lfs: fix the stall and corruption issue when concurrently uploading blobs
We've avoided the issue up to this point by gating worker usage with an
experimental config. See 10e62d5efa73, and the thread linked there for some of
the initial diagnosis, but essentially some data was being read from the blob
before an error occurred and `keepalive` retried, but didn't rewind the file
pointer. So the leading data was lost from the blob on the server, and the
connection stalled, trying to send more data than available.
In trying to recreate this, I was unable to do so uploading from Windows to
CentOS 7. But it reproduced every time going from CentOS 7 to another CentOS 7
over https.
I found recent fixes in the FaceBook repo to address this[1][2]. The commit
message for the first is:
The KeepAlive HTTP implementation is bugged in it's retry logic, it supports
reading from a file pointer, but doesn't support rewinding of the seek cursor
when it performs a retry. So it can happen that an upload fails for whatever
reason and will then 'hang' on the retry event.
The sequence of events that get triggered are:
- Upload file A, goes OK. Keep-Alive caches connection.
- Upload file B, fails due to (for example) failing Keep-Alive, but LFS file
pointer has been consumed for the upload and fd has been closed.
- Retry for file B starts, sets the Content-Length properly to the expected
file size, but since file pointer has been consumed no data will be uploaded,
causing the server to wait for the uploaded data until either client or
server reaches a timeout, making it seem as our mercurial process hangs.
This is just a stop-gap measure to prevent this behavior from blocking Mercurial
(LFS has retry logic). A proper solutions need to be build on top of this
stop-gap measure: for upload from file pointers, we should support fseek() on
the interface. Since we expect to consume the whole file always anyways, this
should be safe. This way we can seek back to the beginning on a retry.
I ported those two patches, and it works. But I see that `url._sendfile()` does
a rewind on `httpsendfile` objects[3], so maybe it's better to keep this all in
one place and avoid a second seek. We may still want the first FaceBook patch
as extra protection for this problem in general. The other two uses of
`httpsendfile` are in the wire protocol to upload bundles, and to upload
largefiles. Neither of these appear to use a worker, and I'm not sure why
workers seem to trigger this, or if this could have happened without a worker.
