If we use those encrypted propagation code paths, we already know from
the discovery phase (and thus the journal db) that the folders are
encrypted so no need to check again.
This will remove another expensive round trip with the server.
Signed-off-by: Kevin Ottens <kevin.ottens@nextcloud.com>
Thanks to the new discovery algorithm, we got all the freshest E2EE
information straight from the database so reuse it instead of going
through an in memory copy.
Signed-off-by: Kevin Ottens <kevin.ottens@nextcloud.com>
This was done because the propagator jobs where running in a thread a long
time ago, but this is no longer the case.
(Also QAtomicInt::load is marked as deprecated now)
- Close the UploadDevice to close the QFile after the PUT job is done.
This allows winvfs to get an oplock on the file later.
- Don't rely on QFile::fileName() to be valid after
openAndSeekFileSharedRead() was called. The way it is openend on
Windows makes it have an empty filename.
If one adds a new file to an online-only folder the previous behavior
was to upload the file in one sync and dehydrate it in the next. Now
these new files get set to Unspecified pin state, making them retain
their data.
Instead of all at once, to reduce peak memory use.
Changing UploadDevice in this way requires keeping the file open for the
duration of the upload. It also means changes to open(), seek(), close()
to ensure that uses of the device work right when a request needs to
be resent.
This was not required with 2.5 because a size of 0 was ignorted when comparing
size by the csync updater, to be compatible with very old version of the database.
But the we discovery will still think the file is changed if the database contains
a size of 0
As far as I'm aware local discovery can be skipped on folders that are
selective-sync blacklisted, so a local discovery is required when an
entry is removed from the blacklist.
Also rename
avoidReadFromDbOnNextSync() -> schedulePathForRemoteDiscovery()
since the old name might also imply it's not read from db in the local
discovery - which is not the case. Use Folder::
schedulePathForLocalDiscovery() for that.
On Linux and Windows the file watcher can't distinguish between changes
that were caused by the process itself, like during a sync operation,
and external changes. To work around that the client keeps a list of
files it has touched and blocks notifications on these files for a bit.
The duration of this block was originally and arbitrarily set at 15
seconds. During manual tests I regularly thought there was a bug when
syncs didn't trigger, when the only problem was that my changes happened
too close to a previous sync operation.
This change reduces the duration to three seconds. I imagine that this
is still enough.
Also use std::chrono while at it.
The E2EE code path would get the engine to go wrong in case of db error.
It's just better to have a failing upload or failing mkdir later in those
cases.
Emitting signals from a ctor is a bad idea anyway
Signed-off-by: Kevin Ottens <kevin.ottens@nextcloud.com>
Issue #6420
Store the X-Request-ID in the SyncFileItem and also in the blacklist.
Note that for consistency reason, the X-Request-ID is also in the
SyncFileItem if the request succeeds.
Currently there is no UI to access it, but it can be queried with sql
commands
The headers() method is used to pass extra headers to the PUT jobs for
instance, definitely needed for uploads now.
Signed-off-by: Kevin Ottens <kevin.ottens@nextcloud.com>
With the current design of the file upload this necessarily pushed to a
lock starvation on the folder. Indeed you could end up with N jobs
asking for the lock at the same time. So just avoid parallelizing for
now even though it will be slow.
We could try to optimize but that'd require some serious changes to the
sync logic on the jobs... let's stabilize first and optimize later.
Signed-off-by: Kevin Ottens <kevin.ottens@nextcloud.com>
Yes... I still wish this would be all driven by the type system, would be
much less error-prone.
Signed-off-by: Kevin Ottens <kevin.ottens@nextcloud.com>
PropagateUploadEncrypted made the assumption of the folder names never
being mangled. This is not true since the previous commits so make sure
we properly deal with that using the journal db.
Signed-off-by: Kevin Ottens <kevin.ottens@nextcloud.com>
There in no "return" in
PropagateUploadFileCommon::slotStartUpload in if (prevModtime != _item-
>_modtime) {... }
There is possibility that
PropagateItemJob::done(status, errorString)
maybe called two times from PropagateUploadFileCommon::slotStartUpload
1. in if (prevModtime != _item->_modtime) {... }
2. in if (fileIsStillChanging(*_item)) {..}
if changes in files are frequent the second call is possible.
This two calls has effect in PropagatorCompositeJob::slotSubJobFinished
and job is removed two times in _runningJobs.remove(i);
(the second time with argumetnt -1 (because first call removed job).
This return was removed in commit
efc039863b - by accident I think.
Good simulation is to synchronize firefox profile with frequent page
refresh.
Signed-off-by: Mariusz Wasak <mawasak@gmail.com>
Some servers have virus scanners and the like that can delay the
response of the final chunked upload assembly significantly, often
breaking the current 5min (!) timeout. See owncloud/enterprise#2480
for details.
Previously it tried to abort even jobs that had already finished, which
was not going to work as they wouldn't emit finished() again.
Also, in some cases the abortCount would never go to zero and that case
wasn't well documented.