Joey Hess [Mon, 25 Jul 2022 16:35:58 +0000 (12:35 -0400)]
comment
arnaud@c1d1cc612a3921dc06a417301be08a3e125478c4 [Mon, 25 Jul 2022 07:52:39 +0000 (07:52 +0000)]
Added a comment
arnaud@c1d1cc612a3921dc06a417301be08a3e125478c4 [Mon, 25 Jul 2022 07:20:13 +0000 (07:20 +0000)]
removed
yarikoptic [Sun, 24 Jul 2022 16:52:33 +0000 (16:52 +0000)]
Added a comment
arnaud@c1d1cc612a3921dc06a417301be08a3e125478c4 [Sun, 24 Jul 2022 16:16:35 +0000 (16:16 +0000)]
Added a comment
arnaud@c1d1cc612a3921dc06a417301be08a3e125478c4 [Sat, 23 Jul 2022 15:24:06 +0000 (15:24 +0000)]
arnaud@c1d1cc612a3921dc06a417301be08a3e125478c4 [Sat, 23 Jul 2022 08:48:17 +0000 (08:48 +0000)]
Added a comment
drunken_sapo [Sat, 23 Jul 2022 02:28:18 +0000 (02:28 +0000)]
Added a comment: Confirmed your suspiction
Atemu [Fri, 22 Jul 2022 23:47:19 +0000 (23:47 +0000)]
Added a comment
Joey Hess [Fri, 22 Jul 2022 17:24:24 +0000 (13:24 -0400)]
close git bug that I reported to git mailing list
Joey Hess [Fri, 22 Jul 2022 16:24:42 +0000 (12:24 -0400)]
reproduced
Joey Hess [Fri, 22 Jul 2022 15:59:15 +0000 (11:59 -0400)]
comment
Joey Hess [Fri, 22 Jul 2022 15:38:33 +0000 (11:38 -0400)]
comment
Joey Hess [Fri, 22 Jul 2022 15:37:35 +0000 (11:37 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Fri, 22 Jul 2022 15:36:21 +0000 (11:36 -0400)]
force fully strict read of journal file again
I was thinking that discardIncompleteAppend would make it strict, since
it looks at the end of the bytestring. But, it's applied lazily..
This probably fixes windows, which was failing:
git-annex.exe: .git\annex\journal\trust.log: DeleteFile "\\\\?\\C:\\Users\\runneradmin\\.t\\5\\tmprepo22\\.git\\annex\\journal\\trust.log": permission denied (The process cannot access the file because it is being used by another process.)
jkniiv [Thu, 21 Jul 2022 20:08:55 +0000 (20:08 +0000)]
Added a comment
yarikoptic [Thu, 21 Jul 2022 14:03:56 +0000 (14:03 +0000)]
Added a comment
arnaud@c1d1cc612a3921dc06a417301be08a3e125478c4 [Thu, 21 Jul 2022 08:23:04 +0000 (08:23 +0000)]
Added a comment: Happens only when core.untrackedCache=true
Joey Hess [Wed, 20 Jul 2022 17:51:26 +0000 (13:51 -0400)]
comment
Joey Hess [Wed, 20 Jul 2022 17:44:24 +0000 (13:44 -0400)]
comment
Joey Hess [Wed, 20 Jul 2022 17:34:37 +0000 (13:34 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Wed, 20 Jul 2022 17:32:26 +0000 (13:32 -0400)]
fixed
Joey Hess [Wed, 20 Jul 2022 17:24:04 +0000 (13:24 -0400)]
Merge branch 'append'
Joey Hess [Wed, 20 Jul 2022 17:19:06 +0000 (13:19 -0400)]
prevent appends except when annex.alwayscompact=false
I would like for a new repo version to enable appends, but to do so
safely would need a v11 followed by a 1 year delay followed by a v12
that does it. Since a similar v9 and v10 transition is currently
happening, and is less than 6 months along in most repos, it does not
feel wise to stack up another year-long transition behind that. What if
I need to hurry up a new repo version for some other change?
Added todo so I remember to make this change at some time when a v11
and probably v12 repo version do make sense.
Sponsored-by: Dartmouth College's DANDI project
Joey Hess [Wed, 20 Jul 2022 16:39:03 +0000 (12:39 -0400)]
handling of interrupted appends
An append that is interrupted and writes part of a line is now dealt
with by subsequent reads and appends. This also handles a read that
happens at the same time as an append to the file.
