This avoids all the complication about redundant work discussed in
the previous try at fixing this. At the expense of needing each command
that could have the problem to be patched to simply wrap the action in
onlyActionOn once the key is known. But there do not seem to be many
such commands.
onlyActionOn' should not be used with a CommandStart (or CommandPerform),
although the types do allow it. onlyActionOn handles running the whole
CommandStart chain. I couldn't immediately see a way to avoid mistken
use of onlyActionOn'.
This commit was supported by the NSF-funded DataLad project.
After a false start, I found a fairly non-intrusive way to deal with it.
Although it only handles transfers -- there may be issues with eg
concurrent dropping of the same key, or other operations.
There is no added overhead when -J is not used, other than an added
inAnnex check. When -J is used, it has to maintain and check a small
Set, which should be negligible overhead.
It could output some message saying that the transfer is being done by
another thread. Or it could even display the same progress info for both
files that are being downloaded since they have the same content. But I
opted to keep it simple, since this is rather an edge case, so it just
doesn't say anything about the transfer of the file until the other
thread finishes.
Since the deferred transfer action still runs, actions that do more than
transfer content will still get a chance to do their other work. (An
example of something that needs to do such other work is P2P.Annex,
where the download always needs to receive the content from the peer.)
And, if the first thread fails to complete a transfer, the second thread
can resume it.
But, this unfortunately means that there's a risk of redundant work
being done to transfer a key that just got transferred.
That's not ideal, but should never cause breakage; the same
thing can occur when running two separate git-annex processes.
The get/move/copy/mirror --from commands had extra inAnnex checks added,
inside the download actions. Without those checks, the first thread
downloaded the content, and then the second thread woke up and
downloaded the same content redundantly.
move/copy/mirror --to is left doing redundant uploads for now. It
would need a second checkPresent of the remote inside the upload
to avoid them, which would be expensive. A better way to avoid
redundant work needs to be found..
This commit was supported by the NSF-funded DataLad project.
git annex add, git annex lock etc make multiple seek passes,
and each seek pass checked that files existed. That was unncessary
redundant work.
Fixed by adding a new WorkTreeItem type, make seek actions use it,
and check that the files exist when constructing it.
This commit was supported by the NSF-funded DataLad project.
Before, there was a window where interrupting an add could result in the
file being moved into the annex, with no symlink yet created.
This commit was supported by the NSF-funded DataLad project.
when storing files there, since that collection is created by initremote.
(This seems to work around some brokenness of the box.com webdav server
which was entering a redirect loop.)
Note that the fix makes locationParent return Nothing instead of "."
when there's no parent directory between the path and the top of the webdav
repo.
This commit was sponsored by André Pereira on Patreon.
In my git-annex repos, I found some stale transfer info files
without lock files.
Pass a mode to tryLockExclusive, so it will create the lock file if
not present, and so not fail to clean up such transfer info files.
Normally, transfer info files are accompanied by a lock file.
But, when alwaysRunTransfer is used, the locking can fail
and it will still write the transfer info file. Perhaps there are other
cases too? Note that mkProgressUpdater's meter
writes to the transfer info file too, and it might be possible for
that meter to fire after runTransfer has cleaned up.
This commit was sponsored by andrea rota.
Fix process and file descriptor leak that was exposed when git-annex was
built with ghc 8.2.1. Apparently ghc has changed its behavior of GC
of open file handles that are pipes to running processes. That
broke git-annex test on OSX due to running out of FDs.
Audited for all uses of Annex.new and made stopCoProcesses be called
once it's done with the state. Fixed several places that might have
leaked in other situations than running the test suite.
This commit was sponsored by Ewen McNeill.
Using annexeval to run probeCrippledFileSystem' caused Git.CurrentRepo.get
to be run. Fixed easily since probeCrippledFileSystem' had no need to use
the Annex monad.
This commit was sponsored by Ethan Aubin.
While the link seems a valid article on recovering a sqlite database,
it's a stealth advertisement for a commercial product. Every post on
that blog is such a stealth advertisement.
Also, the question in this forum post has already been adequatly
answered.
So, I suspect this was a spamming attempt.
Remove closed bugs and todos that were last edited or commented before 2017.
Command line used:
for f in $(grep -l '|done\]\]' -- *.mdwn); do d="$(echo "$f" | sed 's/.mdwn$//')"; if [ -z "$(git log --since=01-01-2017 --pretty=oneline -- "$f")" -a -z "$(git log --since=01-01-2017 --pretty=oneline -- "$d")" ]; then git rm -- "./$f" ; git rm -rf "./$d"; fi; done
for f in $(grep -l '\[\[done\]\]' -- *.mdwn); do d="$(echo "$f" | sed 's/.mdwn$//')"; if [ -z "$(git log --since=01-01-2017 --pretty=oneline -- "$f")" -a -z "$(git log --since=01-01-2017 --pretty=oneline -- "$d")" ]; then git rm -- "./$f" ; git rm -rf "./$d"; fi; done
Also deletes any tagged pushes that the assistant might have done,
since those would also prevent resetting a branch back.
This commit was sponsored by andrea rota.
Motivation is to remove all metadata when it gets copied from a previous
version of the file, and that is not deisrable.
This commit was supported by the NSF-funded DataLad project.
This is similar to the pusher thread, but a separate thread because git
pushes can be done in parallel with exports, and updating a big export
should not prevent other git pushes going out in the meantime.
The exportThread only runs at most every 30 seconds, since updating an
export is more expensive than pushing. This may need to be tuned.
Added a separate channel for export commits; the committer records a
commit in that channel.
Also, reconnectRemotes records a dummy commit, to make the exporter
thread wake up and make sure all exports are up-to-date. So,
connecting a drive with a directory special remote export will
immediately update it, and getting online will automatically
update S3 and WebDAV exports.
The transfer queue is not involved in exports. Instead, failed
exports are retried much like failed pushes.
This commit was sponsored by Ewen McNeill.
The bug occurred when closeDb was not called, and garbage collection of
the DbHandle didn't give the workerThread time to shut down. Fixed by
exiting the runSqlite action when a commit is made.
(MultiWriter mode already forked off a runSqlite action, so avoided the
problem.)
This commit was sponsored by Brock Spratlen on Patreon.
Now when one repository has exported a tree, another repository can get
files from the export, after syncing.
There's a bug: While the database update works, somehow the database on
disk does not get updated, and so the database update is run the next
time, etc. Wasn't able to figure out why yet.
This commit was sponsored by Ole-Morten Duesund on Patreon.
New table needed to look up what filenames are used in the currently
exported tree, for reasons explained in export.mdwn.
Also, added smart constructors for ExportLocation and ExportDirectory to
make sure they contain filepaths with the right direction slashes.
And some code refactoring.
This commit was sponsored by Francois Marier on Patreon.
There does not seem to be a use case for supporting that, and it would
need a lot of complication to support it in a way that allows eventual
consistency when two repositories are updating the same export.
This commit was sponsored by Henrik Riomar on Patreon.