Since `httpsendfile` already has a `close()` method, that is dropped. That
class also explicitly says there's no `__len__` attribute, so that is removed
too. The override for `read()` is necessary to avoid the progressbar usage per
file.
[1] https://github.com/facebookexperimental/eden/commit/c350d6536d90c044c837abdd3675185644481469
[2] https://github.com/facebookexperimental/eden/commit/77f0d3fd0415e81b63e317e457af9c55c46103ee
[3] https://www.mercurial-scm.org/repo/hg/file/5.2.2/mercurial/url.py#l176
Differential Revision: https://phab.mercurial-scm.org/D7962
d40cc5aacc31ed673d9b5b24f98bee78c283062c 0.4f
1c590d34bf61e2ea12c71738e5a746cd74586157 0.4e
7eca4cfa8aad5fce9a04f7d8acadcd0452e2f34e 0.4d
b4d0c3786ad3e47beacf8412157326a32b6d25a4 0.4c
f40273b0ad7b3a6d3012fd37736d0611f41ecf54 0.5
0a28dfe59f8fab54a5118c5be4f40da34a53cdb7 0.5b
12e0fdbc57a0be78f0e817fd1d170a3615cd35da 0.6
4ccf3de52989b14c3d84e1097f59e39a992e00bd 0.6b
eac9c8efcd9bd8244e72fb6821f769f450457a32 0.6c
979c049974485125e1f9357f6bbe9c1b548a64c3 0.7
3a56574f329a368d645853e0f9e09472aee62349 0.8
6a03cff2b0f5d30281e6addefe96b993582f2eac 0.8.1
35fb62a3a673d5322f6274a44ba6456e5e4b3b37 0.9
2be3001847cb18a23c403439d9e7d0ace30804e9 0.9.1
36a957364b1b89c150f2d0e60a99befe0ee08bd3 0.9.2
27230c29bfec36d5540fbe1c976810aefecfd1d2 0.9.3
fb4b6d5fe100b0886f8bc3d6731ec0e5ed5c4694 0.9.4
23889160905a1b09fffe1c07378e9fc1827606eb 0.9.5
bae2e9c838e90a393bae3973a7850280413e091a 1.0
d5cbbe2c49cee22a9fbeb9ea41daa0ac4e26b846 1.0.1
d2375bbee6d47e62ba8e415c86e83a465dc4dce9 1.0.2
2a67430f92f15ea5159c26b09ec4839a0c549a26 1.1
3773e510d433969e277b1863c317b674cbee2065 1.1.1
11a4eb81fb4f4742451591489e2797dc47903277 1.1.2
11efa41037e280d08cfb07c09ad485df30fb0ea8 1.2
02981000012e3adf40c4849bd7b3d5618f9ce82d 1.2.1
196d40e7c885fa6e95f89134809b3ec7bdbca34b 1.3
3ef6c14a1e8e83a31226f5881b7fe6095bbfa6f6 1.3.1
31ec469f9b556f11819937cf68ee53f2be927ebf 1.4
439d7ea6fe3aa4ab9ec274a68846779153789de9 1.4.1
296a0b14a68621f6990c54fdba0083f6f20935bf 1.4.2
4aa619c4c2c09907034d9824ebb1dd0e878206eb 1.4.3
ff2704a8ded37fbebd8b6eb5ec733731d725da8a 1.5
2b01dab594167bc0dd33331dbaa6dca3dca1b3aa 1.5.1
39f725929f0c48c5fb3b90c071fc3066012456ca 1.5.2
fdcf80f26604f233dc4d8f0a5ef9d7470e317e8a 1.5.3
24fe2629c6fd0c74c90bd066e77387c2b02e8437 1.5.4
f786fc4b8764cd2a5526d259cf2f94d8a66924d9 1.6
bf1774d95bde614af3956d92b20e2a0c68c5fec7 1.6.1
c00f03a4982e467fb6b6bd45908767db6df4771d 1.6.2
ff5cec76b1c5b6be9c3bb923aae8c3c6d079d6b9 1.6.3
93d8bff78c96fe7e33237b257558ee97290048a4 1.6.4
333421b9e0f96c7bc788e5667c146a58a9440a55 1.7
