* Add LOAD CSV clause infrastructure
* Add LoadCsv operator
* Update csv::Reader class
* Support csv files with and without header
Co-authored-by: jseljan <josip.seljan@memgraph.io>
* Define additional commit log constructor which takes an oldest active id
* Delay commit log construction until the recovery process is finished
* Add test for commit log with initial id
* Silence the macro redefinition warning
* Set state to invalid after exception
* Add proper locking
* Start background replicating only if in valid state
* Freeze transaction timestamp on replica
* Timeout fixes
* Fix Jepsen run script
* Disable perf checker and enable nemesis
* Add documentation for some chunks of code
* Decrease timeout so main doesn't hang on network partitions too long
* Add config for replication client/server
* Add SSL to replication
* Add semi-sync replication
* Expose necessary information about replication
* Thread pool fix
* Set BasicResult value type to void
* Add basic communication process using commit timestamp
* Add file number to req
* Add proper recovery handling
* Allow loading of WALs with same seq num
* Allow always desired commit timestamp
* Set replica timestamp for operation
* Mark non-transactional timestamp as finished
* Add file transfer over RPC
* Snapshot transfer implementation
* Allow snapshot creation only for MAIN instances
* Replica and main can have replication clients
* Use only snapshots and WALs that are from the Main storage
* Add flush lock and expose buffer
* Add fstat for file size and TryFlushing method
* Use lseek for size
Co-authored-by: Antonio Andelic <antonio.andelic@memgraph.io>
* Add tests for multiple clients
* Use variant for RPC server and clients
* Using synchronized list for replication clients, extracted variant access to a function
* Set MAIN as default, add unregister function, add a name for replication clients
* Use the regular list for clients
* Use test fixture so storage directory is cleaned
* Use seq_cst for replication_state
Co-authored-by: Antonio Andelic <antonio.andelic@memgraph.io>
This implements the initial version of synchronous replication.
Currently, only one replica is supported and that isn't configurable.
To run the main instance use the following command:
```
./memgraph \
--main \
--data-directory main-data \
--storage-properties-on-edges \
--storage-wal-enabled \
--storage-snapshot-interval-sec 300
```
To run the replica instance use the following command:
```
./memgraph \
--replica \
--data-directory replica-data \
--storage-properties-on-edges \
--bolt-port 7688
```
You can then write/read data to Bolt port 7687 (the main instance) and also you
can read the data from the replica instance using Bolt port 7688.
NOTE: The main instance *must* be started without any data and the replica
*must* be started before any data is added to the main instance.
* Add basic synchronous replication test
* Using RWLock for replication stuff
Co-authored-by: Matej Ferencevic <matej.ferencevic@memgraph.io>
Co-authored-by: Antonio Andelic <antonio.andelic@memgraph.io>
Not having one logrotate file produced an error during rpmlint. It makes sense
to have one logrotate file after Memgraph is installed because it's easier to
manage config files. There are two logrotate files in the codebase, one for
Community and one for Enterprise edition. Having rotate files per offering also
makes sense because offerings are affected less often compared to the features.
It's easier to maintain.
* Added handshake support
* Add support for v4 hello and goodbye
* Add support for pulling n results
* Add support for transactions
* Add pull n for the dump
* Add support for NOOP
* Add support for multiple queries
* Update bolt session to support qid
* Update drivers test with multiple versions and go
* Extract failure handling into a function
* Use unique ptr instead of optional for query execution
* Destroy stream before query execution
Co-authored-by: Antonio Andelic <antonio.andelic@memgraph.io>
Summary:
This change only adds streaming support to the client request. The client
response, server request and server response are still handled only when all of
the data is received.
Reviewers: buda
Reviewed By: buda
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2807
Summary:
SLK now correctly handles different CPU architectures (BIG/little endian).
Also, more string encoding functions have been added.
Reviewers: buda
Reviewed By: buda
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2806
Summary:
This diff fixes the variable expand operator to work correctly then the start
and destination nodes use the same symbol or when the destination symbol is an
existing symbol.
Previously, the variable expand operator produced paths that were both
completely wrong (they shouldn't have been produced) and nonexistent (they
didn't even exist in the storage). Invalid data was produced because of a
wrong equality check that was introduced in D1703.
This issue was reported externally and the supplied test case was:
```
CREATE (p1:Person {id: 1})-[:KNOWS]->(:Person {id: 2})-[:KNOWS]->(:Person {id: 3})-[:KNOWS]->(:Person {id: 4})-[:KNOWS]->(p1);
MATCH path = (pers:Person {id: 3})-[:KNOWS*2]->(pers) RETURN path;
```
Also, tests have been added so the behavior remains correct.
Reviewers: buda
Reviewed By: buda
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2793
Summary:
Semantics of 'all' and 'single' were updated to be
consistent with that of 'any' and 'none'
Reviewers: mferencevic
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2789
Summary: Change any function's handling of Null elements in a list
Reviewers: mferencevic
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2787
Summary:
`DUMP DATABASE` used a separate transaction to read database data. That
wouldn't be an issue if the query was correctly disallowed in multicommand
transactions. Because it was allowed the output wasn't transactionally correct.
Instead of disabling `DUMP DATABASE` in multicommand transactions this change
fixes it so that it works properly in multicommand transactions.
Reviewers: buda
Reviewed By: buda
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2781
Summary:
Before this change properties were joined by ", " and returned as a single string,
which was ambiguous for properties that contain ", ". This diff solves this
problem by returning properties as a list type.
