upload-pack: add deepen-since to cut shallow repos based on time
[git/git.git] / Documentation / technical / pack-protocol.txt
1 Packfile transfer protocols
2 ===========================
4 Git supports transferring data in packfiles over the ssh://, git://, http:// 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. The three transports (ssh, git, file) use the same
8 protocol to transfer data. http is documented in http-protocol.txt.
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.
17 pkt-line Format
18 ---------------
20 The descriptions below build on the pkt-line format described in
21 protocol-common.txt. When the grammar indicate `PKT-LINE(...)`, unless
22 otherwise noted the usual pkt-line LF rules apply: the sender SHOULD
23 include a LF, but the receiver MUST NOT complain if it is not present.
25 Transports
26 ----------
27 There are three transports over which the packfile protocol is
28 initiated. The Git transport is a simple, unauthenticated server that
29 takes the command (almost always 'upload-pack', though Git
30 servers can be configured to be globally writable, in which 'receive-
31 pack' initiation is also allowed) with which the client wishes to
32 communicate and executes it and connects it to the requesting
33 process.
35 In the SSH transport, the client just runs the 'upload-pack'
36 or 'receive-pack' process on the server over the SSH protocol and then
37 communicates with that invoked process over the SSH connection.
39 The file:// transport runs the 'upload-pack' or 'receive-pack'
40 process locally and communicates with it over a pipe.
42 Git Transport
43 -------------
45 The Git transport starts off by sending the command and repository
46 on the wire using the pkt-line format, followed by a NUL byte and a
47 hostname parameter, terminated by a NUL byte.
49 0032git-upload-pack /project.git\0host=myserver.com\0
51 --
52 git-proto-request = request-command SP pathname NUL [ host-parameter NUL ]
53 request-command = "git-upload-pack" / "git-receive-pack" /
54 "git-upload-archive" ; case sensitive
55 pathname = *( %x01-ff ) ; exclude NUL
56 host-parameter = "host=" hostname [ ":" port ]
57 --
59 Only host-parameter is allowed in the git-proto-request. Clients
60 MUST NOT attempt to send additional parameters. It is used for the
61 git-daemon name based virtual hosting. See --interpolated-path
62 option to git daemon, with the %H/%CH format characters.
64 Basically what the Git client is doing to connect to an 'upload-pack'
65 process on the server side over the Git protocol is this:
67 $ echo -e -n \
68 "0039git-upload-pack /schacon/gitbook.git\0host=example.com\0" |
69 nc -v example.com 9418
71 If the server refuses the request for some reasons, it could abort
72 gracefully with an error message.
74 ----
75 error-line = PKT-LINE("ERR" SP explanation-text)
76 ----
79 SSH Transport
80 -------------
82 Initiating the upload-pack or receive-pack processes over SSH is
83 executing the binary on the server via SSH remote execution.
84 It is basically equivalent to running this:
86 $ ssh git.example.com "git-upload-pack '/project.git'"
88 For a server to support Git pushing and pulling for a given user over
89 SSH, that user needs to be able to execute one or both of those
90 commands via the SSH shell that they are provided on login. On some
91 systems, that shell access is limited to only being able to run those
92 two commands, or even just one of them.
94 In an ssh:// format URI, it's absolute in the URI, so the '/' after
95 the host name (or port number) is sent as an argument, which is then
96 read by the remote git-upload-pack exactly as is, so it's effectively
97 an absolute path in the remote filesystem.
99 git clone ssh://user@example.com/project.git
100 |
101 v
102 ssh user@example.com "git-upload-pack '/project.git'"
104 In a "user@host:path" format URI, its relative to the user's home
105 directory, because the Git client will run:
107 git clone user@example.com:project.git
108 |
109 v
110 ssh user@example.com "git-upload-pack 'project.git'"
112 The exception is if a '~' is used, in which case
113 we execute it without the leading '/'.
115 ssh://user@example.com/~alice/project.git,
116 |
117 v
118 ssh user@example.com "git-upload-pack '~alice/project.git'"
120 A few things to remember here:
122 - The "command name" is spelled with dash (e.g. git-upload-pack), but
123 this can be overridden by the client;
125 - The repository path is always quoted with single quotes.
127 Fetching Data From a Server
128 ---------------------------
130 When one Git repository wants to get data that a second repository
131 has, the first can 'fetch' from the second. This operation determines
132 what data the server has that the client does not then streams that
133 data down to the client in packfile format.
136 Reference Discovery
137 -------------------
139 When the client initially connects the server will immediately respond
140 with a listing of each reference it has (all branches and tags) along
141 with the object name that each reference currently points to.