Old versions of git-annex will still see a partially written line,
and could get confused. Since appends are currently done for url logs
and location logs, the confusion is limited to a substring of the actual
url or UUID of the remote being read. This will not affect writes, since
the journal file is locked when reading in preparation for writing.
However, the bad data can be output by git-annex and used by other
things, or could cause surprising behavior by git-annex. Including eg,
downloading the content of the wrong url.
So, something needs to be done to prevent old versions of git-annex from
running in a repository where this appending is being done..
Sponsored-by: Dartmouth College's DANDI project
Joey Hess [Wed, 20 Jul 2022 14:57:28 +0000 (10:57 -0400)]
no locking of journal on read after all
Finally have a final design, and it turns out not to need locking on read.
Joey Hess [Tue, 19 Jul 2022 22:10:45 +0000 (18:10 -0400)]
comment
yarikoptic [Tue, 19 Jul 2022 21:29:02 +0000 (21:29 +0000)]
Added a comment
Joey Hess [Tue, 19 Jul 2022 19:45:08 +0000 (15:45 -0400)]
thoughts
akspecs@fae4f3d58a0c6c9d50f01a850ce53d425e1a90ba [Tue, 19 Jul 2022 12:53:46 +0000 (12:53 +0000)]
Added a comment: cannot reproduce
arnaud@c1d1cc612a3921dc06a417301be08a3e125478c4 [Tue, 19 Jul 2022 11:22:26 +0000 (11:22 +0000)]
Joey Hess [Mon, 18 Jul 2022 21:02:41 +0000 (17:02 -0400)]
comment
Joey Hess [Mon, 18 Jul 2022 20:56:31 +0000 (16:56 -0400)]
comment
Joey Hess [Mon, 18 Jul 2022 20:46:01 +0000 (16:46 -0400)]
Merge branch 'master' into append
Joey Hess [Mon, 18 Jul 2022 20:44:32 +0000 (16:44 -0400)]
Joey Hess [Mon, 18 Jul 2022 20:40:25 +0000 (16:40 -0400)]
re-add lock journal before reading journal files
This reverts commit
2e6e9876e3fc420a22eee82eedec54ac3f678878.
This is gonna be needed after all.. The append will only be atomic if
the journal is locked, because the file being appended will have to be
moved out of the way to avoid an old version of git-annex seeing an
incomplete write to it. When git-annex finds that the file is not in the
journal, and checks the append location, locking will be needed to avoid
a race causing it to miss it in the append location too due to it being
moved back to the journal.
Joey Hess [Mon, 18 Jul 2022 19:50:36 +0000 (15:50 -0400)]
add annex.alwayscompact
Added annex.alwayscompact setting which can be unset to speed up writes to
the git-annex branch in some cases.
Sponsored-by: Dartmouth College's DANDI project
yarikoptic [Mon, 18 Jul 2022 19:23:02 +0000 (19:23 +0000)]
Added a comment
Joey Hess [Mon, 18 Jul 2022 18:45:23 +0000 (14:45 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Mon, 18 Jul 2022 18:45:03 +0000 (14:45 -0400)]
comments
Joey Hess [Mon, 18 Jul 2022 18:17:15 +0000 (14:17 -0400)]
Merge branch 'master' into append
Joey Hess [Mon, 18 Jul 2022 17:47:56 +0000 (13:47 -0400)]
efficient but unsafe journal file append
This is only for checking performance, it's not safe.
Sponsored-by: Dartmouth College's DANDI project
Joey Hess [Mon, 18 Jul 2022 17:49:17 +0000 (13:49 -0400)]
minor optimisation
Avoid re-writing the file when the journal directory did not
exist.
Joey Hess [Mon, 18 Jul 2022 17:38:12 +0000 (13:38 -0400)]
Revert "lock journal before reading journal files"
This reverts commit
47358a6f956681346a198cdf27c60bcf9498cc45.
This added overhead, and will not be needed, because appends are going
to have to be made atomic for other reasons than avoiding incomplete
reads of data being appended.
In particular, when git-annex is interrupted in the middle of an append,
it must not leave the file with a partially written line. So appending
has to somehow be made fully atomic.
Joey Hess [Mon, 18 Jul 2022 17:22:50 +0000 (13:22 -0400)]
split out appending to journal from writing, high level only
Currently this is not an improvement, but it allows for optimising
appendJournalFile later. With an optimised appendJournalFile, this will
greatly speed up access patterns like git-annex addurl of a lot of urls
to the same key, where the log file can grow rather large. Appending
rather than re-writing the journal file for each line can save a lot of
disk writes.
