Fixed successfullyFinishedLiveSizeChange to not update the rolling total
when a redundant change is in RecentChanges.
Made setRepoSizes clear RecentChanges that are no longer needed.
It might be possible to clear those earlier, this is only a convenient
point to do it.
The reason it's safe to clear RecentChanges here is that, in order for a
live update to call successfullyFinishedLiveSizeChange, a change must be
made to a location log. If a RecentChange gets cleared, and just after
that a new live update is started, making the same change, the location
log has already been changed (since the RecentChange exists), and
so when the live update succeeds, it won't call
successfullyFinishedLiveSizeChange. The reason it doesn't
clear RecentChanges when there is a reduntant live update is because
I didn't want to think through whether or not all races are avoided in
that case.
The rolling total in SizeChanges is never cleared. Instead,
calcJournalledRepoSizes gets the initial value of it, and then
getLiveRepoSizes subtracts that initial value from the current value.
Since the rolling total can only be updated by updateRepoSize,
which is called with the journal locked, locking the journal in
calcJournalledRepoSizes ensures that the database does not change while
reading the journal.
Each command that first checks preferred content (and/or required
content) and then does something that can change the sizes of
repositories needs to call prepareLiveUpdate, and plumb it through the
preferred content check and the location log update.
So far, only Command.Drop is done. Many other commands that don't need
to do this have been updated to keep working.
There may be some calls to NoLiveUpdate in places where that should be
done. All will need to be double checked.
Not currently in a compilable state.
The idea is that upon a merge of the git-annex branch, or a commit to
the git-annex branch, the reposize database will be updated. So it
should always accurately reflect the location log sizes, but it will
often be behind the actual current sizes.
Annex.reposizes will start with the value from the database, and get
updated with each transfer, so it will reflect a process's best
understanding of the current sizes.
When there are multiple processes all transferring to the same repo,
Annex.reposize will not reflect transfers made by the other processes
since the current process started. So when using balanced preferred
content, it may make suboptimal choices, including trying to transfer
content to the repo when another process has already filled it up.
But this is the same as if there are multiple processes running on
ifferent machines, so is acceptable. The reposize will eventually
get an accurate value reflecting changes made by other processes or in
other repos.
This all works fine. But it doesn't check repository sizes yet, and
without repository size checking, once a repository gets full, there
will be no other repository that will want its files.
Use of sha2 seems unncessary, probably alder2 or md5 or crc would have
been enough. Possibly just summing up the bytes of the key mod the number
of repositories would have sufficed. But sha2 is there, and probably
hardware accellerated. I doubt very much there is any security benefit
to using it though. If someone wants to construct a key that will be
balanced onto a given repository, sha2 is certianly not going to stop
them.
This fixes a problem with datalad's test suite, where loading the cluster
log happened to cause the git-annex branch commits to take a different
shape, with an additional commit.
It's also faster though, since many commands don't need the cluster log.
Just fill Annex.clusters with a thunk.
Sponsored-by: the NIH-funded NICEMAN (ReproNim TR&D3) project
This fixes a problem when git-annex testremote is run against a cluster
accessed via the http server. Annex.Cluster uses the location log
to find nodes that contain a key when checking if the key is present or getting
it. Just after a key was stored to a cluster node, reading the location log
was not getting the UUID of that node.
Apparently the Annex action that wrote to the location log, and the one
that read from it were run with two different Annex states. The http server
does use several different Annex threads.
BranchState was part of the AnnexState, and so two threads could have
different BranchStates.
Moved BranchState to the AnnexRead, so all threads will see the common state.
This might possibly impact performance. If one thread is writing changes to the
branch, and another thread is reading from the branch, the writing thread will
now invalidate the BranchState's cache, which will cause the reading thread to
need to do extra work. But correctness is surely more important. If did is
found to have impacted performance, it could probably be dealt with by doing
smarter BranchState cache invalidation.
Another way this might impact performance is that the BranchState has a small
cache. If several threads were reading from the branch and relying on the value
they just read still being in the case, now a cache miss will be more likely.
Increasing the BranchState cache to the number of jobs might be a good
idea to amelorate that. But the cache is currently an innefficient list,
so making it large would need changes to the data types.
(Commit 4304f1b6ae dealt with a follow-on
effect of the bug fixed here.)
Works down to P2P protocol.
The question now is, how to handle protocol version negotiation for
clusters? Connecting to each node to find their protocol versions and
using the lowest would be too expensive with a lot of nodes. So it seems
that the cluster needs to pick its own protocol version to use with the
client.
