before this change, we provide `boost_test_print_type()` for all types
which can be formatted using {fmt}. these types includes those who
fulfill the concept of range, and their element can be formatted using
{fmt}. if the compilation unit happens to include `fmt/ranges.h`.
the ranges are formatted with `boost_test_print_type()` as well. this
is what we expect. in other words, we use {fmt} to format types which
do not natively support {fmt}, but they fulfill the range concept.
but `boost::unit_test::basic_cstring` is one of them
- it can be formatted using operator<<, but it does not provide
fmt::format specialization
- it fulfills the concept of range
- and its element type is `char const`, which can be formatted using
{fmt}
that's why it's formatted like:
```
test/boost/sstable_directory_test.cc(317): fatal error: in "sstable_directory_test_generation_sanity": critical check ['s', 's', 't', '-', '>', 'g', 'e', 'n', 'e', 'r', 'a', 't', 'i', 'o', 'n', '(', ')', ' ', '=', '=', ' ', 's', 's', 't', '1', '-', '>', 'g', 'e', 'n', 'e', 'r', 'a', 't', 'i', 'o', 'n', '(', ')'] has failed`
```
where the string is formatted as a sequence-alike container. this
is far from readable.
so, in this change, we do not define `boost_test_print_type()` for
the types which natively support `operator<<` anymore. so they can
be printed with `operator<<` when boost::test prints them.
Fixesscylladb/scylladb#19637
Signed-off-by: Kefu Chai <kefu.chai@scylladb.com>
Closesscylladb/scylladb#19638
The equivalent of small-objects, but for large objects (spans).
Allows listing object of a large-class, and therefore investigating a
run-away class, by attempting to identify the owners of the objects in
it.
Written to investigate #16493Closesscylladb/scylladb#16711
When debugging the issue of high LWT contention metric, we (the
drivers team) discovered that at least 3 drivers (Go, Java, Rust)
cause high numbers in that metrics in LWT workloads - we doubted that
all those drivers route LWT queries badly. We tried to understand that
metric and its semantics. It took 3 people over 10 hours to figure out
what it is supposed to count.
People from core team suspected that it was the drivers sending
requests to different shards, causing contention. Then we ran the
workload against a single node single shard cluster... and observed
contention. Finally, we looked into the Scylla code and saw it.
**Uninitialized stack value.**
The core member was shocked. But we, the drivers people, felt we always
knew it. It's yet another time that we are blamed for a server-side
issue. We rebuilt scylla with the variable initialized to 0 and the
metric kept being 0.
To prevent such errors in the future, let's consider some lints that
warn against uninitialized variables. This is such an obvious feature
of e.g. Rust, and yet this has shown to be cause a painful bug in 2024.
Closesscylladb/scylladb#19625
On Ubuntu/Debian, we have to install systemd-coredump before
running has_ztd(), since it detect ZSTD support by running coredumpctl.
Move pkg_install('systemd-coredump') to the head of the script.
Fixes#19643Closesscylladb/scylladb#19648
in cccec07581, we started using a featured introduced by {fmt} v10.
but we are still using the {fmt} cooked using seastar, and it is
9.1.0, so this breaks the build when running the clang-tidy workflow.
in this change, instead of building on ubuntu jammy, we use the
scylladb/scylla-toolchain image based on fedora 40, which provides
{fmt} v10.2.1. since we are have clang 18 in fedora 40, this change
does not sacrifice anything.
after this change, clang-tidy workflow should be back to normal.
Fixesscylladb/scylladb#19621
Signed-off-by: Kefu Chai <kefu.chai@scylladb.com>
Closesscylladb/scylladb#19628
Although Scylla already exposes metrics keeping track of various information related to hinted handoff, all of them correspond to either storing or sending hints. However, when debugging, it's also crucial to be aware of how many hints are coming to a given node and what their size is. Unfortunately, the existing metrics are not enough to obtain that information.