Reviewers: mferencevic
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2778
Summary:
This diff fixes the issue for label name (and edge type/property)
with spaces and special characters to avoid possible OpenCypher injections.
Consider an example where label name is 'hello :world'. `DUMP DATABASE`
used to return query which creates a node (u:hello :world) - i.e. node
that contains two labels 'hello' and 'world'. This fix escapes names to
create the following node with exactly one label as expected:
```
(u:`hello :world`)
```
Reviewers: mferencevic
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2774
Summary:
This diff improves the performance of `PropertyStore` with two main
techniques:
First:
`PropertyValue` has a very expensive constructor and destructor. The
`PropertyValue` was previously passed as a return value from many functions
wrapped in a `std::optional`. That caused the `PropertyValue`
constructor/destructor to be called for each intermediary value that was passed
between functions. This diff changes the functions to return a `bool` value
that imitates the `std::optional` "emptyness" flag and the `PropertyValue` is
modified using a pointer to it so that its constructor/destructor is called
only once.
Second:
The `PropertyStore` buffer was previously iterated through at least twice.
First to determine the exact position of the encoded property and then to
actually decode the property. This diff combines the two passes into a single
pass so that the property is immediately loaded if it is found.
Reviewers: buda
Reviewed By: buda
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2766
Summary:
Use Unwind + ScanAllByLabelPropertyValue logical operator to
accelerate the execution of queries like the following one:
`MATCH (n:Label) WHERE n.property IN [] ...`
Reviewers: llugovic, mferencevic
Reviewed By: llugovic, mferencevic
Subscribers: llugovic, pullbot
Differential Revision: https://phabricator.memgraph.io/D2758
Summary: All integers are signed in openCypher.
Reviewers: buda, llugovic
Reviewed By: buda
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2746
Summary:
- Add the `mgp_must_abort(const mgp_graph *graph)` C API.
- Add the `ProcCtx.must_abort()` Python API.
The usage is very simple -- the function returns a boolean indicating whether
the procedure should abort.
Reviewers: mferencevic, dsantl
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2742
Summary:
The Python `ExceptionInfo` data has the potential to extend lifetime of Python
objects that were on the Python frame while the execution was thrown. This
lifetime extension is very dangerous when executing Python procedures because
we use a custom internal memory allocator that is destroyed immediately when
the procedure is done with its execution.
This diff only keeps the formatted Python traceback (as a string) to avoid any
lifetime extension for the Python objects so that the memory allocator can
safely be destroyed after the procedure has finished its execution.
Reviewers: llugovic
Reviewed By: llugovic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2739
Summary:
`PyObject_Init` shouldn't be used in conjunction with `PyObject_New`.
The official Python documentation doesn't say this explicitly and many examples
on the web suggest that they must be used together. When they are used together
everything works when Memgraph is built against the release Python binary, but
doesn't work when built against the debug Python binary.
An implicit example that `PyObject_Init` shouldn't be used with `PyObject_New`
can be found in the Python source, in the file `Include/objimpl.h`:
```
This example code implements an object constructor with a custom
allocator, where PyObject_New is inlined, and shows the important
distinction between two steps (at least):
1) the actual allocation of the object storage;
2) the initialization of the Python specific fields
in this storage with PyObject_{Init, InitVar}.
PyObject *
YourObject_New(...)
{
PyObject *op;
op = (PyObject *) Your_Allocator(_PyObject_SIZE(YourTypeStruct));
if (op == NULL)
return PyErr_NoMemory();
PyObject_Init(op, &YourTypeStruct);
op->ob_field = value;
...
return op;
}
Note that in C++, the use of the new operator usually implies that
the 1st step is performed automatically for you, so in a C++ class
constructor you would start directly with PyObject_Init/InitVar
```
It explains that `PyObject_New` is actually equal to `malloc` +
`PyObject_Init`.
Reviewers: llugovic
Reviewed By: llugovic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2737
Summary:
This fixes an issue in Py(Vertex|Edge)GetProperty and prevents any
further issues of that type at the cost of additional typing effort.
Reviewers: ipaljak, llugovic
Reviewed By: ipaljak, llugovic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2735
Summary:
In order for this race condition to cause damage, an index/constraint must be
created/dropped at the exact moment that the GC is cleaning
indices/constraints.
Reviewers: teon.banek
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2733
Summary:
Currently, when starting Memgraph with the production package (DEB/RPM),
Memgraph always outputs an error for not being able to replace an existing
query module (`example.so` with `example.c`). This diff introduces a precheck
so that the error message is correct - so that Memgraph doesn't try to replace
an `.so` file with a `.c` file before verifying that the `.c` file is a valid
query module (which it obviously isn't). Also, I have moved the source of the
example into a subdirectory so that it isn't even considered while loading
modules.
Reviewers: teon.banek
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2724
Summary:
The storage now uses a file in the data directory (`.lock`) to determine
whether there is another instance of the storage running with the same data
directory. That helps notify the user/administrator that the system is running
in an unsupported configuration.
Reviewers: teon.banek, ipaljak
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2719
Summary:
The importer now supports all of the flags that the modern Neo4j CSV importer
supports.
Reviewers: teon.banek, ipaljak
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2709
Summary:
This diff contains simple tests for unique constraints which tries to
change property values or labels in multiple threads at the same time.
During testing, a bug has been encountered in unique constraints, i.e.
one guard lock on vertices was missing.