It still has to read the current journal or branch file, to check
if it can append to it, and so when the journal file does not exist yet,
it can write the old content from the branch to it. Probably the re-reads
are better cached by the filesystem than repeated writes. (If the
re-reads turn out to keep performance bad, they could be eliminated, at
the cost of not being able to compact the log when replacing old
information in it. That could be enabled by a switch.)
While the immediate need is to affect addurl writes, it was implemented
at the level of presence logs, so will also perhaps speed up location logs.
The only added overhead is the call to isNewInfo, which only needs to
compare ByteStrings. Helping to balance that out, it avoids compactLog
when it's able to append.
Sponsored-by: Dartmouth College's DANDI project
Ilya_Shlyakhter [Mon, 18 Jul 2022 16:43:28 +0000 (16:43 +0000)]
added suggestion to record ETags in URL- keys
g@aaed65f19d6c3a2a18c33da828e66c7bb915e65a [Mon, 18 Jul 2022 16:42:10 +0000 (16:42 +0000)]
removed duplicated content
Joey Hess [Mon, 18 Jul 2022 16:38:17 +0000 (12:38 -0400)]
Merge branch 'master' into append
g@aaed65f19d6c3a2a18c33da828e66c7bb915e65a [Mon, 18 Jul 2022 16:18:42 +0000 (16:18 +0000)]
Add new bug "creating failure getting/copying on git-lfs remote (gcrypt)"
g@aaed65f19d6c3a2a18c33da828e66c7bb915e65a [Mon, 18 Jul 2022 16:17:33 +0000 (16:17 +0000)]
Add new bug "creating failure getting/copying on git-lfs remote (gcrypt)"
Ilya_Shlyakhter [Mon, 18 Jul 2022 16:01:41 +0000 (16:01 +0000)]
Added a comment: checksums and addurl --fast
cehteh [Mon, 18 Jul 2022 13:19:44 +0000 (13:19 +0000)]
Added a comment
oliv5 [Sun, 17 Jul 2022 20:41:51 +0000 (20:41 +0000)]
Added a comment
Atemu [Sun, 17 Jul 2022 13:49:11 +0000 (13:49 +0000)]
Atemu [Sun, 17 Jul 2022 13:17:22 +0000 (13:17 +0000)]
Lukey [Sun, 17 Jul 2022 09:33:04 +0000 (09:33 +0000)]
Added a comment
tomdhunt [Sat, 16 Jul 2022 20:12:20 +0000 (20:12 +0000)]
Added a comment: Hashes for files added via addurl
Joey Hess [Sat, 16 Jul 2022 18:57:29 +0000 (14:57 -0400)]
increase cabal-version to work with recent cabal
It started complaining about custom setup needing too old a version of
cabal, a very confusing error message.
1.12 is the version of Cabal on the i386ancient builder.
Sponsored-by: Jack Hill on Patreon
yarikoptic [Fri, 15 Jul 2022 19:33:50 +0000 (19:33 +0000)]
Added a comment
Joey Hess [Fri, 15 Jul 2022 19:07:05 +0000 (15:07 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Fri, 15 Jul 2022 19:06:40 +0000 (15:06 -0400)]
design work
Joey Hess [Fri, 15 Jul 2022 18:43:29 +0000 (14:43 -0400)]
lock journal before reading journal files
This is not currently necessary; journal files are updated atomically.
However, for faster appends to large journal files, locking on read will
be needed, because appends are not atomic.
Sponsored-by: Dartmouth College's DANDI project
Joey Hess [Fri, 15 Jul 2022 17:43:46 +0000 (13:43 -0400)]
disable journalIgnorable in enableInteractiveBranchAccess
Fix a reversion that prevented --batch commands (and the assistant)
from noticing data written to the journal by other commands.
I have not identified which commit broke this for sure,
but probably it was
aeca7c2207a652c8a72f044204a5c0edb4782aac
--batch commands that wrote to the journal avoided the problem since
journalIgnorable sets unset on write. It's a little bit surprising that
nobody noticed that query --batch commands did not see data written by
other commands.