This PR introduces the following new metrics:
* `sent_bytes_total` – the total size of the hints that have been sent from a given shard,
* `received_hints_total` – the total number of hints that a given shard has received,
* `received_hints_bytes_total` – the total size of the hints a given shard has received.
It also renames `hints_manager_sent` to `hints_manager_sent_total` to avoid conflicts of prefixes between that metric and `sent_bytes_total` in tests.
Fixesscylladb/scylladb#10987Closesscylladb/scylladb#18976
* github.com:scylladb/scylladb:
db/hints: Add a metric for the size of sent hints
service/storage_proxy: Add metrics for received hints
The view builder is doing write operations to the database.
In order for the view builder to shutdown gracefully without errors, we
need to ensure the database can handle writes while it is drained.
The commit changes the drain order, so that view builder is drained
before the database shuts down.
Fixesscylladb/scylladb#18929Closesscylladb/scylladb#19609
Recently, the code in paxos_state::prepare(), paxos_state::accept() and
paxos_state::learn() was coroutinized by 58912c2cc1, 887a5a8f62 and
2b7acdb32c respectively. This introduced a regression: the latency
histogram updater code, was moved from a finally() to a defer(). Unlike
the former, the latter runs in a noexcept context so the possible
replica::no_such_column_family raised from the latency update code now
crashes the node, instead of failing just the paxos operation as before.
Fix by only updating the latency histogram if the table still exists.
Fixes: scylladb/scylladb#19620Closesscylladb/scylladb#19623
The page was missing from the docs. I created the page based on
the information in the download center (which will be closed down soon)
and other ScyllaDB resources.
Closesscylladb/scylladb#19577
In 2446cce, we stopped trying to attempt to create
endpoint managers for invalid hint directories
even when their names represented IP addresses or
host IDs. In this commit, we add logging informing
the user about it.
Refs scylladb/scylladb#19173Closesscylladb/scylladb#19618
Now that the CPU concurency limit is configurable, new reads might be
ready to execute right after the current one was executed. So move the
poll for admitting new reads into the inner loop, to prevent the
situation where the inner loop yields and a concurrent
do_wait_admission() finds that there are waiters (queued because at the
time they arrived to the semaphore, the _ready_list was not empty) but it
is is possible to admit a new read. When this happens the semaphore will
dump diagnostics to help debug the apparent contradiction, which can
generate a lot of log spam. Moving the poll into the inner loop prevents
the false-positive contradiction detection from firing.
Refs: scylladb/scylladb#19017Closesscylladb/scylladb#19600
This is the first patch from series which would allow us to unify raft command code. Property we want to achieve is that all modifications performed by a single raft command can be made visible atomically. This helps to exclude accidental dependencies across subsystem updates and make easier to reason about state.
Here we alter functions schema code so that changes are first applied to a copy of declared functions and then made visible atomically. Later work will apply similar strategy to the whole schema.
Relates scylladb/scylladb#19153Closesscylladb/scylladb#19598
* github.com:scylladb/scylladb:
cql3: functions: make modification functions accessible only via batch class
db: replica: batch functions schema modifications
cql3: functions: introduce class for batching functions modifications
cql3: functions: make functions class non-static
cql3: functions: remove reduntant class access specifiers
cql3: functions: remove unused java snippet
Before each function change was immediately visible as
during event notification logic yielded.
Now we first gather the modifications and then commit them.
Further work will broaden the scope of atomicity to the whole
schema and even across other subsystems.
When balancer fails to find a node to balance drained tablets into, it
throws an exception with tablet id and node id, but it's also good to
know more details about the balancing state that lead to failure
refs: #19504
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
Closesscylladb/scylladb#19588
It will hold a temporary shallow copy of declared functions.
Then each modification adds/removes/replaces stored function object.
At the end change is commited by moving temporary copy to the
main functions class instance.
This series is another approach of https://github.com/scylladb/scylladb/pull/18646 and https://github.com/scylladb/scylladb/pull/19181. In this series we only change where the view backlog gets
updated - we do not assure that the view update backlog returned in a response is necessarily the backlog
that increased due to the corresponding write, the returned backlog may be outdated up to 10ms. Because
this series does not include this change, it's considerably less complex and it doesn't modify the common
write patch, so no particular performance considerations were needed in that context. The issue being fixed
is still the same, the full description can be seen below.