Then it can either negotiate that same version with the nodes when
it comes time to use them, or it can translate between multiple protocol
versions. That seems complicated. Thinking it would be ok to refuse to
use a node if it is not able to negotiate the same protocol version with
it as with the client. That will mean that sometimes need nodes to be
upgraded when upgrading the cluster's proxy. But protocol versions
rarely change.
For NotifyChanges and also for the fallthrough case where
git-annex-shell passes a command off to git-shell, proxying is currently
ignored. So every remote that is accessed via a proxy will be treated as
the same git repository.
Every other command listed in cmdsMap will need to check if
Annex.proxyremote is set, and if so handle the proxying appropriately.
Probably only P2PStdio will need to support proxying. For now,
everything else refuses to work when proxying.
The part of that I don't like is that there's the possibility a command
later gets added to the list that doesn't check proxying.
When proxying is not enabled, it's important that git-annex-shell not
leak information that it would not have exposed before. Such as the
names or uuids of remotes.
I decided that, in the case where a repository used to have proxying
enabled, but no longer supports any proxies, it's ok to give the user a
clear error message indicating that proxying is not configured, rather
than a confusing uuid mismatch message.
Similarly, if a repository has proxying enabled, but not for the
requested repository, give a clear error message.
A tricky thing here is how to handle the case where there is more than
one remote, with proxying enabled, with the specified uuid. One way to
handle that would be to plumb the proxyRemoteName all the way through
from the remote git-annex to git-annex-shell, eg as a field, and use
only a remote with the same name. That would be very intrusive though.
Instead, I decided to let the proxy pick which remote it uses to access
a given Remote. And so it picks the least expensive one.
The client after all doesn't necessarily know any details about the
proxy's configuration. This does mean though, that if the least
expensive remote is not accessible, but another remote would have
worked, an access via the proxy will fail.
Currently it only displays explanations of options like --in and --copies.
In the future, it should explain preferred content expression evaluation
and other decisions.
The explanations of a few things could be better. In particular,
"standard" will just appear as-is (or as "!standard" if it doesn't
match), rather than explaining why the standard preferred content expression
for the group matches or not.
Currently as implemented, it goes to stdout, and so commands like
git-annex find that have custom output will not display --explain
information. Perhaps that should change, dunno.
Sponsored-by: Dartmouth College's DANDI project
Speeds up eg git-annex sync --content by up to 50%. When it does not need
to transfer or drop anything, it now noops a lot more quickly.
I didn't see anything else in sync --content noop loop that could really
be sped up. It has to cat git objects to keys, stat object files, etc.
Sponsored-by: unqueued on Patreon
The flush was only done Annex.run' to make sure that the queue was flushed
before git-annex exits. But, doing it there means that as soon as one
change gets queued, it gets flushed soon after, which contributes to
excessive writes to the database, slowing git-annex down.
(This does not yet speed git-annex up, but it is a stepping stone to
doing so.)
Database queues do not autoflush when garbage collected, so have to
be flushed explicitly. I don't think it's possible to make them
autoflush (except perhaps if git-annex sqitched to using ResourceT..).
The comment in Database.Keys.closeDb used to be accurate, since the
automatic flushing did mean that all writes reached the database even
when closeDb was not called. But now, closeDb or flushDb needs to be
called before stopping using an Annex state. So, removed that comment.
In Remote.Git, change to using quiesce everywhere that it used to use
stopCoProcesses. This means that uses on onLocal in there are just as
slow as before. I considered only calling closeDb on the local git remotes
when git-annex exits. But, the reason that Remote.Git calls stopCoProcesses
in each onLocal is so as not to leave git processes running that have files
open on the remote repo, when it's on removable media. So, it seemed to make
sense to also closeDb after each one, since sqlite may also keep files
open. Although that has not seemed to cause problems with removable
media so far. It was also just easier to quiesce in each onLocal than
once at the end. This does likely leave performance on the floor, so
could be revisited.
In Annex.Content.saveState, there was no reason to close the db,
flushing it is enough.
The rest of the changes are from auditing for Annex.new, and making
sure that quiesce is called, after any action that might possibly need
it.
After that audit, I'm pretty sure that the change to Annex.run' is
safe. The only concern might be that this does let more changes get
queued for write to the db, and if git-annex is interrupted, those will be
lost. But interrupting git-annex can obviously already prevent it from
writing the most recent change to the db, so it must recover from such
lost data... right?
Sponsored-by: Dartmouth College's Datalad project
When concurrency is enabled, there can be worker threads still running
when the time limit is checked. Exiting right there does not
give those threads time to finish what they're doing. Instead, the seeking
is wrapped up, and git-annex then shuts down cleanly.