4438875ec01bd0fc32be92b0872eb6daeed4d44f 1.7.1
6aff4f144ad356311318b0011df0bb21f2c97429 1.7.2
e3bf16703e2601de99e563cdb3a5d50b64e6d320 1.7.3
a6c855c32ea081da3c3b8ff628f1847ff271482f 1.7.4
2b2155623ee2559caf288fd333f30475966c4525 1.7.5
2616325766e3504c8ae7c84bd15ee610901fe91d 1.8
aa1f3be38ab127280761889d2dca906ca465b5f4 1.8.1
b032bec2c0a651ca0ddecb65714bfe6770f67d70 1.8.2
3cb1e95676ad089596bd81d0937cad37d6e3b7fb 1.8.3
733af5d9f6b22387913e1d11350fb8cb7c1487dd 1.8.4
de9eb6b1da4fc522b1cab16d86ca166204c24f25 1.9
4a43e23b8c55b4566b8200bf69fe2158485a2634 1.9.1
d629f1e89021103f1753addcef6b310e4435b184 1.9.2
351a9292e430e35766c552066ed3e87c557b803b 1.9.3
384082750f2c51dc917d85a7145748330fa6ef4d 2.0-rc
41453d55b481ddfcc1dacb445179649e24ca861d 2.0
195dbd1cef0c2f9f8bcf4ea303238105f716bda3 2.0.1
6344043924497cd06d781d9014c66802285072e4 2.0.2
db33555eafeaf9df1e18950e29439eaa706d399b 2.1-rc
2aa5b51f310fb3befd26bed99c02267f5c12c734 2.1
53e2cd303ecf8ca7c7eeebd785c34e5ed6b0f4a4 2.1.1
b9bd95e61b49c221c4cca24e6da7c946fc02f992 2.1.2
d9e2f09d5488c395ae9ddbb320ceacd24757e055 2.2-rc
00182b3d087909e3c3ae44761efecdde8f319ef3 2.2
5983de86462c5a9f42a3ad0f5e90ce5b1d221d25 2.2.1
85a358df5bbbe404ca25730c9c459b34263441dc 2.2.2
b013baa3898e117959984fc64c29d8c784d2f28b 2.2.3
a06e2681dd1786e2354d84a5fa9c1c88dd4fa3e0 2.3-rc
7f5094bb3f423fc799e471aac2aee81a7ce57a0b 2.3
072209ae4ddb654eb2d5fd35bff358c738414432 2.3.1
b3f0f9a39c4e1d0250048cd803ab03542d6f140a 2.3.2
d118a4f4fd16d9b558ec3f3e87bfee772861d2b7 2.4-rc
195ad823b5d58c68903a6153a25e3fb4ed25239d 2.4
0c10cf8191469e7c3c8844922e17e71a176cb7cb 2.4.1
a4765077b65e6ae29ba42bab7834717b5072d5ba 2.4.2
f5fbe15ca7449f2c9a3cf817c86d0ae68b307214 2.5-rc
a6088c05e43a8aee0472ca3a4f6f8d7dd914ebbf 2.5
7511d4df752e61fe7ae4f3682e0a0008573b0402 2.5.1
5b7175377babacce80a6c1e12366d8032a6d4340 2.5.2
50c922c1b5145dab8baefefb0437d363b6a6c21c 2.5.3
8a7bd2dccd44ed571afe7424cd7f95594f27c092 2.5.4
292cd385856d98bacb2c3086f8897bc660c2beea 2.6-rc
23f785b38af38d2fca6b8f3db56b8007a84cd73a 2.6
ddc7a6be20212d18f3e27d9d7e6f079a66d96f21 2.6.1
cceaf7af4c9e9e6fa2dbfdcfe9856c5da69c4ffd 2.6.2
009794acc6e37a650f0fae37872e733382ac1c0c 2.6.3
f0d7721d7322dcfb5af33599c2543f27335334bb 2.7-rc
f37b5a17e6a0ee17afde2cdde5393dd74715fb58 2.7
335a558f81dc73afeab4d7be63617392b130117f 2.7.1
e7fa36d2ad3a7944a52dca126458d6f482db3524 2.7.2
1596f2d8f2421314b1ddead8f7d0c91009358994 2.8-rc
d825e4025e39d1c39db943cdc89818abd0a87c27 2.8
209e04a06467e2969c0cc6501335be0406d46ef0 2.8.1
ca387377df7a3a67dbb90b6336b781cdadc3ef41 2.8.2
8862469e16f9236208581b20de5f96bd13cc039d 2.9-rc