When a replica applies a write on a table which has a materialized view
it generates view updates. These updates take memory which is tracked
by `database::_view_update_concurrency_sem`, separate on each shard.
The fraction of units taken from the semaphore to the semaphore limit
is the shard's view update backlog. Based on these backlogs, we want
to estimate how busy a node is with its view updates work. We do that
by taking the max backlog across all shards.
To avoid excessive cross-shard operations, the node's (max) backlog isn't
calculated each time we need it, but up to 1 time per 10ms (the `_interval`) with an optimization where the backlog of the calculating shard is immediately up-to-date (we don't need cross-shard operations for it):
```
update_backlog node_update_backlog::fetch() {
auto now = clock::now();
if (now >= _last_update.load(std::memory_order_relaxed) + _interval) {
_last_update.store(now, std::memory_order_relaxed);
auto new_max = boost::accumulate(
_backlogs,
update_backlog::no_backlog(),
[] (const update_backlog& lhs, const per_shard_backlog& rhs) {
return std::max(lhs, rhs.load());
});
_max.store(new_max, std::memory_order_relaxed);
return new_max;
}
return std::max(fetch_shard(this_shard_id()), _max.load(std::memory_order_relaxed));
}
```
For the same reason, even when we do calculate the new node's backlog,
we don't read from the `_view_update_concurrency_sem`. Instead, for
each shard we also store a update_backlog atomic which we use for
calculation:
```
struct per_shard_backlog {
// Multiply by 2 to defeat the prefetcher
alignas(seastar::cache_line_size * 2) std::atomic<update_backlog> backlog = update_backlog::no_backlog();
need_publishing need_publishing = need_publishing::no;
update_backlog load() const {
return backlog.load(std::memory_order_relaxed);
}
};
std::vector<per_shard_backlog> _backlogs;
```
Due to this distinction, the update_backlog atomic need to be updated
separately, when the `_view_update_concurrency_sem` changes.
This is done by calling `storage_proxy::update_view_update_backlog`, which reads the `_view_update_concurrency_sem` of the shard (in `database::get_view_update_backlog`)
and then calls node`_update_backlog::add` where the read backlog
is stored in the atomic:
```
void storage_proxy::update_view_update_backlog() {
_max_view_update_backlog.add(get_db().local().get_view_update_backlog());
}
void node_update_backlog::add(update_backlog backlog) {
_backlogs[this_shard_id()].backlog.store(backlog, std::memory_order_relaxed);
_backlogs[this_shard_id()].need_publishing = need_publishing::yes;
}
```
For this implementation of calculating the node's view update backlog to work,
we need the atomics to be updated correctly when the semaphores of corresponding
shards change.
The main event where the view update backlog changes is an incoming write
request. That's why when handling the request and preparing a response
we update the backlog calling `storage_proxy::get_view_update_backlog` (also
because we want to read the backlog and send it in the response):
backlog update after local view updates (`storage_proxy::send_to_live_endpoints` in `mutate_begin`)
```
auto lmutate = [handler_ptr, response_id, this, my_address, timeout] () mutable {
return handler_ptr->apply_locally(timeout, handler_ptr->get_trace_state())
.then([response_id, this, my_address, h = std::move(handler_ptr), p = shared_from_this()] {
// make mutation alive until it is processed locally, otherwise it
// may disappear if write timeouts before this future is ready
got_response(response_id, my_address, get_view_update_backlog());
});
};
backlog update after remote view updates (storage_proxy::remote::handle_write)
auto f = co_await coroutine::as_future(send_mutation_done(netw::messaging_service::msg_addr{reply_to, shard}, trace_state_ptr,
shard, response_id, p->get_view_update_backlog()));
```
Now assume that on a certain node we have a write request received on shard A,
which updates a row on shard B (A!=B). As a result, shard B will generate view
updates and consume units from its `_view_update_concurrency_sem`, but will
not update its atomic in `_backlogs` yet. Because both shards in the example
are on the same node, shard A will perform a local write calling `lmutate` shown
above. In the `lmutate` call, the `apply_locally` will initiate the actual write on
shard B and the `storage_proxy::update_view_update_backlog` will be called back
on shard A. In no place will the backlog atomic on shard B get updated even
though it increased in size due to the view updates generated there.
