Sometimes the backup worker has not updated progress to the system space and a
master recovery happens. As a result, next epoch doesn't know the progress of
previous ones. This change is to check for such missing gaps and fill them with
the whole range [startVersion, endVersion).
The code is refactored into BackupProgress.actor.* to consolidate backup
progress processing for the master server.
- When a shard has a read bandwidth larger than a threshold value(configurable via knob), and it's read-bandwidth/byte-size ratio is also larger than a threshold(configurable via knob), the corresponding shard tracker will run the algorithm
- The algorithm will divide the shard into 10MB(configurable via knob) chunks and try to find the chunk(s) that has large aforementioned ratio
- Then those ranges will be logged into TraceEvents. This will later do more like actually cache them.
Not only can we now compile this giant code block in parallel with the
rest of the code, we also no longer compile this giant function twice!
(Once for the master, once for the proxy.)
1) Use the runRYWTransaction for simple DB access
2) Replace some printf with TraceEvent
3) Remove printf not used in debugging
4) Avoid wait inside the condition in loop-choose-when for
the core routine of restore worker, loader and applier.
5) Rename Restore.actor.cpp to RestoreWorker.actor.cpp since
the file only has functionalities related to restore worker.
Passed correctness test
Add a NotifiedVersion into an applier data which represents
the smallest version the applier is at.
When a loader sends mutation vector to appliers, it sends
the request that contains prevVersion and commitVersion.
This commits also put actor into an actorCollector for
loop-choose-when situation.
Add .h and .cpp files for RestoreLoader and RestoreApplier roles.
We will split the code for each restore role into a separate file.
This commit also fixes the bug in including RestoreCommon.actor.h, and
remove the unused code.
This change allows a user to write a workload in Java.
The way this is implemented is by creating a JVM within the
simulator and calling the corresponding workload class. A
workload can then run in the simulator or on a testing cluster.
If the workload is executed within the simulator, the resulting
test will not be deterministic anymore as it will execute in a
different thread (and even without that it is not clear, whether
we could get determinism as the JVM does a lot of stuff that are
not deterministic).
This is intendet to get better testing of the Java client and
layer authors can use the simulator to test their layers on a single
machine but they can still simulate failing machines etc.
Up unto here this code is only very rudiemantery tested.
This is a firest attempt of making cpack more user-friendly.
The basic idea is to generate a component for package type so
that we can have different paths depending on whether we build
an RPM, a DEB, a TGZ, or a MacOS installer. The cpack package
config file will then chose the correct components to use.
In a later point this should make it possible to build these
with `make packages` and the ugly iteration with calling cmake
between each package would be obsolete. While this solution is
a bit more bloated, it is also much more flexible and it will be
much easier to use.
Another benefit is, that this will get rid of all warnings during
a cpack run
Add a new role for ratekeeper.
Remove StorageServerChanges from data distribution.
Ratekeeper monitors storage servers, which borrows the idea from
DataDistribution.
- A set of CMake variables controls whether to keep
the simfdb directory and the traces and whether we
want to aggregate the traces into a single file
- Test labels now contain the directory they are in
so that one can now run `ctest -R fast/`
- A different binary can be used for restart tests. CMake
will automatically look for an installed fdb and use that
by default. If none is found, it will use the built one
but it will also print a warning
- CMake will throw an error if there are any text files in
the tests directory that are not associated with a test.
- Moved testing from fdbserver/CMakeLists.txt to
tests/CMakeLists.txt
- Moved fdb testing functions to its own cmake module
This isn't ideal, as it makes `restarting/from_5.2.0/potato.txt` have
the label "from_5.2.0" instead of "restarting", but it does make the
fast label work right.
negoyal