Reviewers: mferencevic
Reviewed By: mferencevic
Subscribers: mferencevic, pullbot
Differential Revision: https://phabricator.memgraph.io/D2711
Summary:
You should now be able to invoke query procedures written in Python. To
test the example you can run memgraph with PYTHONPATH set to `include`.
For example, assuming you are in the root of the repo, run this command.
PYTHONPATH=$PWD/include ./build/memgraph --query-modules-directory=./query_modules/
Alternatively, you can set a symlink inside the ./query_modules to point
to `include/mgp.py`, so there's no need to set PYTHONPATH. For example,
assuming you are in the root of the repo, run the following.
cd ./query_modules
ln -s ../include/mgp.py
cd ..
./build/memgraph --query-modules-directory=./query_modules/
Depends on D207
Reviewers: mferencevic, ipaljak, dsantl
Reviewed By: ipaljak
Subscribers: buda, tlastre, pullbot
Differential Revision: https://phabricator.memgraph.io/D2708
Summary:
This diff contains a necessary functionality to save and restore unique
constraint operations. The previous snapshot/WAL version is backward
compatible. Integration tests for migration from older snapshot and WAL
versions are also included.
Reviewers: mferencevic
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2680
Summary:
The new CSV parser in `mg_import_csv` behaves the same when importing a CSV
file as the standard Python CSV importer. Tests are added for all CSV field
edge-cases.
Reviewers: teon.banek, ipaljak
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2702
Summary:
When invoking a Python registered procedures we want to convert
`mgp_value` types to user facing 'mgp' types.
Depends on D2706
Reviewers: mferencevic, ipaljak
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2707
Summary:
This issue was already fixed in D2119 for the iterator, but I missed to fix the
const iterator in that diff...
Reviewers: teon.banek
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2703
Summary:
This diff restores (and fixes) the old mg_import_csv implementation. The
importer now supports the new storage engine.
Reviewers: teon.banek, ipaljak
Reviewed By: teon.banek, ipaljak
Subscribers: buda, pullbot
Differential Revision: https://phabricator.memgraph.io/D2690
Summary:
Now that unique constraint feature is added to Memgraph database,
we should update `DUMP DATABASE` with list of existing unique constraints.
Reviewers: mferencevic
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2698
Summary:
With this diff you should now be able to register `example.py` and read
procedures found there. The procedures will be listed through `CALL
mg.procedures() YIELD *` query, but invoking them will raise
`NotYetImplemented`.
If you wish to test this, you will need to run the Memgraph executable
with PYTHONPATH set to the `include` directory where `mgp.py` is found.
Additionally, you need to pass `--query-modules-directory` flag to
Memgraph, such that it points to where it will find the `example.py`.
For example, when running from the root directory of Memgraph repo, the
shell invocation below should do the trick (assuming `./build/memgraph`
is where is the executable). Make sure that `./query_modules/` does not
have `example.so` built, as that may interfere with loading
`example.py`.
PYTHONPATH=$PWD/include ./build/memgraph --query-modules-directory=./query_modules/
Reviewers: mferencevic, ipaljak, llugovic
Reviewed By: mferencevic, ipaljak, llugovic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2678
Summary:
Before this change, unique constraints supported only pairs of label
and a single property. With this change, unique constraints can be
created for label and set of properties.
Better tests for unique constraints in general are also included in
this diff.
Reviewers: mferencevic, teon.banek
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2653
Summary:
With this diff, each build of Memgraph has a version that uniquely identifies
it. The given version uniquely identifies both official release builds and
development builds. Enterprise/community builds are also differentiated in the
version. Also, support for custom suffixes is added to support custom builds
for customers.
Reviewers: teon.banek
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2662
Summary:
`query::procedure::Module` is now an interface class through which
users interact with loaded modules. At the moment, the interface is implemented
by three concrete classes: `query::procedure::BuiltinModule`,
`query::procedure::SharedLibraryModule` and `query::procedure::PythonModule`.
Reviewers: teon.banek
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2664
Summary:
This diff contains a basic implementation of unique constraints consistent with
the MVCC storage.
Stale records in the unique constraints are collected by the garbage collector.
Tests for checking correctness of unique constraints and violations are included.
Note: currently we only support a pair of label and a single property. Support for
multiple properties will be added later.
Reviewers: mferencevic, teon.banek
Reviewed By: mferencevic, teon.banek
Subscribers: buda, ipaljak, pullbot
Differential Revision: https://phabricator.memgraph.io/D2608
Summary:
This only imports the `.py` files into the global interpreter. These
modules are not exposed to query execution. A later diff will add that
support.
Reviewers: mferencevic, ipaljak
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2660
Summary:
The antlr openCypher parser generation is moved to query. Also, header files
have been added to the list of generated files so that if any header file is
deleted CMake will know that it has to regenerate it.
Reviewers: teon.banek
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2652
Summary:
The property store stores a map of `PropertyId` to `PropertyValue` mappings. It
compresses all of the values in order to use as little memory as possible.
Reviewers: teon.banek, ipaljak
Reviewed By: teon.banek
Subscribers: buda, pullbot
Differential Revision: https://phabricator.memgraph.io/D2604
Summary:
Most of the examples require properties to be enabled on edges so this change
specifies that flag explicitly. Also, errors are handled better by replacing
`bolt_client` with `mg_client`. Missing indices are created to improve import
speed.