Currently, what we calculate there doesn't really matter - it's only used for the
MV flow control delays, so currently, in this scenario, we may only overload
a replica causing failed replica writes which will be later retried as hints. However,
when we add MV admission control, the calculated backlog will be the difference
between an accepted and a rejected request.
Fixes: https://github.com/scylladb/scylladb/issues/18542
Without admission control (https://github.com/scylladb/scylladb/pull/18334), this patch doesn't affect much, so I'm marking it as backport/none
Closesscylladb/scylladb#19341
* github.com:scylladb/scylladb:
test: add test for view backlog not being updated on correct shard
test: move auxiliary methods for waiting until a view is built to util
mv: update view update backlog when it increases on correct shard
This is done to ease code reuse in the following commit.
It'd also help should we ever want properly mount functions
class to schema object instead of static storage.
since fedora 38 is EOL. and fedora 39 comes with fmt v10.0.0, also,
we've switched to the build image based on fedora 40, which ships
fmt-devel v10.2.1, there is no need to use fmt::streamed() when
the corresponding format_as() as available.
simpler this way.
Signed-off-by: Kefu Chai <kefu.chai@scylladb.com>
Closesscylladb/scylladb#19594
to avoid warning like
```
DeprecationWarning: datetime.datetime.utcfromtimestamp() is deprecated and scheduled for removal in a future version. Use timezone-aware objects to represent datetimes in UTC: datetime.datetime.fromtimestamp(timestamp, datetime.UTC).
```
and to be future-proof, let's use the offset-aware timestamp.
Signed-off-by: Kefu Chai <kefu.chai@scylladb.com>
Closesscylladb/scylladb#19536
zstd.h is a header provided by libzstd, so let's include it with
brackets, more consistent this way.
Signed-off-by: Kefu Chai <kefu.chai@scylladb.com>
Closesscylladb/scylladb#19538
We disabled coredump compression by default because it was too slow,
but recent versions of systemd-coredump supports faster zstd based compression,
so let's enable compression by default when zstd support detected.
Related scylladb/scylla-machine-image#462Closesscylladb/scylladb#18854
This short series fixes Alternator's "/localnodes" request to allow a node's external IP address - configured with `broadcast_rpc_address` - to be listed instead of its usual, internal, IP address.
The first patch fixes a bug in gossiper::get_rpc_address(), which the second patch needs to implement the feature. The second patch also contains regression tests.
Fixes#18711.
Closesscylladb/scylladb#18828
* github.com:scylladb/scylladb:
alternator: fix "/localnodes" to use broadcast_rpc_address
gossiper: fix get_rpc_address() for this node
They don't need to modify the captured objects. In fact, they must not
do it in the first place, because the request can be called more than
once and the buffers must not change between those invocations.
For the memory_sink_buffers there must be const method to get the vector
of temporary_buffers themselves.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
Closesscylladb/scylladb#19599
Log the sstable origin when its bloom filter is being rebuilt. The
origin has to be passed to the method by the caller as it is not
available in the sstable object when the filter is rebuilt.
Signed-off-by: Lakshmi Narayanan Sreethar <lakshmi.sreethar@scylladb.com>
Closesscylladb/scylladb#19601
This patch adds a test for reproducing issue https://github.com/scylladb/scylladb/issues/18542
The test performs writes on a table with a materialized view and
checks that the view backlog increases. To get the current view
update backlog, a new metric "view_update_backlog" is added to
the `storage_proxy` metrics. The metric differs from the metric
from `database` metric with the same name by taking the backlog
from the max_view_update_backlog which keeps view update backlogs
from all shards which may be a bit outdated, instead of taking
the backlog by checking the view_update_semaphore which the backlog
is based on directly.
