* 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
This avoids import with --no-content and with --content potentially
generating two different trees, leading to a merge conflict when run in
two different clones of a repo. And it's necessary groundwork to make
git-annex sync --no-content import from special remotes that support
importKey.
Only the directory special remote currently supports importKey, and it
generates the same key as git-annex usually does, so there is no
behavior change for it.
Future special remotes will need to take care when adding importKey,
if it generates different keys. Added some warnings about that to
comments.
This commit was sponsored by Noam Kremen on Patreon.
TVars were not updated atomically, which was ok when each thread got its
own External that was the only thing using these TVars. But, with the
async extension, several External instances can share the same var, so
it needs to be a TMVar to avoid read/write conflicts.
In particular, this makes PREPARE only be sent once.
Moving jobid generation to the git-annex side lets it be simplified a
lot.
Note that it will also be possible to generate one jobid per connection,
rather than a new job per request. That will make overflow not an issue,
and will avoid some work, and will simplify some of the code.
Receive loop looks right. Still need the send loop.
And, a complication is that some messages git-annex
sends need to be wrapped in REPLY_ASYNC, while others
do not. So will probably need to split externalSend
into two.
Now that I've started implementation, I see it's really necessary that
every message the special remote sends use the protocol, otherwise
nasty edge cases abound.
Since an external process can be in the middle of some operation when an
async exception is received, it has to be shut down then. Using
cleanupProcess will close its IO handles and send it a SIGTERM.
If a special remote choses to catch SIGTERM, it's fine for it to do some
cleanup then, but until it finishes, git-annex will be blocked waiting
for it. If a special remote blocked SIGTERM, it would cause a hang.
Mentioned in docs.
Also, in passing, fixed a FD leak, it was not closing the error handle
when shutting down the external. In practice that didn't matter before because
it was only run when git-annex was itself shutting down, but now that it
can run on exception, it would have been a problem.