Reviewers: teon.banek
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2594
Summary:
The configuration file that was used for the Debian/CentOS package was manually
written. This diff adds a configuration file generator that extracts all of the
necessary information about the flags directly from the built binary and uses
that information to generate the configuration file for the packages.
Reviewers: teon.banek
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2589
Summary:
The function will also be used during AST construction in order to
support `CALL ... YIELD *` syntax.
Reviewers: mferencevic, ipaljak
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2586
Summary:
The garbage collector had a race condition when it would delete deltas that
were in the middle of an object's delta chain. In the process of deleting
(unlinking) the delta, the garbage collector previously wouldn't acquire any
locks. That operation was then racing with the standard MVCC
`CreateAndLinkDelta` function that adds a new delta into the chain.
Fortunately, `CreateAndLinkDelta` always does its modifications while holding a
lock to the owner of the chain (either a vertex or an edge) so this change just
adds the lock acquiring to the garbage collector.
Reviewers: teon.banek
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2582
Summary:
The functions that previously had locks in them are always called while the
vertex lock is already being held. Also, the lock guards were implemented
incorrectly.
Reviewers: teon.banek
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2580
Summary:
Edge filters (edge type and destination vertex) are now handled natively in the
storage API. The API is implemented to be the fastest possible when using the
filters with the assumption that the number of edge types will be small.
Reviewers: teon.banek
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2576
Summary:
This diff renames `__reload__` procedure to be `mg.reload` accepting a
module name. The main CallCustomProcedure function is now split into
multiple parts, so that there's more control over finding a procedure,
type checking its arguments and finally checking the returned result
set.
Depends on D2572
Reviewers: mferencevic, ipaljak
Reviewed By: ipaljak
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2573
Summary:
The query execution now has a timeout for each Cypher query it executes. The
timeout is implemented using TSC and will work only when TSC is available (same
as PROFILE). TSC is used to mitigate the performance impact of reading the
current time constantly.
Reviewers: teon.banek
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2562
Summary:
Now when iterating over a label+property index the index verifies that the
bounds meet the criteria imposed by openCypher.
Reviewers: teon.banek
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2552
Summary:
This simplifies the C API and reduces total allocations done when
constructing a type.
Reviewers: mferencevic, ipaljak
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2550
Summary:
Previously, when accessing the labels/properties/edges of a vertex/edge that
was just created the NEW view would correctly display the change, but the OLD
view would be invalid and would crash the database. With this change the OLD
view of a freshly created vertex/edge won't cause a crash, but will instead
report an error.
Reviewers: teon.banek
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2549
Summary:
The type system is modelled after "CIP2015-09-16"
https://github.com/opencypher/openCypher/blob/master/cip/1.accepted/CIP2015-09-16-public-type-system-type-annotation.adoc
This is needed for registering procedures and their signatures. The
users will be able to specify what a custom procedure accepts and
returns. All of this needs to be available for inspection during
runtime. Therefore, this diff implements printing types as a user
presentable string. In the future, we will probably want to add type
checking through these types, because openCypher requires type checking
on values passed in and returned from custom procedures.
Reviewers: mferencevic, ipaljak, dsantl
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2544
Summary: Also test Explain and Profile through Intepreter.
Reviewers: mferencevic
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2546
Summary:
The dumper is now a function and doesn't have to worry about any state. The
function streams the Cypher queries directly to the client. This diff also
makes the dumper work with storage v2.
Reviewers: teon.banek
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2545
Summary:
The atomic memory order should be `acquire` for `load` operations, `release`
for `store` operations and `acq_rel` for any RMW (read-modify-write) operations
(like `fetch_add`).
Reviewers: teon.banek
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2540
Summary:
- Add the `AnyStream` wrapper
- Remove the `Results` struct and store a function (handler) instead
Reviewers: teon.banek, mferencevic
Reviewed By: teon.banek, mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2497
Summary:
All mgp_* symbols are exported from Memgraph executable, no other
symbols should be visible.
The primary C API header, mg_procedure.h, is now part of the
installation. Also, added a shippable query module example.
Directory `query_modules` is meant to contain sources of modules we
write and ship as part of the installation. Currently, there's only an
example module, but there may be potentially more. Some modules could
only be installed as part of the enterprise release.
For Memgraph to load custom procedures, it needs to be started with a
flag pointing to a directory with compiled shared libraries implementing
those procedures.
Reviewers: mferencevic, ipaljak, llugovic, dsantl, buda
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2538
Summary:
This change increases the throughput of the storage v2 durability 20x. With
this change, the storage v2 durability is 3x faster than the storage v1
durability in both recovery and snapshotting (before the change v2 durability
is slower than v1 durability).
Reviewers: teon.banek
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2529
Summary:
Store accumulated results as `communication::bolt::Value`s instead of
`TypedValue`s.
Add additional overloads for `Result` and `Summary` which accept `TypedValue`s
but internally perform conversions.
Reviewers: teon.banek, mferencevic
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2514
Summary:
Depends on D2471
- Add pointer to storage to `InterpreterContext`
- Rename `operator()` to `Prepare`
- Use `Interpret` instead of `operator()` (`Interpret` will be removed soon)
- Remove the `in_explicit_transaction` parameter
- Remove the memory resource parameter from `Interpret`
- Remove the storage accessor parameter from `Interpret`
- Fix up tests (remove the `Interpreter` from `database_transaction_timeout`)
Reviewers: teon.banek, mferencevic
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2482
Summary: Make `InterpreterContext` a top level instead of a nested struct
Reviewers: teon.banek
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2512
Summary: For example, the aggregate element produced for `COUNT(*)` has its `value` set to `NULL`.