The whole point of --time-limit existing, rather than using timeout(1)
when running git-annex is to let git-annex finish the action(s) it is
working on when the time limit is reached, and shut down cleanly.
I noticed this problem when investigating why restagePointerFile might
not have run after get/drop of an unlocked file. With --time-limit -J,
a worker thread may have finished updating a work tree file, and be killed
by the time limit check before it can run restagePointerFile. So despite
--time-limit running the shutdown actions, the work tree file didn't get
restaged.
Sponsored-by: Dartmouth College's DANDI project
When accessing a git remote over http needs a git credential prompt for a
password, cache it for the lifetime of the git-annex process, rather than
repeatedly prompting.
The git-lfs special remote already caches the credential when discovering
the endpoint. And presumably commands like git pull do as well, since they
may download multiple urls from a remote.
The TMVar CredentialCache is read, so two concurrent calls to
getBasicAuthFromCredential will both prompt for a credential.
There would already be two concurrent password prompts in such a case,
and existing uses of `prompt` probably avoid it. Anyway, it's no worse
than before.
WIP: This is mostly complete, but there is a problem: createDirectoryUnder
throws an error when annex.dbdir is set to outside the git repo.
annex.dbdir is a workaround for filesystems where sqlite does not work,
due to eg, the filesystem not properly supporting locking.
It's intended to be set before initializing the repository. Changing it
in an existing repository can be done, but would be the same as making a
new repository and moving all the annexed objects into it. While the
databases get recreated from the git-annex branch in that situation, any
information that is in the databases but not stored in the branch gets
lost. It may be that no information ever gets stored in the databases
that cannot be reconstructed from the branch, but I have not verified
that.
Sponsored-by: Dartmouth College's Datalad project
At this point I've checked all AnnexState values and these were all that
remained that could move.
Pity that Annex.repo can't move, but it gets modified sometimes..
A couple of AnnexState values are set by options and could be AnnexRead,
but happen to use Annex when being set.
Sponsored-by: Max Thoursie on Patreon
Capstone of the v10 upgrade process.
Tested with a git-annex drop in a v8 repo that had a local v8 remote.
Upgrading the repo to v10 (with --force) immedaitely caused it to notice
and switch over to v10 locking. Upgrading the remote also caused it to
switch over when operating on the remote.
The InodeCache makes this fairly efficient, just an added stat call per
lock of an object file. After the v10 upgrade, there is no more
overhead.
Sponsored-by: Dartmouth College's Datalad project
This eliminates the distinction between decodeBS and decodeBS', encodeBS
and encodeBS', etc. The old implementation truncated at NUL, and the
primed versions had to do extra work to avoid that problem. The new
implementation does not truncate at NUL, and is also a lot faster.
(Benchmarked at 2x faster for decodeBS and 3x for encodeBS; more for the
primed versions.)
Note that filepath-bytestring 1.4.2.1.8 contains the same optimisation,
and upgrading to it will speed up to/fromRawFilePath.
AFAIK, nothing relied on the old behavior of truncating at NUL. Some
code used the faster versions in places where I was sure there would not
be a NUL. So this change is unlikely to break anything.
Also, moved s2w8 and w82s out of the module, as they do not involve
filesystem encoding really.
Sponsored-by: Shae Erisson on Patreon
* Deal with clock skew, both forwards and backwards, when logging
information to the git-annex branch.
* GIT_ANNEX_VECTOR_CLOCK can now be set to a fixed value (eg 1)
rather than needing to be advanced each time a new change is made.
* Misuse of GIT_ANNEX_VECTOR_CLOCK will no longer confuse git-annex.
When changing a file in the git-annex branch, the vector clock to use is now
determined by first looking at the current time (or GIT_ANNEX_VECTOR_CLOCK
when set), and comparing it to the newest vector clock already in use in
that file. If a newer time stamp was already in use, advance it forward by
a second instead.
When the clock is set to a time in the past, this avoids logging with
an old timestamp, which would risk that log line later being ignored in favor
of "newer" line that is really not newer.
When a log entry has been made with a clock that was set far ahead in the
future, this avoids newer information being logged with an older timestamp
and so being ignored in favor of that future-timestamped information.
Once all clocks get fixed, this will result in the vector clocks being
incremented, until finally enough time has passed that time gets back ahead
of the vector clock value, and then it will return to usual operation.
(This latter situation is not ideal, but it seems the best that can be done.
The issue with it is, since all writers will be incrementing the last
vector clock they saw, there's no way to tell when one writer made a write
significantly later in time than another, so the earlier write might
arbitrarily be picked when merging. This problem is why git-annex uses
timestamps in the first place, rather than pure vector clocks.)
