needs. In multi_ack mode, the canonical implementation
276 will send up to 32 of these at a time, then will send a flush-pkt. The
277 canonical implementation will skip ahead and send the next 32 immediately,
278 so that there is always a block of 32 "in-flight on the wire" at a time.
279
280 ----
281 upload-haves = have-list
282 compute-end
283
284 have-list = *have-line
285 have-line = PKT-LINE("have" SP obj-id LF)
286 compute-end = flush-pkt / PKT-LINE("done")
287 ----
288
289 If the server reads 'have' lines, it then will respond by ACKing any
290 of the obj-ids the client said it had that the server also has. The
291 server will ACK obj-ids differently depending on which ack mode is
292 chosen by the client.
293
294 In multi_ack mode:
295
296 * the server will respond with 'ACK obj-id continue' for any common
297 commits.
298
299 * once the server has found an acceptable common base commit and is
300 ready to make a packfile, it will blindly ACK all 'have' obj-ids
301 back to the client.
302
303 * the server will then send a 'NACK' and then wait for another response
304 from the client - either a 'done' or another list of 'have' lines.
305
306 In multi_ack_detailed mode:
307
308 * the server will differentiate the ACKs where it is signaling
309 that it is ready to send data with 'ACK obj-id ready' lines, and
310 signals the identified common commits with 'ACK obj-id common' lines.
311
312 Without either multi_ack or multi_ack_detailed:
313
314 * upload-pack sends "ACK obj-id" on the first common object it finds.
315 After that it says nothing until the client gives it a "done".
316
317 * upload-pack sends "NAK" on a flush-pkt if no common object
318 has been found yet. If one has been found, and thus an ACK
319 was already sent, it's silent on the flush-pkt.
320
321 After the client has gotten enough ACK responses that it can determine
322 that the server has enough information to send an efficient packfile
323 (in the canonical implementation, this is determined when it has received
324 enough ACKs that it can color everything left in the --date-order queue
325 as common with the server, or the --date-order queue is empty), or the
326 client determines that it wants to give up (in the canonical implementation,
327 this is determined when the client sends 256 'have' lines without getting
328 any of them ACKed by the server - meaning there is nothing in common and
329 the server should just send all of its objects), then the client will send
330 a 'done' command. The 'done' command signals to the server that the client
331 is ready to receive its packfile data.
332
333 However, the 256 limit *only* turns on in the canonical client
334 implementation if we have received at least one "ACK %s continue"
335 during a prior round. This helps to ensure that at least one common
336 ancestor is found before we give up entirely.
337
338 Once the 'done' line is read from the client, the server will either
339 send a final 'ACK obj-id' or it will send a 'NAK'. The server only sends
340 ACK after 'done' if there is at least one common base and multi_ack or
341 multi_ack_detailed is enabled. The server always sends NAK after 'done'
342 if there is no common base found.
343
344 Then the server will start sending its packfile data.
345
346 ----
347 server-response = *ack_multi ack / nak
348 ack_multi = PKT-LINE("ACK" SP obj-id ack_status LF)
349 ack_status = "continue" / "common" / "ready"
350 ack = PKT-LINE("ACK SP obj-id LF)
351 nak = PKT-LINE("NAK" LF)
352 ----
353
354 A simple clone may look like this (with no 'have' lines):
355
356 ----
357 C: 0054want 74730d410fcb6603ace96f1dc55ea6196122532d multi_ack \
358 side-band-64k ofs-delta\n
359 C: 0032want 7d1665144a3a975c05f1f43902ddaf084e784dbe\n
360 C: 0032want 5a3f6be755bbb7deae50065988cbfa1ffa9ab68a\n
361 C: 0032want 7e47fe2bd8d01d481f44d7af0531bd93d3b21c01\n
362 C: 0032want 74730d410fcb6603ace96f1dc55ea6196122532d\n
363 C: 0000
364 C: 0009done\n
365
366 S: 0008NAK\n
367 S: [PACKFILE]
368 ----
369
370 An incremental update (fetch) response might look like this:
371
372 ----
373 C: 0054want 74730d410fcb6603ace96f1dc55ea6196122532d multi_ack \
374 side-band-64k ofs-delta\n
375 C: 0032want 7d1665144a3a975c05f1f43902ddaf084e784dbe\n
376 C: 0032want 5a3f6be755bbb7deae50065988cbfa1ffa9ab68a\n
377 C: 0000
378 C: 0032have 7e47fe2bd8d01d481f44d7af0531bd93d3b21c01\n
379 C: [30 more have lines]
380 C: 0032have 74730d410fcb6603ace96f1dc55ea6196122532d\n
381 C: 0000
382
383 S: 003aACK 7e47fe2bd8d01d481f44d7af0531bd93d3b21c01 continue\n
384 S: 003aACK 74730d410fcb6603ace96f1dc55ea6196122532d continue\n
385 S: 0008NAK\n
386
387 C: 0009done\n
388
389 S: 0031ACK 74730d410fcb6603ace96f1dc55ea6196122532d\n
390 S: [PACKFILE]
391 ----
392
393
394 Packfile Data
395 -------------
396
397 Now that the client and server have finished negotiation about what
398 the minimal amount of data that needs to be sent to the client is, the server
399 will construct and send the required data in packfile format.
400
401 See pack-format.txt for what the packfile itself actually looks like.
402
403 If 'side-band' or 'side-band-64k' capabilities have been specified by
404 the client, the server will send the packfile data multiplexed.
405
406 Each packet starting with the packet-line length of the amount of data
407 that follows, followed by a single byte specifying the sideband the
408 following data is coming in on.