143 $ echo -e -n "0039git-upload-pack /schacon/gitbook.git\0host=example.com\0" |
144 nc -v example.com 9418
145 00887217a7c7e582c46cec22a130adf4b9d7d950fba0 HEAD\0multi_ack thin-pack
146 side-band side-band-64k ofs-delta shallow no-progress include-tag
147 00441d3fcd5ced445d1abc402225c0b8a1299641f497 refs/heads/integration
148 003f7217a7c7e582c46cec22a130adf4b9d7d950fba0 refs/heads/master
149 003cb88d2441cac0977faf98efc80305012112238d9d refs/tags/v0.9
150 003c525128480b96c89e6418b1e40909bf6c5b2d580f refs/tags/v1.0
151 003fe92df48743b7bc7d26bcaabfddde0a1e20cae47c refs/tags/v1.0^{}
152 0000
154 The returned response is a pkt-line stream describing each ref and
155 its current value. The stream MUST be sorted by name according to
156 the C locale ordering.
158 If HEAD is a valid ref, HEAD MUST appear as the first advertised
159 ref. If HEAD is not a valid ref, HEAD MUST NOT appear in the
160 advertisement list at all, but other refs may still appear.
162 The stream MUST include capability declarations behind a NUL on the
163 first ref. The peeled value of a ref (that is "ref^{}") MUST be
164 immediately after the ref itself, if presented. A conforming server
165 MUST peel the ref if it's an annotated tag.
167 ----
168 advertised-refs = (no-refs / list-of-refs)
169 *shallow
170 flush-pkt
172 no-refs = PKT-LINE(zero-id SP "capabilities^{}"
173 NUL capability-list)
175 list-of-refs = first-ref *other-ref
176 first-ref = PKT-LINE(obj-id SP refname
177 NUL capability-list)
179 other-ref = PKT-LINE(other-tip / other-peeled)
180 other-tip = obj-id SP refname
181 other-peeled = obj-id SP refname "^{}"
183 shallow = PKT-LINE("shallow" SP obj-id)
185 capability-list = capability *(SP capability)
186 capability = 1*(LC_ALPHA / DIGIT / "-" / "_")
187 LC_ALPHA = %x61-7A
188 ----
190 Server and client MUST use lowercase for obj-id, both MUST treat obj-id
191 as case-insensitive.
193 See protocol-capabilities.txt for a list of allowed server capabilities
194 and descriptions.
196 Packfile Negotiation
197 --------------------
198 After reference and capabilities discovery, the client can decide to
199 terminate the connection by sending a flush-pkt, telling the server it can
200 now gracefully terminate, and disconnect, when it does not need any pack
201 data. This can happen with the ls-remote command, and also can happen when
202 the client already is up-to-date.
204 Otherwise, it enters the negotiation phase, where the client and
205 server determine what the minimal packfile necessary for transport is,
206 by telling the server what objects it wants, its shallow objects
207 (if any), and the maximum commit depth it wants (if any). The client
208 will also send a list of the capabilities it wants to be in effect,
209 out of what the server said it could do with the first 'want' line.
211 ----
212 upload-request = want-list
213 *shallow-line
214 *1depth-request
215 flush-pkt
217 want-list = first-want
218 *additional-want
220 shallow-line = PKT-LINE("shallow" SP obj-id)
222 depth-request = PKT-LINE("deepen" SP depth) /
223 PKT-LINE("deepen-since" SP timestamp)
225 first-want = PKT-LINE("want" SP obj-id SP capability-list)
226 additional-want = PKT-LINE("want" SP obj-id)
228 depth = 1*DIGIT
229 ----
231 Clients MUST send all the obj-ids it wants from the reference
232 discovery phase as 'want' lines. Clients MUST send at least one
233 'want' command in the request body. Clients MUST NOT mention an
234 obj-id in a 'want' command which did not appear in the response
235 obtained through ref discovery.
237 The client MUST write all obj-ids which it only has shallow copies
238 of (meaning that it does not have the parents of a commit) as
239 'shallow' lines so that the server is aware of the limitations of
240 the client's history.
242 The client now sends the maximum commit history depth it wants for
243 this transaction, which is the number of commits it wants from the
244 tip of the history, if any, as a 'deepen' line. A depth of 0 is the
245 same as not making a depth request. The client does not want to receive
246 any commits beyond this depth, nor does it want objects needed only to
247 complete those commits. Commits whose parents are not received as a
248 result are defined as shallow and marked as such in the server. This
249 information is sent back to the client in the next step.