Advancing forward by 1 second is somewhat arbitrary. setDead
advances a timestamp by just 1 picosecond, and the vector clock could
too. But then it would interfere with setDead, which wants to be
overrulled by any change. So it could use 2 picoseconds or something,
but that seems weird. It could just as well advance it forward by a
minute or whatever, but then it would be harder for real time to catch
up with the vector clock when forward clock slew had happened.
A complication is that many log files contain several different peices of
information, and it may be best to only use vector clocks for the same peice
of information. For example, a key's location log file contains
InfoPresent/InfoMissing for each UUID, and it only looks at the vector
clocks for the UUID that is being changed, and not other UUIDs.
Although exactly where the dividing line is can be hard to determine.
Consider metadata logs, where a field "tag" can have multiple values set
at different times. Should it advance forward past the last tag?
Probably. What about when a different field is set, should it look at
the clocks of other fields? Perhaps not, but currently it does, and
this does not seems like it will cause any problems.
Another one I'm not entirely sure about is the export log, which is
keyed by (fromuuid, touuid). So if multiple repos are exporting to the
same remote, different vector clocks can be used for that remote.
It looks like that's probably ok, because it does not try to determine
what order things occurred when there was an export conflict.
Sponsored-by: Jochen Bartl on Patreon
When this option is not used, there should be effectively no added
overhead, thanks to the optimisation in
b3cd0cc6ba.
When an action fails on a file, the size of the file still counts toward
the size limit. This was necessary to support concurrency, but also
generally seems like the right choice.
Most commands that operate on annexed files support the option.
export and import do not, and I don't know if it would make sense for
export to.. Why would you want an incomplete export? sync doesn't, and
while it would be easy to make it support it for transferring files,
it's not clear if dropping files should also take the size limit into
account. Commands like add that don't operate on annexed files don't
support the option either.
Exiting 101 not yet implemented.
Sponsored-by: Denis Dzyubenko on Patreon
Which could happen occasionally before when concurrency is enabled.
While not much of a problem when it did happen, better to avoid it. Also,
since it seems likely the gpg-agent sometimes fails in such a situation,
this makes it not happen when running a single git-annex command with
concurrency enabled.
This commit was sponsored by Jake Vosloo on Patreon.
Had to add to AnnexRead an indication of whether debugging is enabled.
Could have just made setupConsole not install a debug output action that
outputs, and have enableDebug be what installs that, but then in the
common case where there is no debug selector, and so all debug output is
selected, it would run the debug output action every time, which entails
an IORef access. Which would make fastDebug too slow..
Most of the changes here involve global option parsing: GlobalSetter
changed so it can both run an Annex action to set state, but can also
change the AnnexRead value, which is immutable once the Annex monad is
running.
That allowed a debugselector value to be added to AnnexRead, seeded
from the git config. The --debugfilter option's GlobalSetter then updates
the AnnexRead.
This improved GlobalSetter can later be used to move more stuff to
AnnexRead. Things that don't involve a git config will be easier to
move, and probably a *lot* of things can be moved eventually.
fastDebug, while implemented, is not used anywhere yet. But it should be
fast..
Values in AnnexRead can be read more efficiently, without MVar overhead.
Only a few things have been moved into there, and the performance
increase so far is not likely to be noticable.
This is groundwork for putting more stuff in there, particularly a value
that indicates if debugging is enabled.
The obvious next step is to change option parsing to not run in the
Annex monad to set values in AnnexState, and instead return a pure value
that gets stored in AnnexRead.
When a git remote is configured with an absolute path, use that path,
rather than making it relative. If it's configured with a relative path,
use that.
Git.Construct.fromPath changed to preserve the path as-is,
rather than making it absolute. And Annex.new changed to not
convert the path to relative. Instead, Git.CurrentRepo.get
generates a relative path.
A few things that used fromAbsPath unncessarily were changed in passing to
use fromPath instead. I'm seeing fromAbsPath as a security check,
while before it was being used in some cases when the path was
known absolute already. It may be that fromAbsPath is not really needed,
but only git-annex-shell uses it now, and I'm not 100% sure that there's
not some input that would cause a relative path to be used, opening a
security hole, without the security check. So left it as-is.
Test suite passes and strace shows the configured remote url is used
unchanged in the path into it. I can't be 100% sure there's not some code
somewhere that takes an absolute path to the repo and converts it to
relative and uses it, but it seems pretty unlikely that the code paths used
for a git remote would call such code. One place I know of is gitAnnexLink,
but I'm pretty sure that git remotes never deal with annex symlinks. If
that did get called, it generates a path relative to cwd, which would have
been wrong before this change as well, when operating on a remote.