409
410 In 'side-band' mode, it will send up to 999 data bytes plus 1 control
411 code, for a total of up to 1000 bytes in a pkt-line. In 'side-band-64k'
412 mode it will send up to 65519 data bytes plus 1 control code, for a
413 total of up to 65520 bytes in a pkt-line.
414
415 The sideband byte will be a '1', '2' or a '3'. Sideband '1' will contain
416 packfile data, sideband '2' will be used for progress information that the
417 client will generally print to stderr and sideband '3' is used for error
418 information.
419
420 If no 'side-band' capability was specified, the server will stream the
421 entire packfile without multiplexing.
422
423
424 Pushing Data To a Server
425 ------------------------
426
427 Pushing data to a server will invoke the 'receive-pack' process on the
428 server, which will allow the client to tell it which references it should
429 update and then send all the data the server will need for those new
430 references to be complete. Once all the data is received and validated,
431 the server will then update its references to what the client specified.
432
433 Authentication
434 --------------
435
436 The protocol itself contains no authentication mechanisms. That is to be
437 handled by the transport, such as SSH, before the 'receive-pack' process is
438 invoked. If 'receive-pack' is configured over the Git transport, those
439 repositories will be writable by anyone who can access that port (9418) as
440 that transport is unauthenticated.
441
442 Reference Discovery
443 -------------------
444
445 The reference discovery phase is done nearly the same way as it is in the
446 fetching protocol. Each reference obj-id and name on the server is sent
447 in packet-line format to the client, followed by a flush-pkt. The only
448 real difference is that the capability listing is different - the only
449 possible values are 'report-status', 'delete-refs' and 'ofs-delta'.
450
451 Reference Update Request and Packfile Transfer
452 ----------------------------------------------
453
454 Once the client knows what references the server is at, it can send a
455 list of reference update requests. For each reference on the server
456 that it wants to update, it sends a line listing the obj-id currently on
457 the server, the obj-id the client would like to update it to and the name
458 of the reference.
459
460 This list is followed by a flush-pkt and then the packfile that should
461 contain all the objects that the server will need to complete the new
462 references.
463
464 ----
465 update-request = command-list [pack-file]
466
467 command-list = PKT-LINE(command NUL capability-list LF)
468 *PKT-LINE(command LF)
469 flush-pkt
470
471 command = create / delete / update
472 create = zero-id SP new-id SP name
473 delete = old-id SP zero-id SP name
474 update = old-id SP new-id SP name
475
476 old-id = obj-id
477 new-id = obj-id
478
479 pack-file = "PACK" 28*(OCTET)
480 ----
481
482 If the receiving end does not support delete-refs, the sending end MUST
483 NOT ask for delete command.
484
485 The pack-file MUST NOT be sent if the only command used is 'delete'.
486
487 A pack-file MUST be sent if either create or update command is used,
488 even if the server already has all the necessary objects. In this
489 case the client MUST send an empty pack-file. The only time this
490 is likely to happen is if the client is creating
491 a new branch or a tag that points to an existing obj-id.
492
493 The server will receive the packfile, unpack it, then validate each
494 reference that is being updated that it hasn't changed while the request
495 was being processed (the obj-id is still the same as the old-id), and
496 it will run any update hooks to make sure that the update is acceptable.
497 If all of that is fine, the server will then update the references.
498
499 Report Status
500 -------------
501
502 After receiving the pack data from the sender, the receiver sends a
503 report if 'report-status' capability is in effect.
504 It is a short listing of what happened in that update. It will first
505 list the status of the packfile unpacking as either 'unpack ok' or
506 'unpack [error]'. Then it will list the status for each of the references
507 that it tried to update. Each line is either 'ok [refname]' if the
508 update was successful, or 'ng [refname] [error]' if the update was not.
509
510 ----
511 report-status = unpack-status
512 1*(command-status)
513 flush-pkt
514
515 unpack-status = PKT-LINE("unpack" SP unpack-result LF)
516 unpack-result = "ok" / error-msg
517
518 command-status = command-ok / command-fail
519 command-ok = PKT-LINE("ok" SP refname LF)
520 command-fail = PKT-LINE("ng" SP refname SP error-msg LF)
521
522 error-msg = 1*(OCTECT) ; where not "ok"
523 ----
524
525 Updates can be unsuccessful for a number of reasons. The reference can have
526 changed since the reference discovery phase was originally sent, meaning
527 someone pushed in the meantime. The reference being pushed could be a
528 non-fast-forward reference and the update hooks or configuration could be
529 set to not allow that, etc. Also, some references can be updated while others
530 can be rejected.
531
532 An example client/server communication might look like this:
533
534 ----
535 S: 007c74730d410fcb6603ace96f1dc55ea6196122532d refs/heads/local\0report-status delete-refs ofs-delta\n
536 S: 003e7d1665144a3a975c05f1f43902ddaf084e784dbe refs/heads/debug\n
537 S: 003f74730d410fcb6603ace96f1dc55ea6196122532d refs/heads/master\n
538 S: 003f74730d410fcb6603ace96f1dc55ea6196122532d refs/heads/team\n
539 S: 0000
540
541 C: 003e7d1665144a3a975c05f1f43902ddaf084e784dbe 74730d410fcb6603ace96f1dc55ea6196122532d refs/heads/debug\n
542 C: 003e74730d410fcb6603ace96f1dc55ea6196122532d 5a3f6be755bbb7deae50065988cbfa1ffa9ab68a refs/heads/master\n
543 C: 0000
544 C: [PACKDATA]
545
546 S: 000eunpack ok\n
547 S: 0018ok refs/heads/debug\n
548 S: 002ang refs/heads/master non-fast-forward\n
549 ----