Reviewers: teon.banek
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2463
Summary:
The registry is now in the `query::procedure` namespace, this makes the
naming more consistent. I.e. we are dealing with custom procedure which
are contained in modules. This naming convention is similar to Python
source code where each file represents a module and each module provides
multiple functions (or procedures in our case). At the moment we only
support exactly 1 procedure per module, but the openCypher syntax allows
for more.
Reviewers: mferencevic, ipaljak, dsantl
Differential Revision: https://phabricator.memgraph.io/D2454
Summary:
This diff implements a mechanism for registering plugins which provide
custom procedures for openCypher. Although the `Plugin` struct already
stores some function pointers, these are not set in stone w.r.t. to
requirements and signatures.
For example, in the future, we may want to allow a single plugin to
register multiple custom procedures instead of just one.
Reviewers: ipaljak, dsantl, mferencevic
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2386
Summary:
This diff adds support for an auth module. The module is used to provide
authentication and authorization (only user to role mappings). The module can
be written in any language and uses a simple protocol to communicate with
Memgraph.
Reviewers: teon.banek, buda
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2359
Summary:
Switching to Storage V2 API will require passing storage::View when
serializing VertexAccessor and EdgeAccessor, so this is just the first
step in adapting the code.
Reviewers: mferencevic, ipaljak
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2352
Summary:
Most instances of `@throw std::bad_alloc` are left unexplained as these
functions perform general heap allocations are it's obvious from the
function name that it will do so. Basically anything with `Create`, `Make` or
`Build` implies allocations. Additionally, which parts exactly perform
allocations are an implementation detail. Functions which do unexpected
heap allocations have the reason stated in the documentation, these
functions typically have exactly one spot which could raise such an
exception.
Some functions are marked as `noexcept`, these are usually "special
functions" such as constructors and operators. This could potentially
improve performance because STL may use API overloads that work faster
with `noexcept` stuff. Remaining non-throwing functions aren't marked as
`noexcept` as that wasn't our practice nor is common in our codebase. On
the other hand, if we continue enforcing the documentation of thrown
exceptions, perhaps we should start using `noexcept`.
Reviewers: mferencevic, ipaljak
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2350
Summary:
This makes Gid the same as the one in storage/v2. Before they can be
merge into one implementation, we probably want to have a similar
transition for remaining ID types.
Depends on D2346
Reviewers: mferencevic, ipaljak
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2347
Summary:
It never made sense that a global ID is its own namespace in the storage
directory tree.
Reviewers: mferencevic, ipaljak
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2346
Summary:
This effectively replaces the old PropertyValue implementation from the
one in storage/v2
Depends on D2333
Reviewers: mferencevic, ipaljak
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2335
Summary:
This is a different scheme for setting up a bookkeeping object while
still supporting arbitrary allocation alignment requests. The previous
scheme was simpler as it always allocated a power of 2 bytes, but the
trade-off was increased memory usage. This should waste less memory.
Reviewers: mtomic, mferencevic, ipaljak
Reviewed By: mtomic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2321
Summary:
There's a possible race condition where we add a deleted vertex into
index and garbage collection removes it from main storage before indices are
cleaned-up.
Reviewers: mferencevic, teon.banek
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2314
Summary:
Utils now contain a BasicResult implementation which supports different
types of error. This will make it useful for other parts of the code as
well as during the transition from old to new storage.
Reviewers: mtomic, msantl, mferencevic
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2291
Summary:
The documentation includes `std` exceptions like `std::bad_alloc` or
`std::system_error`, for which there's probably nothing we can do. This
may seem unnecessary, but it will be really helpful when writing the C
API for interfacing with custom modules and plugins, as well as when
switching to storage v2 API.
In general, we should start updating the documentation of functions
which may throw exceptions. This ought to be enforced in code review, so
that the implementation and documentation are kept in sync.
Reviewers: mferencevic, mtomic, msantl
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2288
Summary:
This ought to simplify the code which needs to work with any kind of
vertex iteration, be it through an index store or regular.
Reviewers: mtomic, mferencevic
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2258
Summary: this will make GC for indices easier
Reviewers: teon.banek, mferencevic
Reviewed By: teon.banek, mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2223
Summary:
`std::unordered_map` is 56 bytes in size, `std::map` is 48 bytes in size.
Also, `std::map` doesn't require the key type to be hashable.
Reviewers: mtomic, teon.banek
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2218
Summary:
For proper client interaction, we need to expose the (term_id, log_index)
pair for the transaction that's about to be replicated and we need to be able
to retrieve the status of a transaction defined by that pair. Transaction
status can be one of the following:
1) REPLICATED (self-explanatory)
2) WAITING (waiting for replication)
3) ABORTED (self-explanatory)
4) INVALID (received request with either invalid term_id or invalid log_index)
Reviewers: mferencevic
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2201
Summary:
The first 2 tuple elements are redundant as they are available through
EdgeAccessor and they needlessly complicate the usage of the API.
Reviewers: mtomic, mferencevic
Reviewed By: mtomic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2200
Summary:
The newly added methods are modeled after the ones being added in C++20 to
`std::pmr::polymorphic_allocator`.