This is conceptually very simple, just making a 1 that was hard coded be
exposed as a config option. The hard part was plumbing all that, and
dealing with complexities like reading it from git attributes at the
same time that numcopies is read.
Behavior change: When numcopies is set to 0, git-annex used to drop
content without requiring any copies. Now to get that (highly unsafe)
behavior, mincopies also needs to be set to 0. It seemed better to
remove that edge case, than complicate mincopies by ignoring it when
numcopies is 0.
This commit was sponsored by Denis Dzyubenko on Patreon.
It got broken in several ways by the streaming seeking optimisations
around version 8.20201007.
Moved time limit checking out of the matcher, which was a hack in the
first place. So everywhere that uses Limit.getMatcher needs to check
time limit. Well, almost everywhere. Command.Info uses it, but it does
not make sense to time limit getting info. And Command.MultiCast uses it
just to build up a list of files that then get passed to a command, so
it would never have hit the timeout in a useful way.
This implementation is a little more expensive when at time limit than
necessary, since it continues seeking only to discard everything after the
time limit. I did try making it close the file handles to force a faster
shutdown, but that didn't work and hung. Could certianly be improved
somehow, but seeking is probably not the expensive bit when a time limit
is hit, so this seems acceptable for now.
Especially from borg, where the content identifier logs
all end up being the same identical file!
But also, for other imports, the location tracking logs can,
in some cases, be identical files.
Bonus optimisation: Avoid looking up (and parsing when set)
GIT_ANNEX_VECTOR_CLOCK env var every time a log is written to.
Although the lookup does happen at startup even when no
log will be written now.
Done on unix, could not implement it on windows quite.
The signal library gets part of the way needed for windows.
But I had to open https://github.com/pmlodawski/signal/issues/1 because
it lacks raiseSignal.
Also, I don't know what the equivilant of getProcessGroupIDOf is on
windows. And System.Process does not provide a way to send any signal to
a process group except for SIGINT.
This commit was sponsored by Boyd Stephen Smith Jr. on Patreon.
Added annex.adjustedbranchrefresh git config to update adjusted branches
set up by git-annex adjust --unlock-present/--hide-missing.
Note, in a few cases, I was not able to make the adjusted branch
be updated in calls to moveAnnex, because information about what
file corresponds to a key is not available. They are:
* If two files point to one file, then eg, `git annex get foo` will
update the branch to unlock foo, but will not unlock bar, because it
does not know about it. Might be fixable by making `git annex get
bar` do something besides skipping bar?
* git-annex-shell recvkey likewise (so sends over ssh from old versions
of git-annex)
* git-annex setkey
* git-annex transferkey if the user does not use --file
* git-annex multicast sends keys with no associated file info
Doing a single full refresh at the end, after any incremental refresh,
will deal with those edge cases.
All properties changed to use them, except for
prop_encode_c_decode_c_roundtrip, which already filtered to ascii
for other reasons.
A few modules had to be split out, because Setup does not build-depend
on QuickCheck.
Always run Git.Config.store, so when the git config gets reloaded,
the override gets re-added to it, and changeGitRepo then calls extractGitConfig
on it and sees the annex.* settings from the override.
Remove any prior occurance of -c v and add it to the end. This way,
-c foo=1 -c foo=2 -c foo=1 will pass -c foo=1 to git, rather than -c foo=2
Note that, if git had some multiline config that got built up by
multiple -c's, this would not work still. But it never worked because
before the bug got fixed in the first place, the -c value was repeated
many times, so the multivalue thing would have been wrong. I don't think
-c can be used with multiline configs anyway, though git-config does
talk about them?
Fix a deadlock that could occur after git-annex got an unlocked file,
causing the command to hang indefinitely.
Known to happen on vfat filesystems, possibly others.
Note that a deadlock is still theoretically possible, if anything
smudge --clean does causes it to run the git queue for some other
reason.
Apparently that doesn't happen, but will need to keep an eye on it.
Limited to min of -JN or number of CPU cores, because it will often be
CPU bound, once it's read the gitignore file for a directory.
In some situations it's more disk bound, but in any case it's unlikely
to be the main bottleneck that -J is used to avoid. Eg, when dropping,
this is used for numcopies checks, but the main bottleneck will be
accessing the remotes to verify presence. So the user might decide to
-J32 that, but having 32 check-attr processes would just waste however
many filehandles they open, and probably worsen their performance due to
CPU contention.
Note that, I first tried just letting up to the -JN be started. However,
even when it's no bottleneck at all, that still results in all of them
being started. Why? Well, all the worker threads start up nearly
simulantaneously, so there's a thundering herd..