251 Once all the 'want's and 'shallow's (and optional 'deepen') are
252 transferred, clients MUST send a flush-pkt, to tell the server side
253 that it is done sending the list.
255 Otherwise, if the client sent a positive depth request, the server
256 will determine which commits will and will not be shallow and
257 send this information to the client. If the client did not request
258 a positive depth, this step is skipped.
260 ----
261 shallow-update = *shallow-line
262 *unshallow-line
263 flush-pkt
265 shallow-line = PKT-LINE("shallow" SP obj-id)
267 unshallow-line = PKT-LINE("unshallow" SP obj-id)
268 ----
270 If the client has requested a positive depth, the server will compute
271 the set of commits which are no deeper than the desired depth. The set
272 of commits start at the client's wants.
274 The server writes 'shallow' lines for each
275 commit whose parents will not be sent as a result. The server writes
276 an 'unshallow' line for each commit which the client has indicated is
277 shallow, but is no longer shallow at the currently requested depth
278 (that is, its parents will now be sent). The server MUST NOT mark
279 as unshallow anything which the client has not indicated was shallow.
281 Now the client will send a list of the obj-ids it has using 'have'
282 lines, so the server can make a packfile that only contains the objects
283 that the client needs. In multi_ack mode, the canonical implementation
284 will send up to 32 of these at a time, then will send a flush-pkt. The
285 canonical implementation will skip ahead and send the next 32 immediately,
286 so that there is always a block of 32 "in-flight on the wire" at a time.
288 ----
289 upload-haves = have-list
290 compute-end
292 have-list = *have-line
293 have-line = PKT-LINE("have" SP obj-id)
294 compute-end = flush-pkt / PKT-LINE("done")
295 ----
297 If the server reads 'have' lines, it then will respond by ACKing any
298 of the obj-ids the client said it had that the server also has. The
299 server will ACK obj-ids differently depending on which ack mode is
300 chosen by the client.
302 In multi_ack mode:
304 * the server will respond with 'ACK obj-id continue' for any common
305 commits.
307 * once the server has found an acceptable common base commit and is
308 ready to make a packfile, it will blindly ACK all 'have' obj-ids
309 back to the client.
311 * the server will then send a 'NACK' and then wait for another response
312 from the client - either a 'done' or another list of 'have' lines.
314 In multi_ack_detailed mode:
316 * the server will differentiate the ACKs where it is signaling
317 that it is ready to send data with 'ACK obj-id ready' lines, and
318 signals the identified common commits with 'ACK obj-id common' lines.
320 Without either multi_ack or multi_ack_detailed:
322 * upload-pack sends "ACK obj-id" on the first common object it finds.
323 After that it says nothing until the client gives it a "done".
325 * upload-pack sends "NAK" on a flush-pkt if no common object
326 has been found yet. If one has been found, and thus an ACK
327 was already sent, it's silent on the flush-pkt.
329 After the client has gotten enough ACK responses that it can determine
330 that the server has enough information to send an efficient packfile
331 (in the canonical implementation, this is determined when it has received
332 enough ACKs that it can color everything left in the --date-order queue
333 as common with the server, or the --date-order queue is empty), or the
334 client determines that it wants to give up (in the canonical implementation,
335 this is determined when the client sends 256 'have' lines without getting
336 any of them ACKed by the server - meaning there is nothing in common and
337 the server should just send all of its objects), then the client will send
338 a 'done' command. The 'done' command signals to the server that the client
339 is ready to receive its packfile data.
341 However, the 256 limit *only* turns on in the canonical client
342 implementation if we have received at least one "ACK %s continue"
343 during a prior round. This helps to ensure that at least one common
344 ancestor is found before we give up entirely.
346 Once the 'done' line is read from the client, the server will either
347 send a final 'ACK obj-id' or it will send a 'NAK'. 'obj-id' is the object
348 name of the last commit determined to be common. The server only sends
349 ACK after 'done' if there is at least one common base and multi_ack or
350 multi_ack_detailed is enabled. The server always sends NAK after 'done'
351 if there is no common base found.