Sponsored-by: Dartmouth College's DANDI project
Joey Hess [Fri, 15 Jul 2022 16:59:59 +0000 (12:59 -0400)]
complete a comment
Added a comment
Joey Hess [Fri, 15 Jul 2022 15:14:13 +0000 (11:14 -0400)]
comment
Joey Hess [Fri, 15 Jul 2022 15:10:44 +0000 (11:10 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Fri, 15 Jul 2022 15:10:40 +0000 (11:10 -0400)]
comment
oliv5 [Fri, 15 Jul 2022 06:33:25 +0000 (06:33 +0000)]
jkniiv [Thu, 14 Jul 2022 20:32:06 +0000 (20:32 +0000)]
Added a comment
Joey Hess [Thu, 14 Jul 2022 20:09:48 +0000 (16:09 -0400)]
comment
yarikoptic [Thu, 14 Jul 2022 19:42:58 +0000 (19:42 +0000)]
Added a comment
Joey Hess [Thu, 14 Jul 2022 19:08:01 +0000 (15:08 -0400)]
comments
Joey Hess [Thu, 14 Jul 2022 19:05:51 +0000 (15:05 -0400)]
S3: Avoid writing or checking the uuid file in the S3 bucket when importtree=yes or exporttree=yes
It does not make sense for either; importing from an existing bucket should
not write to it. And the user may not have write access at all. And exporting to
a bucket should not write other files.
Also this prevents the uuid file being imported after being written.
Sponsored-by: Dartmouth College's DANDI project
Joey Hess [Thu, 14 Jul 2022 17:54:50 +0000 (13:54 -0400)]
comment
Joey Hess [Thu, 14 Jul 2022 17:53:32 +0000 (13:53 -0400)]
close
Joey Hess [Thu, 14 Jul 2022 17:51:59 +0000 (13:51 -0400)]
comment
yarikoptic [Thu, 14 Jul 2022 17:00:40 +0000 (17:00 +0000)]
Added a comment
yarikoptic [Thu, 14 Jul 2022 16:50:15 +0000 (16:50 +0000)]
Added a comment
Joey Hess [Thu, 14 Jul 2022 16:28:16 +0000 (12:28 -0400)]
optimise journal writes to not mkdir journal directory when it already exists
Sponsored-by: Dartmouth College's DANDI project
Joey Hess [Thu, 14 Jul 2022 16:13:28 +0000 (12:13 -0400)]
comment with a question
yarikoptic [Thu, 14 Jul 2022 13:40:16 +0000 (13:40 +0000)]
initial request for more efficient registerurl
yarikoptic [Wed, 13 Jul 2022 22:05:46 +0000 (22:05 +0000)]
Added a comment
Joey Hess [Wed, 13 Jul 2022 22:00:47 +0000 (18:00 -0400)]
adb: Added configuration setting oldandroid=true
To avoid using find -printf, which was first supported in Android around
2019-2020.
Probing seems too fragile, and execing stat once per file is too slow to do
when there's a faster way available, which brought me to an option...
Sponsored-by: Brett Eisenberg on Patreon
Joey Hess [Wed, 13 Jul 2022 21:27:30 +0000 (17:27 -0400)]
comment
Joey Hess [Wed, 13 Jul 2022 21:26:10 +0000 (17:26 -0400)]
comment
Joey Hess [Wed, 13 Jul 2022 21:17:44 +0000 (17:17 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Wed, 13 Jul 2022 21:13:54 +0000 (17:13 -0400)]
filter-process: Fix protocol for empty files
This caused git to complain that filter-process failed and kill it with
signal 15. Because it wrote an extra flushPkt for an empty file, which
git did not expect, and so git saw an unexpected response to the next
request.
Luckily, filter-process is only used by default in v9 and up, and v8 is
still the default. Also, git had to be updating an empty file, followed
by another file, which is a fairly unlikely situation. And git restarts
filter-process after this happens and uses it to filter the rest of the
files. So this isn't a crippling bug.
Sponsored-by: Luke Shumaker on Patreon
yarikoptic [Wed, 13 Jul 2022 20:05:34 +0000 (20:05 +0000)]
Added a comment
yarikoptic [Wed, 13 Jul 2022 19:27:06 +0000 (19:27 +0000)]
Added a comment
Joey Hess [Wed, 13 Jul 2022 19:17:08 +0000 (15:17 -0400)]
revert accidental change
yarikoptic [Wed, 13 Jul 2022 19:04:40 +0000 (19:04 +0000)]
Added a comment
Joey Hess [Wed, 13 Jul 2022 18:58:04 +0000 (14:58 -0400)]
response to misplaced bug report
Joey Hess [Wed, 13 Jul 2022 18:53:46 +0000 (14:53 -0400)]
followup
Joey Hess [Wed, 13 Jul 2022 17:44:43 +0000 (13:44 -0400)]
comment
Joey Hess [Wed, 13 Jul 2022 17:09:15 +0000 (13:09 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Wed, 13 Jul 2022 17:09:04 +0000 (13:09 -0400)]
comment