Joey Hess [Wed, 20 Nov 2024 17:18:25 +0000 (13:18 -0400)]
add proxied uuids to http server state map
This fixes support for proxying after last commit broke it.
Note that withP2PConnections is called at server startup, and so only
proxies seen at that point will appear in the map and be used. It was
already the case that a proxy added after p2phttp was running would not
be served.
I think that is possibly a bug, but at least this commit doesn't
introduce the problem, though it might make it harder to fix it.
As bugs go, it's probably not a big deal, because after all,
git configs needs to be set in the local repository, followed by
git-annex updateproxy being run, to set up proxying. If someone is doing
that, they can restart their http server I suppose.
Joey Hess [Tue, 19 Nov 2024 19:27:26 +0000 (15:27 -0400)]
reuse http url password for p2phttp url when on same host
When remote.name.annexUrl is an annex+http(s) url, that uses the same
hostname as remote.name.url, which is itself a http(s) url, they are
assumed to share a username and password.
This avoids unnecessary duplicate password prompts.
Joey Hess [Tue, 19 Nov 2024 17:01:12 +0000 (13:01 -0400)]
skip git-remote-annex tests on windows
The NullSoftInstaller does not install git-remote-annex. For that
matter, it does not install git-annex-shell either. I don't know quite
how it would make sense to do so, without hard links.
It could contain 3 copies of the same binary.
Joey Hess [Fri, 15 Nov 2024 19:11:56 +0000 (15:11 -0400)]
avoid storing same filename repeatedly in versioned import from S3
Logically, this should make it need a lot less memory when files have
been changed many times. In my tests, it didn't seem to change memory
use at all. Unsure why, it is working. It's possible the Response is not
getting garbage collected due to pinning. But as far as I can see, all
parts of it that are retained get copied in a way that won't keep the
whole thing pinned in memory.
Joey Hess [Fri, 15 Nov 2024 18:19:02 +0000 (14:19 -0400)]
use 80% less memory when importing from a versioned S3 bucket
Same idea as commit eb714c107ba81d805458c770db8d4f22ae12a077, but even
better, because a lot of the response is DeleteMarker, that can be garbage
collected now.
Joey Hess [Thu, 14 Nov 2024 20:09:27 +0000 (16:09 -0400)]
S3 use last Key when there is no Marker element
Fix infinite loop and memory blowup when importing from an unversioned S3
bucket that is large enough to need pagination.
I don't think there actually ever will be a Marker element, a delimiter is
not set.
Probably this code path was never tested with pagination! Also the aws
library's lack of any docs made it easy to mess up.
Versioned buckets seem to not have the same problem. The API docs for
ListObjectVersions say that NextKeyMarker will always be provided when
paginating.
Joey Hess [Wed, 13 Nov 2024 18:09:24 +0000 (14:09 -0400)]
add: When adding a dotfile as a non-large file, mention that it's a dotfile
This is to reduce user confusion when their annex.largefiles matches it,
or is not set.
Note that, when annex.dotfiles is set, but a dotfile is not matched by
annex.largefiles, the "non-large file" message will be displayed. That
makes sense because whether the file is a dotfile does not matter with that
configuration.
Also, this slightly optimised the annex.dotfiles path in passing,
by avoiding the slight slowdown caused by the check added in commit 876d5b6c6fed7d540767783eb3d14469f41bd1c2 in that case.
Joey Hess [Mon, 11 Nov 2024 16:42:26 +0000 (12:42 -0400)]
git-remote-annex: Fix a reversion
Introduced in version 10.20241031 that broke cloning from a special remote
retrieveKeyFile changed to use createAnnexDirectory, which means that the
path passed to it needs to be under .git
git-remote-annex is probably the only thing in git-annex where that was not
the case. And there's no real reason it cannot be the case with it either.
Just use withOtherTmp.
Joey Hess [Wed, 30 Oct 2024 17:51:58 +0000 (13:51 -0400)]
DATA-PRESENT working for exporttree=yes remotes
Since the annex-tracking-branch is pushed first, git-annex has already
updated the export database when the DATA-PRESENT arrives. Which means
that just using checkPresent is enough to verify that there is some file
on the special remote in the export location for the key.
So, the simplest possible implementation of this happened to work!
(I also tested it with chunked specialremotes, which also works, as long
as the chunk size used is the same as the configured chunk size. In that
case, the lack of a chunk log is not a problem. Doubtful this will ever
make sense to use with a chunked special remote though, that gets pretty
deep into re-implementing git-annex.)
Updated the client side upload tip with a missing step, and reorged for clarity.
Joey Hess [Wed, 30 Oct 2024 17:09:50 +0000 (13:09 -0400)]
fix recording present on the proxy when proxying DATA-PRESENT
Apparently the protoerrhandler parameter never runs. Also the const
typo prevented the type checker from complaining that relayPUTRecord was
being called with 1 less parameter than needed.
Joey Hess [Wed, 30 Oct 2024 16:29:37 +0000 (12:29 -0400)]
Fix hang when receiving a large file into a proxied special remote
Only indicate that we're done with the bytestring once it all gets written.
Otherwise, the end of it may get garbage collected before we can process
it, leading to a hang.
This seems to have been introduced in commit cdc4bd7443a4921dd0573a0babedc51c1d62f918. Which oddly was trying to fix a
very similar problem, but specific to a cluster node. In that commit,
things got out of order, with it signaling it was done with the bytestring
before it has written all of it to the file.
My test case for this bug is a directory special remote
with a file being sent to it via a proxy accessed via ssh or http.
The file was 10 mb, and it hung on the last few kb of it not being
received.
I've also tested this fix in the case of proxying to a cluster node
directory special remote over http, which was the case cdc4bd7443a4921dd0573a0babedc51c1d62f918 was dealing with.
projects / git-annex.git / log