Compare commits

...

5 Commits

Author SHA1 Message Date
copilot-swe-agent[bot]
7ebf5a02d1 Fix variable quoting and profile path in optimized_clang.sh
Co-authored-by: mykaul <4655593+mykaul@users.noreply.github.com>
2025-12-14 15:57:50 +00:00
copilot-swe-agent[bot]
03daab52ec Initial plan 2025-12-14 15:55:30 +00:00
Botond Dénes
7e7e378a4b Merge 'Revert "Merge 'Add option to use sstable identifier in snapshot' from Benny Halevy"' from null
Reverts commit 8192f45e84.

The merge exposed a critical bug where truncate operations during table drop with auto-snapshot fail, causing Raft applier fiber to stop with unhandled exceptions. This leads to schema inconsistencies across nodes and test failures with "Keyspace does not exist" errors.

**Root Cause**

Commit 19b6207f modified `truncate_table_on_all_shards` to set `use_sstable_identifier = true`:

```cpp
// Before (working)
co_await table::snapshot_on_all_shards(sharded_db, table_shards, name);

// After (broken)
auto opts = db::snapshot_options{.use_sstable_identifier = true};
co_await table::snapshot_on_all_shards(sharded_db, table_shards, name, opts);
```

This triggers exceptions during snapshot that propagate through Raft state machine, causing:
- Raft applier stops: `raft::state_machine_error` at `raft/server.cc:1369`
- Schema changes fail to propagate
- Nodes report non-existent keyspaces for valid schemas

**Changes**

Reverts 15 files (200 deletions, 74 insertions):
- Removes `use_sstable_identifier` from truncate/snapshot code paths
- Reverts `snapshot_options` struct back to simple `skip_flush` boolean
- Removes REST API and nodetool `--use-sstable-identifier` parameter
- Removes feature tests from `test/boost/database_test.cc`

No backport required - the original feature was merged to master only and never released.

<!-- START COPILOT ORIGINAL PROMPT -->

<details>

<summary>Original prompt</summary>

----

*This section details on the original issue you should resolve*

<issue_title>test_table_drop_with_auto_snapshot failed with InvalidRequest</issue_title>
<issue_description>Seen in: https://jenkins.scylladb.com/job/scylla-master/job/next/9968//testReport
Logs: [download](https://downloads.scylladb.com/unstable/scylla/master/testLogs/2025-12-08T15:05:42Z/)

Error message:
```
cassandra.InvalidRequest: Error from server: code=2200 [Invalid query] message="Keyspace test does not exist"
```

Stacktrace:
```
manager = <test.pylib.manager_client.ManagerClient object at 0xffff734e6c10>

    @pytest.mark.asyncio
    async def test_table_drop_with_auto_snapshot(manager: ManagerClient):
        logger.info("Bootstrapping cluster")
        cfg = { 'auto_snapshot': True }
        servers = await manager.servers_add(3, config = cfg)

        cql = manager.get_cql()

        # Increases the chance of tablet migration concurrent with schema change
        await inject_error_on(manager, "tablet_allocator_shuffle", servers)

        for i in range(3):
            await cql.run_async("DROP KEYSPACE IF EXISTS test;")
            await cql.run_async("CREATE KEYSPACE IF NOT EXISTS test WITH replication = {'class': 'NetworkTopologyStrategy', 'replication_factor': 1} AND tablets = {'initial': 8 };")
            await cql.run_async("CREATE TABLE IF NOT EXISTS test.tbl_sample_kv (id int, value text, PRIMARY KEY (id));")
>           await cql.run_async("INSERT INTO test.tbl_sample_kv (id, value) VALUES (1, 'ala');")
E           cassandra.InvalidRequest: Error from server: code=2200 [Invalid query] message="Keyspace test does not exist"

test/cluster/test_tablets2.py:173: InvalidRequest
```
</issue_description>

<agent_instructions>this issue was exposed by commit 8192f45e84, please send a pull request reverting that merge commit and mark it as fixing this github issue.</agent_instructions>

<comments>
<comment_new><author>@yaronkaikov</author><body>
@denesb is this something in your team area? if not , please feel free to delegate it or un-assign yourself :-)</body></comment_new>
<comment_new><author>@nyh</author><body>
This is very strange. Clearly the keyspace `test` does exist at this point, because we created it two lines above and also we ran `CREATE TABLE .. test.tbl_sample_kv` which would have failed if the keyspace `test` didn't exist - so it must exit, no?

In the past, we had a bug where the running `CREATE KEYSPACE IF NOT EXISTS` forgot to set the "schema modified" event in the response so it failed to wait for schema agreement, but 1. we fixed this bug (https://github.com/scylladb/scylladb/pull/18819 by @nuivall ) and 2. this bug didn't happen in this case, where CREATE TABLE deed had work to do.

But I just realized something... Our fix in https://github.com/scylladb/scylladb/pull/18819 only applies to CREATE KEYSPACE / TABLE / VIEW / TYPE statements. It wasn't applied to `DROP KEYSPACE` - and it should have been....

But I don't have a good theory how a bug like https://github.com/scylladb/scylladb/pull/18819 can explain this specific test failure. Different schema operations are already linearized, so if a `CREATE TABLE test.tbl_sample_kv` succeeded, I don't see how there could possibly be any earlier `DROP KEYSPACE test` that suddenly springs to life. Unless we have a serious bug in our raft-based schema operations.</body></comment_new>
<comment_new><author>@nyh</author><body>
Another bug we could have in theory is that the Python driver's async `cql.run_async` might have a bug where it is not waiting for the schema agreement despite being told to wait. If it doesn't wait for schema agreement, this can easily explain this bug:
1. the CREATE KEYSPACE, CREATE TABLE both are sent to node A, but
2. the last INSERT INTO is sent to node B which is not yet aware of this new keyspace and table, and fails.