In many materialized view tests we need to wait until a view is built before
actually working on it, future tests will also need it. In existing tests
we use the same, duplicated method for achieving that.
In this patch the method is deduplicated and moved to pylib/util.py
and existing tests are modified to use it instead.
When performing a write, we should update the view update backlog
on the shard where the mutation is actually applied. Instead,
currently we only update it on the shard that initially received
the write request (which didn't change at all) and as a result,
the backlog on the correct shard and the aggregated max view update
backlog are not updated at all.
This patch enables updating the backlog on the correct shard. The
update is now performed just after the view generation and propagation
finishes, so that all backlog increases are noted and the backlog is
ready to be used in the write response.
Additionally, after this patch, we no longer (falsely) assume that
the backlog is modified on the same shard as where we later read it
to attach to a response. However, we still compare the aggregated
backlog from all shards and the backlog from the shard retrieving
the max, as with a shard-aware driver, it's likely the exact shard
whose backlog changed.
forward_service is nondescriptive and misnamed, as it does more than
forward requests. It's a classic map/reduce algorithm (and in fact one
of its parameters is "reducer"), so name it accordingly.
The name "forward" leaked into the wire protocol for the messaging
service RPC isolation cookie, so it's kept there. It's also maintained
in the name of the logger (for "nodetool setlogginglevel") for
compatibility with tests.
Closesscylladb/scylladb#19444
before this change, when linking scylla-main, the linker discards
the unreferenced symbols defined by zstd.cc. but we use constructor
of static variable `registerator` to register the zstd compressor,
this variable is not used from the linker's point of view. but we
do rely on the side effect of its constructor.
that's why the rules generated by CMake fails to build tests and
scylla executables with zstd support. that's why we have following
test failure:
```
boost.sstable_3_x_test.test_uncompressed_collections_read
...
[Exception] - no_such_class: unable to find class 'org.apache.cassandra.io.compress.ZstdCompressor'
== [File] - seastar/src/testing/seastar_test.cc
== [Line] - 43
```
in this change, we single out zstd.cc and build it as an archive,
so that scylla-main can include as a whole. an alternative is to
link scylla-main as a whole archive, but that might increase the disk
foot print when building lots of tests -- some of them do not use all
symbols exposed by scylla-main, and can potentially have smaller
size if linker can discard the unused symbols.
Refs https://github.com/scylladb/scylladb/issues/2717
---
cmake related change, hence no need to backport.
Closesscylladb/scylladb#19539
* github.com:scylladb/scylladb:
build: cmake: include the whole archive of zstd.a
build: cmake: find libzstd before using it
If an httpd body writer is called with output_stream<>, it mist close the stream on its own regardless of any exceptions it may generate while working, otherwise stream destructor may step on non-closed assertion. Stepped on with different handler, see #19541
Coroutinize the handler as the first step while at it (though the fix would have been notably shorter if done with .finally() lambda)
Closesscylladb/scylladb#19543
* github.com:scylladb/scylladb:
api: Close response stream of get_compaction_history()
api: Flush output stream in get_compaction_history() call
api: Coroutinize get_compaction_history inner function
when running `make setup`, we could have following failure:
```
Installing the current project: scylla (4.3.0)
The current project could not be installed: No file/folder found for package scylla
If you do not want to install the current project use --no-root
```
because docs is not a proper python project named "scylla",
and do not have a directory structure expected by poetry. what we
expect from poetry, is to manage the dependencies for building
the document.
so, in this change, we install in the `non-package` mode when running
`poetry install`, this skips the root package, which does not exist.
as an alternative, we could put an empty `scylla.py` under `docs`
directory, but that'd be overkill. or we could pass `--no-root`
to `poetry install`, but would be ideal if we can keep the settings
in a single place.
see also https://python-poetry.org/docs/basic-usage/#operating-modes,
and https://python-poetry.org/docs/cli/#options-2, for more
details on the settings and command line options of poetry.
please note this setting was added to poetry 1.8, so the required
poetry version is updated. we might need to upgrade poetry in existing
installation.