Reviewers: mtomic, mferencevic
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2197
Summary:
`lock_guard` is holding vertex lock while we're deleting the vertex,
and then it might try to unlock it in its destructor and access freed memory.
Reviewers: teon.banek, mferencevic
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2192
Summary:
This is a preparation step in case we want to have a custom allocator in
SkipList. For example, pool based allocator for SkipListNode.
Introduction of MemoryResource and removal of `calloc` has reduced the
performance a bit according to micro benchmarks. This performance hit is
not visible on benchmarks which do more concurrent operations.
Reviewers: mferencevic, mtomic
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2140
Summary:
This change implements full edges support in storage v2. Edges can be created
and deleted. Support for detach-deleting vertices is added and regular vertex
deletion verifies existance of edges.
Reviewers: mtomic, teon.banek
Reviewed By: mtomic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2180
Summary:
Instead of calling the wanted method, this diff takes copies of both `is_replicated(x)` and `is_active(x)` results so that they can't change during the method execution.
The problem with the previous implementation was that the information about a
log could change during two consecutive queries as ilustrated in the example
below.
```
Thread 1 Thread2
rlog->set_active(1);
if (rlog->is_replicated(1)) return true;
rlog->set_replicated(1);
if (rlog->is_active(1)) return false;
throw InvalidLogReplicationLookup();
```
The snippet above would throw as we don't have any information about the log
`1`, because the `set_replicated` call would "shadow" the active bit.
Reviewers: ipaljak
Reviewed By: ipaljak
Subscribers: pullbot, mferencevic
Differential Revision: https://phabricator.memgraph.io/D2165
Summary:
We forgot to update `modified_vertices` in the case when the vertex
has an empty version chain. It didn't manifest before because it was impossible
for a vertex to have an empty version chain without garbage collection.
Reviewers: mferencevic, teon.banek
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2154
Summary:
We forgot to add the newly created vertex into `modified_vertices`
list.
Reviewers: teon.banek, mferencevic
Reviewed By: mferencevic
Differential Revision: https://phabricator.memgraph.io/D2153
Summary:
Initial implementation of new storage engine. It implements snapshot isolation
for transactions. All changes in the database are stored as deltas instead of
making full copies. Currently, the storage supports full transaction
functionality (commit, abort, command advancement). Also, support has been
implemented only for vertices that have only labels.
Reviewers: teon.banek, mtomic
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2138
Summary:
Micro benchmarks show some minor variations compared to the previous
commit. Smaller cases are a bit worse while larger data cases are a bit
better.
Reviewers: mtomic, mferencevic
Reviewed By: mtomic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2136
Summary:
Micro benchmarks show improvements in performance of MapLiteral from 5%
to 40% depending on the size of the input. On the other hand, a sequence
of AdditionOperators behaves the same with both allocation schemes.
Reviewers: mtomic, mferencevic, msantl
Reviewed By: mtomic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2132
Summary:
The global variable may hide the fact that it uses the default
utils::NewDeleteResource() for allocations.
Reviewers: mtomic, llugovic, mferencevic
Reviewed By: mtomic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2121
Summary:
Split up `process-file` so that looking at the generated code for an LCP form is
easier from the REPL.
`process-lcp`, `generate-hpp` and `generate-cpp` now perform the generation of
C++ code, but take a list of "C++ elements" (the results of LCP forms) as input
and write their output to streams. They do no reading/evaluating of LCP forms of
their own.
`read-lcp` and `read-lcp-file` are used to read and evaluate a stream of LCP
forms. The latter is a specialized version for file streams which also reports
the position of the form within the file when an error happens.
`process-lcp-string` and `process-lcp-file` are convenient wrappers around the
main functionality that take a string (file) and output to strings (files).
Using `read-lcp` and `read-lcp-file` they process LCP forms and pass them off to
`process-lcp` for code generation.
Reviewers: mtomic, teon.banek
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2097
Summary:
`dump()` awesome Memgraph function is removed.
`mg_dump` executable now uses "dump database" command instead of "return dump()"
and dumps database to standard output instead of a file.
Reviewers: teon.banek
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2122
Summary:
Stream queries to the output table.
For effective output streaming, a simple operator, `OutputTableStream` is implemented which fetches and
produces a single row on each Pull.
Reviewers: teon.banek
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2099
Summary:
This is unfortunately needed in the C++17 standard, so that the
allocator is correctly propagated to elements of pair which respect the
"Uses Allocator" protocol. C++20 standard resolves this issue, but we
still have a long way before it is released and implemented by the
compiler and standard library vendors.
Reviewers: mtomic, llugovic, mferencevic
Reviewed By: mtomic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2107
Summary: Forward declare the needed struct and include in the correct file.
Reviewers: ipaljak
Reviewed By: ipaljak
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2110
Summary:
This diff only introduces a MemoryResource member to TypedValue and correctly
propagates through various constructors and assignments. At the moment,
MemoryResource is not used to actually allocate anything.
Reviewers: mtomic, llugovic, mferencevic, msantl
Reviewed By: mtomic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2085
Summary:
HA should now support constraints in the same way the SM version does.
I only tested this thing manually, but I plan to add a new integration test for
this also.
Reviewers: ipaljak, vkasljevic, mferencevic
Reviewed By: ipaljak, mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2083
Summary:
Implemented a simple dump executable which runs `RETURN dump()` on the database using Bolt protocol.
The `dump()` function returns a list of openCypher queries needed for database reconstruction (note: in the future we should use more appropriate method that supports streaming of data).