353 Then the server will start sending its packfile data.
355 ----
356 server-response = *ack_multi ack / nak
357 ack_multi = PKT-LINE("ACK" SP obj-id ack_status)
358 ack_status = "continue" / "common" / "ready"
359 ack = PKT-LINE("ACK" SP obj-id)
360 nak = PKT-LINE("NAK")
361 ----
363 A simple clone may look like this (with no 'have' lines):
365 ----
366 C: 0054want 74730d410fcb6603ace96f1dc55ea6196122532d multi_ack \
367 side-band-64k ofs-delta\n
368 C: 0032want 7d1665144a3a975c05f1f43902ddaf084e784dbe\n
369 C: 0032want 5a3f6be755bbb7deae50065988cbfa1ffa9ab68a\n
370 C: 0032want 7e47fe2bd8d01d481f44d7af0531bd93d3b21c01\n
371 C: 0032want 74730d410fcb6603ace96f1dc55ea6196122532d\n
372 C: 0000
373 C: 0009done\n
375 S: 0008NAK\n
377 ----
379 An incremental update (fetch) response might look like this:
381 ----
382 C: 0054want 74730d410fcb6603ace96f1dc55ea6196122532d multi_ack \
383 side-band-64k ofs-delta\n
384 C: 0032want 7d1665144a3a975c05f1f43902ddaf084e784dbe\n
385 C: 0032want 5a3f6be755bbb7deae50065988cbfa1ffa9ab68a\n
386 C: 0000
387 C: 0032have 7e47fe2bd8d01d481f44d7af0531bd93d3b21c01\n
388 C: [30 more have lines]
389 C: 0032have 74730d410fcb6603ace96f1dc55ea6196122532d\n
390 C: 0000
392 S: 003aACK 7e47fe2bd8d01d481f44d7af0531bd93d3b21c01 continue\n
393 S: 003aACK 74730d410fcb6603ace96f1dc55ea6196122532d continue\n
394 S: 0008NAK\n
396 C: 0009done\n
398 S: 0031ACK 74730d410fcb6603ace96f1dc55ea6196122532d\n
400 ----
403 Packfile Data
404 -------------
406 Now that the client and server have finished negotiation about what
407 the minimal amount of data that needs to be sent to the client is, the server
408 will construct and send the required data in packfile format.
410 See pack-format.txt for what the packfile itself actually looks like.
412 If 'side-band' or 'side-band-64k' capabilities have been specified by
413 the client, the server will send the packfile data multiplexed.
415 Each packet starting with the packet-line length of the amount of data
416 that follows, followed by a single byte specifying the sideband the
417 following data is coming in on.
419 In 'side-band' mode, it will send up to 999 data bytes plus 1 control
420 code, for a total of up to 1000 bytes in a pkt-line. In 'side-band-64k'
421 mode it will send up to 65519 data bytes plus 1 control code, for a
422 total of up to 65520 bytes in a pkt-line.
424 The sideband byte will be a '1', '2' or a '3'. Sideband '1' will contain
425 packfile data, sideband '2' will be used for progress information that the
426 client will generally print to stderr and sideband '3' is used for error
427 information.
429 If no 'side-band' capability was specified, the server will stream the
430 entire packfile without multiplexing.
433 Pushing Data To a Server
434 ------------------------
436 Pushing data to a server will invoke the 'receive-pack' process on the
437 server, which will allow the client to tell it which references it should
438 update and then send all the data the server will need for those new
439 references to be complete. Once all the data is received and validated,
440 the server will then update its references to what the client specified.
442 Authentication
443 --------------
445 The protocol itself contains no authentication mechanisms. That is to be
446 handled by the transport, such as SSH, before the 'receive-pack' process is
447 invoked. If 'receive-pack' is configured over the Git transport, those
448 repositories will be writable by anyone who can access that port (9418) as
449 that transport is unauthenticated.
451 Reference Discovery
452 -------------------
454 The reference discovery phase is done nearly the same way as it is in the
455 fetching protocol. Each reference obj-id and name on the server is sent
456 in packet-line format to the client, followed by a flush-pkt. The only
457 real difference is that the capability listing is different - the only
458 possible values are 'report-status', 'delete-refs' and 'ofs-delta'.
460 Reference Update Request and Packfile Transfer
461 ----------------------------------------------
463 Once the client knows what references the server is at, it can send a
464 list of reference update requests. For each reference on the server
465 that it wants to update, it sends a line listing the obj-id currently on
466 the server, the obj-id the client would like to update it to and the name
467 of the reference.
469 This list is followed by a flush-pkt and then the packfile that should
470 contain all the objects that the server will need to complete the new
471 references.