Signed-off-by: Kefu Chai <kefu.chai@scylladb.com>
Closesscylladb/scylladb#19498
Currently, a pending replica that applies a write on a table that has
materialized views, will build all the view updates as a normal replica,
only to realize at a late point, in db::view::get_view_natural_endpoint(),
that it doesn't have a paired view replica to send the updates to. It will
then either drop the view updates, or send them to a pending view
replica, if such exists.
This work is unnecessary since it may be dropped, and even if there is a
pending view replica to send the updates to, the updates that are built
by the pending replica may be wrong since it may have incomplete
information.
This commit fixes the inefficiency by skipping the view update building
step when applying an update on a pending replica.
The metric total_view_updates_on_wrong_node is added to count the cases
that a view update is determined to be unnecessary.
The test reproduces the scenario of writing to a table and applying
the update on a pending replica, and verifies that the pending replica
doesn't try to build view updates.
Fixesscylladb/scylladb#19152Closesscylladb/scylladb#19488
The reader concurrency semaphore restricts the concurrency of reads that require CPU (intention: they read from the cache) to 1, meaning that if there is even a single active read which declares that it needs just CPU to proceed, no new read is admitted. This is meant to keep the concurrency of reads in the cache at 1. The idea is that concurrency in the cache is not useful: it just leads to the reactor rotating between these reads, all of the finishing later then they could if they were the only active read in the cache.
This was observed to backfire in the case where there reads from a single table are mostly very fast, but on some keys are very slow (hint: collection full of tombstones). In this case the slow read keeps up the fast reads in the queue, increasing the 99th percentile latencies significantly.
This series proposes to fix this, by making the CPU concurrency configurable. We don't like tunables like this and this is not a proper fix, but a workaround. The proper fix would be to allow to cut any page early, but we cannot cut a page in the middle of a row. We could maybe have a way of detecting slow reads and excluding them from the CPU concurrency. This would be a heuristic and it would be hard to get right. So in this series a robust and simple configurable is offered, which can be used on those few clusters which do suffer from the too strict concurrency limit. We have seen it in very few cases so far, so this doesn't seem to be wide-spread.
Fixes: https://github.com/scylladb/scylladb/issues/19017
This fixes a regression introduced in 5.0, so we have to backport to all currently supported releases
Closesscylladb/scylladb#19018
* github.com:scylladb/scylladb:
test/boost/reader_concurrency_semaphore_test: add test for live-configurable cpu concurrenc Please enter the commit message for your changes. Lines starting
test/boost/reader_concurrency_semaphore_test: hoist require_can_admit
reader_concurrency_semaphore: wire in the configurable cpu concurrency
reader_concurrency_semaphore: add cpu_concurrency constructor parameter
db/config: introduce reader_concurrency_semahore_cpu_concurrency
Co-routinize paxos_state functions to make them more readable.
* 'gleb/coroutineze-paxos-state' of github.com:scylladb/scylla-dev:
paxos: simplify paxos_state::prepare code to not work with raw futures
paxos: co-routinize paxos_state::learn function
paxos: remove no longer used with_locked_key functions
paxos: co-routinize paxos_state::accept function
paxos: co-routinize paxos_state::prepare function
paxos: introduce get_replica_lock() function to take RAII guard for local paxos table access
this changeset adds a filter to customize the rendering of default
values, and enables the `scylladb_cc_properties` extension to display
the logging message related options. it prepares for the further
improvements in
https://opensource.docs.scylladb.com/master/reference/configuration-parameters.html.
this changeset also prepare for the improvements requested by #19463
---
it's an improvement in the document, hence no need to backport.
Closesscylladb/scylladb#19483
* github.com:scylladb/scylladb:
config: add descriptions for default_log_level and friends
config: define log_to_syslog in a different line
docs: parse log_legacy_value as declarations of config option