3cec5134e9c4bceab6a00c60f52a4f80677a78f2 2.9
b96cb15ec9e04d8ac5ee08b34fcbbe4200588965 2.9.1
3f83fc5cfe715d292069ee8417c83804f6c6c1e4 2.9.2
564f55b251224f16508dd1311452db7780dafe2b 3.0-rc
2195ac506c6ababe86985b932f4948837c0891b5 3.0
269c80ee5b3cb3684fa8edc61501b3506d02eb10 3.0.1
2d8cd3d0e83c7336c0cb45a9f88638363f993848 3.0.2
6c36dc6cd61a0e1b563f1d51e55bdf4dacf12162 3.1-rc
3178e49892020336491cdc6945885c4de26ffa8b 3.1
5dc91146f35369949ea56b40172308158b59063a 3.1.1
f768c888aaa68d12dd7f509dcc7f01c9584357d0 3.1.2
7f8d16af8cae246fa5a48e723d48d58b015aed94 3.2-rc
ced632394371a36953ce4d394f86278ae51a2aae 3.2
643c58303fb0ec020907af28b9e486be299ba043 3.2.1
902554884335e5ca3661d63be9978eb4aec3f68a 3.2.2
6dad422ecc5adb63d9fa649eeb8e05a5f9bc4900 3.2.3
1265a3a71d75396f5d4cf6935ae7d9ba5407a547 3.2.4
db8e3f7948b1fdeb9ad12d448fc3525759908b9f 3.3-rc
fbdd5195528fae4f41feebc1838215c110b25d6a 3.3
5b4ed033390bf6e2879c8f5c28c84e1ee3b87231 3.3.1
07a92bbd02e5e3a625e0820389b47786b02b2cea 3.3.2
2e2e9a0750f91a6fe0ad88e4de34f8efefdcab08 3.3.3
e89f909edffad558b56f4affa8239e4832f88de0 3.4-rc
8cc6036bca532e06681c5a8fa37efaa812de67b5 3.4
ed18f4acf435a2824c6f49fba40f42b9df5da7ad 3.4.1
540cd0ddac49c1125b2e013aa2ff18ecbd4dd954 3.4.2
96a38d44ba093bd1d1ecfd34119e94056030278b 3.5-rc
21aa1c313b05b1a85f8ffa1120d51579ddf6bf24 3.5
1a45e49a6bed023deb229102a8903234d18054d3 3.5.1
9a466b9f9792e3ad7ae3fc6c43c3ff2e136b718d 3.5.2
b66e3ca0b90c3095ea28dfd39aa24247bebf5c20 3.6-rc
47dd34f2e7272be9e3b2a5a83cd0d20be44293f4 3.6
1aa5083cbebbe7575c88f3402ab377539b484897 3.6.1
2d437a0f3355834a9485bbbeb30a52a052c98f19 3.6.2
ea389970c08449440587712117f178d33bab3f1e 3.6.3
158bdc8965720ca4061f8f8d806563cfc7cdb62e 3.7-rc
2408645de650d8a29a6ce9e7dce601d8dd0d1474 3.7
b698abf971e7377d9b7ec7fc8c52df45255b0329 3.7.1
d493d64757eb45ada99fcb3693e479a51b7782da 3.7.2
ae279d4a19e9683214cbd1fe8298cf0b50571432 3.7.3
740156eedf2c450aee58b1a90b0e826f47c5da64 3.8-rc
f85de28eae32e7d3064b1a1321309071bbaaa069 3.8
a56296f55a5e1038ea5016dace2076b693c28a56 3.8.1
aaabed77791a75968a12b8c43ad263631a23ee81 3.8.2
a9764ab80e11bcf6a37255db7dd079011f767c6c 3.8.3
26a5d605b8683a292bb89aea11f37a81b06ac016 3.8.4
519bb4f9d3a47a6e83c2b414d58811ed38f503c2 3.9-rc
299546f84e68dbb9bd026f0f3a974ce4bdb93686 3.9
ccd436f7db6d5d7b9af89715179b911d031d44f1 3.9.1
149433e68974eb5c63ccb03f794d8b57339a80c4 3.9.2
438173c415874f6ac653efc1099dec9c9150e90f 4.0-rc
eab27446995210c334c3d06f1a659e3b9b5da769 4.0
b3b1ae98f6a0e14c1e1ba806a6c18e193b6dae5c 4.0.1
e69874dc1f4e142746ff3df91e678a09c6fc208c 4.0.2