473 ----
474 update-request = *shallow ( command-list | push-cert ) [packfile]
476 shallow = PKT-LINE("shallow" SP obj-id)
478 command-list = PKT-LINE(command NUL capability-list)
479 *PKT-LINE(command)
480 flush-pkt
482 command = create / delete / update
483 create = zero-id SP new-id SP name
484 delete = old-id SP zero-id SP name
485 update = old-id SP new-id SP name
487 old-id = obj-id
488 new-id = obj-id
490 push-cert = PKT-LINE("push-cert" NUL capability-list LF)
491 PKT-LINE("certificate version 0.1" LF)
492 PKT-LINE("pusher" SP ident LF)
493 PKT-LINE("pushee" SP url LF)
494 PKT-LINE("nonce" SP nonce LF)
496 *PKT-LINE(command LF)
497 *PKT-LINE(gpg-signature-lines LF)
498 PKT-LINE("push-cert-end" LF)
500 packfile = "PACK" 28*(OCTET)
501 ----
503 If the receiving end does not support delete-refs, the sending end MUST
504 NOT ask for delete command.
506 If the receiving end does not support push-cert, the sending end
507 MUST NOT send a push-cert command. When a push-cert command is
508 sent, command-list MUST NOT be sent; the commands recorded in the
509 push certificate is used instead.
511 The packfile MUST NOT be sent if the only command used is 'delete'.
513 A packfile MUST be sent if either create or update command is used,
514 even if the server already has all the necessary objects. In this
515 case the client MUST send an empty packfile. The only time this
516 is likely to happen is if the client is creating
517 a new branch or a tag that points to an existing obj-id.
519 The server will receive the packfile, unpack it, then validate each
520 reference that is being updated that it hasn't changed while the request
521 was being processed (the obj-id is still the same as the old-id), and
522 it will run any update hooks to make sure that the update is acceptable.
523 If all of that is fine, the server will then update the references.
525 Push Certificate
526 ----------------
528 A push certificate begins with a set of header lines. After the
529 header and an empty line, the protocol commands follow, one per
530 line. Note that the the trailing LF in push-cert PKT-LINEs is _not_
531 optional; it must be present.
533 Currently, the following header fields are defined:
535 `pusher` ident::
536 Identify the GPG key in "Human Readable Name <email@address>"
537 format.
539 `pushee` url::
540 The repository URL (anonymized, if the URL contains
541 authentication material) the user who ran `git push`
542 intended to push into.
544 `nonce` nonce::
545 The 'nonce' string the receiving repository asked the
546 pushing user to include in the certificate, to prevent
547 replay attacks.
549 The GPG signature lines are a detached signature for the contents
550 recorded in the push certificate before the signature block begins.
551 The detached signature is used to certify that the commands were
552 given by the pusher, who must be the signer.
554 Report Status
555 -------------
557 After receiving the pack data from the sender, the receiver sends a
558 report if 'report-status' capability is in effect.
559 It is a short listing of what happened in that update. It will first
560 list the status of the packfile unpacking as either 'unpack ok' or
561 'unpack [error]'. Then it will list the status for each of the references
562 that it tried to update. Each line is either 'ok [refname]' if the
563 update was successful, or 'ng [refname] [error]' if the update was not.
565 ----
566 report-status = unpack-status
567 1*(command-status)
568 flush-pkt
570 unpack-status = PKT-LINE("unpack" SP unpack-result)
571 unpack-result = "ok" / error-msg
573 command-status = command-ok / command-fail
574 command-ok = PKT-LINE("ok" SP refname)
575 command-fail = PKT-LINE("ng" SP refname SP error-msg)
577 error-msg = 1*(OCTECT) ; where not "ok"
578 ----
580 Updates can be unsuccessful for a number of reasons. The reference can have
581 changed since the reference discovery phase was originally sent, meaning
582 someone pushed in the meantime. The reference being pushed could be a
583 non-fast-forward reference and the update hooks or configuration could be
584 set to not allow that, etc. Also, some references can be updated while others
585 can be rejected.
587 An example client/server communication might look like this:
589 ----
590 S: 007c74730d410fcb6603ace96f1dc55ea6196122532d refs/heads/local\0report-status delete-refs ofs-delta\n
591 S: 003e7d1665144a3a975c05f1f43902ddaf084e784dbe refs/heads/debug\n
592 S: 003f74730d410fcb6603ace96f1dc55ea6196122532d refs/heads/master\n
593 S: 003f74730d410fcb6603ace96f1dc55ea6196122532d refs/heads/team\n
594 S: 0000
596 C: 003e7d1665144a3a975c05f1f43902ddaf084e784dbe 74730d410fcb6603ace96f1dc55ea6196122532d refs/heads/debug\n
597 C: 003e74730d410fcb6603ace96f1dc55ea6196122532d 5a3f6be755bbb7deae50065988cbfa1ffa9ab68a refs/heads/master\n
598 C: 0000
601 S: 000eunpack ok\n
602 S: 0018ok refs/heads/debug\n
603 S: 002ang refs/heads/master non-fast-forward\n
604 ----