Reviewers: teon.banek, mferencevic
Reviewed By: teon.banek
Subscribers: pullbot, mferencevic
Differential Revision: https://phabricator.memgraph.io/D2067
Summary:
Fix condition variable notifications
Fix vote requested invalid memory access (size off by one)
Fix blocking wait in RaftPeer while candidate
Don't copy large log entries when only the term is needed
Reviewers: ipaljak, msantl
Reviewed By: ipaljak
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2088
Summary:
Preparing unique constraint code to be implemented in HA. To do so, I'm
moving everything to `stograge/common/` folder. I also added a new header,
`gid.hpp` which does a `ifdef` include of the correct `gid.hpp` based on the
product.
Reviewers: ipaljak, mferencevic, vkasljevic, teon.banek
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2079
Summary:
Instead of asking rocksdb for snapshot metadata, keep it in memory for
faster access.
Reviewers: ipaljak
Reviewed By: ipaljak
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2075
Summary:
During it's leadership, one peer can receive RPC messages from other peers that his reign is over.
The problem is when this happens during a transaction commit.
This is handled in the following way.
If we're the current leader and we want to commit a transaction, we need to make sure the Raft Log is replicated before we can tell the client that the transaction is committed.
During that wait, we can only notice that the replication takes too long, and we report that with `LOG(WARNING)` messages.
If we change the Raft mode during the wait, our Raft implementation will internally commit this transaction, but won't be able to acquire the Raft lock because the `db.Reset` has been called.
This is why there is an manual lock acquire. If we pick up that the `db.Reset` has been called, we throw an `UnexpectedLeaderChangeException` exception to the client.
Another thing with long running transactions, if someone decides to kill a `memgraph_ha` instance during the commit, the transaction will have `abort` hint set. This will cause the `src/query/operator.cpp` to throw a `HintedAbortError`. We need to catch this during the shutdown, because the `memgraph_ha` isn't dead from the user perspective, and the transaction wasn't aborted because it took too long, but we can differentiate between those two.
Reviewers: mferencevic, ipaljak
Reviewed By: mferencevic, ipaljak
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D1956
Summary:
Micro benchmarks show no change compared to global new & delete. This is
to be expected, because Unwind relies only on `std::vector` which ought
to reserve the memory in reasonable chunks.
Reviewers: mtomic, llugovic
Reviewed By: mtomic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2064
Summary:
Micro benchmarks show an improvement to performance of about 10%
compared to global new & delete.
Reviewers: mtomic, llugovic
Reviewed By: mtomic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2061
Summary: We should now exchange around O(log(n)) messages to get a follower that is n transactions behind back in sync.
Reviewers: msantl, mferencevic
Reviewed By: msantl
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2051
Summary:
I was a bit stupid and thought I can use `std::swap`, while `std::swap`
is implemented in terms of move itself.
Reviewers: mtomic, mferencevic
Reviewed By: mtomic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2054
Summary:
A `valgrind` run of a modified client that connects multiple times to the
database now shows no memory leaks, previously the context created with
`SSL_CTX_new` was leaked every time.
Reviewers: teon.banek, mtomic
Reviewed By: teon.banek, mtomic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2052
Summary:
Micro benchmarks show that MonotonicBufferResource improves performance
by a factor of 1.5.
Reviewers: mtomic, mferencevic, llugovic
Reviewed By: mtomic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2048
Summary:
Benchmarks show minor improvements. Perhaps it makes sense at some later date
to use another allocator for things lasting only in a single `Pull`.
Reviewers: mferencevic, mtomic, llugovic
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2018
Summary:
Unfortunately, the written micro benchmark only reports minor
improvements compared to default allocator. The results are in some
cases even a tiny bit worse.
Reviewers: mtomic, mferencevic, llugovic
Reviewed By: mtomic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2039
Summary:
This change introduces dumping of indices keys. During the dump process,
an internal label is assigned to each vertex and index on vertex's
internal property id is created for faster matching during edge creation.
Reviewers: teon.banek
Reviewed By: teon.banek
Subscribers: msantl, pullbot
Differential Revision: https://phabricator.memgraph.io/D2046
Summary:
Prior to this change, a huge query was returned by DumpGenerator that
dumped the entire graph. This change split the single query to multiple
queries, each dumping a single vertex/edge. For easier vertex matching
when dumping edge, an internal property id is assigned to each vertex and
removed after the whole graph is dumped.
Reviewers: teon.banek
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2038
Summary:
# Summary
## Concepts and terminology
A **namestring for <object>** is a Lisp string that names the C++ language element
`<object>` (such as a namespace, a variable, a class, etc.). Therefore we have a
"namestring for a namespace", a "namestring for a variable", etc.
A **designator for a namestring for <object>** is a Lisp object that denotes a
"namestring for <object>". These are symbols and strings themselves.
A **typestring** is a Lisp string that represents a C++ type, i.e. its
declaration.
A **typestring designator** is a Lisp object that denotes a typestring.
A typestring (and the corresponding C++ type) is said to be **supported** if it
can be parsed using `parse-cpp-type-declaration`. This concept is important and
should form the base of our design because we can't really hope to ever support
all of C++'s type declarations.
A typestring (and the corresponding C++ type) that is not supported is
**unsupported**.
A **processed typestring** is a typestring that is either fully qualified or
unqualified.