a1dd2c0c479e0550040542e392e87bc91262517e 4.1-rc
e1526da1e6d84e03146151c9b6e6950fe9a83d7d 4.1
25703b624d27e3917d978af56d6ad59331e0464a 4.1.1
ed5b25874d998ababb181a939dd37a16ea644435 4.1.2
77eaf9539499a1b8be259ffe7ada787d07857f80 4.1.3
616e788321cc4ae9975b7f0c54c849f36d82182b 4.2-rc
bb96d4a497432722623ae60d9bc734a1e360179e 4.2
c850f0ed54c1d42f9aa079ad528f8127e5775217 4.2.1
26c49ed51a698ec016d2b4c6b44ca3c3f73cc788 4.2.2
857876ebaed4e315f63157bd157d6ce553c7ab73 4.3-rc
5544af8622863796a0027566f6b646e10d522c4c 4.3
943c91326b23954e6e1c6960d0239511f9530258 4.2.3
3fee7f7d2da04226914c2258cc2884dc27384fd7 4.3.1
920977f72c7b70acfdaf56ab35360584d7845827 4.3.2
2f427b57bf9019c6dc3750baa539dc22c1be50f6 4.3.3
1e2454b60e5936f5e77498cab2648db469504487 4.4-rc
0ccb43d4cf01d013ae05917ec4f305509f851b2d 4.4
cabc840ffdee8a72f3689fb77dd74d04fdc2bc04 4.4.1
a92b9f8e11ba330614cdfd6af0e03b15c1ff3797 4.4.2
27b6df1b5adbdf647cf5c6675b40575e1b197c60 4.5-rc
d334afc585e29577f271c5eda03378736a16ca6b 4.5
369aadf7a3264b03c8b09efce715bc41e6ab4a9b 4.5.1
8bba684efde7f45add05f737952093bb2aa07155 4.5.2
7de7bd407251af2bc98e5b809c8598ee95830daf 4.5.3
ed5448edcbfa747b9154099e18630e49024fd47b 4.6rc0
1ec874717d8a93b19e0d50628443e0ee5efab3a9 4.6rc1
6614cac550aea66d19c601e45efd1b7bd08d7c40 4.6
9c5ced5276d6e7d54f7c3dadf5247b7ee98ec79c 4.6.1
0b63a6743010dfdbf8a8154186e119949bdaa1cc 4.6.2
e90130af47ce8dd53a3109aed9d15876b3e7dee8 4.7rc0
33ac6a72308a215e6086fbced347ec10aa963b0a 4.7
ede3bf31fe63677fdf5bd8db687977d4e3d792ed 4.7.1
5405cb1a79010ac50c58cd84e6f50c4556bf2a4c 4.7.2
956ec6f1320df26f3133ec40f3de866ea0695fd7 4.8rc0
a91a2837150bdcb27ae76b3646e6c93cd6a15904 4.8
1c8c54cf97256f4468da2eb4dbee24f7f3888e71 4.8.1
197f092b2cd9691e2a55d198f717b231af9be6f9 4.8.2
593718ff5844cad7a27ee3eb5adad89ac8550949 4.9rc0
83377b4b4ae0e9a6b8e579f7b0a693b8cf5c3b10 4.9
4ea21df312ec7159c5b3633096b6ecf68750b0dd 4.9.1
4a8d9ed864754837a185a642170cde24392f9abf 5.0rc0
07e479ef7c9639be0029f00e6a722b96dcc05fee 5.0
c3484ddbdb9621256d597ed86b90d229c59c2af9 5.0.1
97ada9b8d51bef24c5cb4cdca4243f0db694ab6e 5.0.2
e386b5f4f8360dbb43a576dd9b1368e386fefa5b 5.1rc0
e91930d712e8507d1bc1b2dffd96c83edc4cbed3 5.1
a4e32fd539ab41489a51b2aa88bda9a73b839562 5.1.1
181e52f2b62f4768aa0d988936c929dc7c4a41a0 5.1.2
59338f9561099de77c684c00f76507f11e46ebe8 5.2rc0
ca3dca416f8d5863ca6f5a4a6a6bb835dcd5feeb 5.2
a50fecefa691c9b72a99e49aa6fe9dd13943c2bf 5.2.1
b4c82b70418022e67cc0e69b1aa3c3aa43aa1d29 5.2.2
84a0102c05c7852c8215ef6cf21d809927586b69 5.3rc0
e4344e463c0c888a2f437b78b5982ecdf3f6650a 5.3rc1
7f5410dfc8a64bb587d19637deb95d378fd1eb5c 5.3