A C++ type is said to be **known** if LCP knows extra information about the type,
rather than just what kind of type it is, in which namespace it lives, etc. For
now, the only known types are those that are defined within LCP itself using
`define-class` & co.
A C++ type is **unknown** if it is not known.
**Typestring resolution** is the process of resolving a (processed) typestring
into an instance of `cpp-type` or `unsupported-cpp-type`.
**Resolving accessors** are accessors which perform typestring resolution.
## Changes
Explicitly introduce the concept of supported and known types. `cpp-type` models
supported types while `unsupported-cpp-type` models unsupported types.
Subclasses of `cpp-type` model known types. `general-cpp-type` is either a
`cpp-type` or an `unsupported-cpp-type`.
Add various type queries.
Fix `define-rpc`'s `:initarg` (remove it from the `cpp-member` struct).
Introduce namestrings and their designators in `names.lisp`.
Introduce typestrings and their designators.
Introduce **processed typestrings**. Our DSL's macros (`define-class` & co.)
convert all of the given typestrings into processed typestrings because we don't
attempt to support partially qualified names and relative name lookup yet. A
warning is signalled when a partially qualified name is treated as a fully
qualified name.
The slots of `cpp-type`, `cpp-class`, `cpp-member` and `cpp-capnp-opts` now
store processed typestrings which are lazily resolved into their corresponding
C++ types.
The only thing that instances of `unsupported-cpp-type` are good for is getting
the typestring that was used to construct them. Most of LCP's functions only
work with known C++ types, i.e. `cpp-type` instances. The only function so far
that works for both of them is `cpp-type-decl`.
Since "unsupportedness" is now explicitly part of LCP, client code is expected
to manually check whether a returned type is unsupported or not (or face
receiving an error otherwise), unless a function is documented to return only
`cpp-type` instances.
A similar thing goes for "knowness". Client code is expected to manually check
whether a returned type is known or not, unless a function is documented to
return only (instances of `cpp-type` subclasses) known types.
## TODO
Resolution still has to be done for the following slots of the
following structures:
- `slk-opts`
- `save-args`
- `load-args`
- `clone-opts`
- `args`
- `return-type`
Reviewers: teon.banek, mtomic
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D1962
Summary:
This should hopefully resolve multithreading issues when multiple LCP
invocations tried to process the same file.
Reviewers: mferencevic, msantl
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2029
Summary:
Read throughput dropped by about 50% from the last diff.
This should fix it (at least according to local measurements).
Reviewers: msantl
Reviewed By: msantl
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2027
Summary:
According to the written benchmark, using MonotonicBufferResource yields
significant improvements to performance of Distinct. The setup fills the
database with vertices depending on the benchmark state. No edges are
created. Then we run DISTINCT on that. Since each vertex is unique, we
will store everything in the `DistinctCursor::seen_rows_`, which is
backed by a MemoryResource. This setup, on my machine, yields 10 times
better performance when run with MonotonicBufferResource.
Reviewers: mferencevic, mtomic, msantl
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D1894
Summary:
This allows us to decouple issuing heartbeats from the server mode
which is useful when we know we will transition to LEADER but cannot yet
change the mode due to some Raft internals.
It can now happen that a couple of HBs are sent when in FOLLOWER or CANDIDATE
mode, but this doesn't affect the correctness of the protocol.
Reviewers: msantl
Reviewed By: msantl
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2025
Summary:
There will be a lot of leftover files, execute the following commands inside
`src/` to remove them:
```
git clean -xf
rm -r rpc/ storage/single_node_ha/rpc/
```
Reviewers: teon.banek
Reviewed By: teon.banek
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2011
Summary:
Clang-analyzer-core found NullDereference issue in LruList. I don't see
how that can occur, but as a precaution I'll add a CHECK.
Reviewers: msantl, ipaljak
Reviewed By: msantl
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D2015
Summary:
Queries such as `UNWIND RANGE(1, 90000) AS x CREATE(:node{id:x});` work
now. At some point (cca 100000 state deltas) the messages become too large for
Cap'n Proto to handle and that exception kills the cluster. This is fine since
we are about to replace it anyway.
Also, the issue with leadership change during replication still remains if the
user tempers with the constants (election timeout more or less the same to hb
interval). This is unavoidable, so we need to address that issue in the future.
Also, I've removed the unnecessary `backoff_until_` variable and the logic around it.
It's a small change so I kept it within this diff (sorry).
Reviewers: msantl, mferencevic
Reviewed By: msantl
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D1998
Summary:
This API change is needed in order to propagate the memory allocation
scheme for the execution of LogicalOperator::Cursor
Depends on D1990
Reviewers: mtomic, mferencevic
Reviewed By: mtomic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D1980
Summary:
The new distributed directory is inside the query, and mirrors the query
structure. This groups all of the distributed (query) source code
together, which should make the potential directory extraction easier.
Reviewers: mferencevic, llugovic, mtomic
Reviewed By: mferencevic
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D1923
Summary:
Manually call `CompactRange` on KVStore (rocksDB) to free disk space
upon Raft log compaction.
Raft log takes surprisingly large amount of disk space (1MB for 1000 entries) so
we need to free disk space as we compact the log into snapshot.
When `log_size_snapshot_threshold` set to `5000` the disk usage of the
`durability_dir` goes from `4.8 MB` to `240 KB`.
Reviewers: ipaljak
Reviewed By: ipaljak
Subscribers: pullbot
Differential Revision: https://phabricator.memgraph.io/D1939