Copilot claims that **execute_async() does have this bug!**

> For schema-altering statements, schema agreement (meaning all nodes agree on the new schema) is important before running follow-up operations, but this is enforced only by synchronous helpers like Session.execute(), not the asynchronous version.
> If you use execute_async() for schema operations, you are responsible for checking schema agreement yourself, using [Session.check_schema_agreement()](https://docs.datastax.com/en/developer/python-driver/latest/api/cassandra/cluster/#cassandra.cluster.Session.check_schema_agreement) or (in newer code) ResponseFuture.check_schema_agreement.
> According to [a discussion on the DataStax support forum](https://support.datastax.com/s/article/Does-the-Python-Driver-for-Cassandra-Wait-for-Schema-Agreement-after-a-Schema-Change?language=en_US) and the [driver’s source code](7f12a5e1c6/cassandra/cluster.py (L487)), schema agreement is not ch...

</details>

<!-- START COPILOT CODING AGENT SUFFIX -->

- Fixes scylladb/scylladb#27501

<!-- START COPILOT CODING AGENT TIPS -->
---

Closes scylladb/scylladb#27604

* github.com:scylladb/scylladb:
  Revert "Merge 'Add option to use sstable identifier in snapshot' from Benny Halevy"
  Initial plan
2025-12-12 13:20:49 +02:00
copilot-swe-agent[bot]
77ee7f3417 Revert "Merge 'Add option to use sstable identifier in snapshot' from Benny Halevy"
This reverts commit 8192f45e84.

The merge exposed a bug where truncate (via drop) fails and causes Raft
errors, leading to schema inconsistencies across nodes. This results in
test_table_drop_with_auto_snapshot failures with 'Keyspace test does not exist'
errors.

The specific problematic change was in commit 19b6207f which modified
truncate_table_on_all_shards to set use_sstable_identifier = true. This
causes exceptions during truncate that are not properly handled, leading
to Raft applier fiber stopping and nodes losing schema synchronization.
2025-12-12 03:55:13 +00:00
copilot-swe-agent[bot]
0ff89a58be Initial plan 2025-12-12 03:48:12 +00:00
16 changed files with 76 additions and 202 deletions

View File

@@ -729,14 +729,6 @@
"allowMultiple":false,
"type":"boolean",
"paramType":"query"
},
{
"name":"use_sstable_identifier",
"description":"Use the sstable identifier UUID, if available, rather than the sstable generation.",
"required":false,
"allowMultiple":false,
"type":"boolean",
"paramType":"query"
}
]
},

View File

@@ -2020,16 +2020,12 @@ void set_snapshot(http_context& ctx, routes& r, sharded<db::snapshot_ctl>& snap_
auto tag = req->get_query_param("tag");
auto column_families = split(req->get_query_param("cf"), ",");
auto sfopt = req->get_query_param("sf");
auto usiopt = req->get_query_param("use_sstable_identifier");
db::snapshot_options opts = {
.skip_flush = strcasecmp(sfopt.c_str(), "true") == 0,
.use_sstable_identifier = strcasecmp(usiopt.c_str(), "true") == 0
};
auto sf = db::snapshot_ctl::skip_flush(strcasecmp(sfopt.c_str(), "true") == 0);
std::vector<sstring> keynames = split(req->get_query_param("kn"), ",");
try {
if (column_families.empty()) {
co_await snap_ctl.local().take_snapshot(tag, keynames, opts);
co_await snap_ctl.local().take_snapshot(tag, keynames, sf);
} else {
if (keynames.empty()) {
throw httpd::bad_param_exception("The keyspace of column families must be specified");
@@ -2037,7 +2033,7 @@ void set_snapshot(http_context& ctx, routes& r, sharded<db::snapshot_ctl>& snap_
if (keynames.size() > 1) {
throw httpd::bad_param_exception("Only one keyspace allowed when specifying a column family");
}
co_await snap_ctl.local().take_column_family_snapshot(keynames[0], column_families, tag, opts);
co_await snap_ctl.local().take_column_family_snapshot(keynames[0], column_families, tag, sf);
}
co_return json_void();
} catch (...) {
@@ -2072,8 +2068,7 @@ void set_snapshot(http_context& ctx, routes& r, sharded<db::snapshot_ctl>& snap_
auto info = parse_scrub_options(ctx, std::move(req));
if (!info.snapshot_tag.empty()) {
db::snapshot_options opts = {.skip_flush = false, .use_sstable_identifier = false};
co_await snap_ctl.local().take_column_family_snapshot(info.keyspace, info.column_families, info.snapshot_tag, opts);
co_await snap_ctl.local().take_column_family_snapshot(info.keyspace, info.column_families, info.snapshot_tag, db::snapshot_ctl::skip_flush::no);
}
compaction::compaction_stats stats;

View File

@@ -146,8 +146,7 @@ void set_tasks_compaction_module(http_context& ctx, routes& r, sharded<service::
auto info = parse_scrub_options(ctx, std::move(req));
if (!info.snapshot_tag.empty()) {
db::snapshot_options opts = {.skip_flush = false, .use_sstable_identifier = false};
co_await snap_ctl.local().take_column_family_snapshot(info.keyspace, info.column_families, info.snapshot_tag, opts);
co_await snap_ctl.local().take_column_family_snapshot(info.keyspace, info.column_families, info.snapshot_tag, db::snapshot_ctl::skip_flush::no);
}
auto& compaction_module = db.local().get_compaction_manager().get_task_manager_module();

View File

@@ -65,7 +65,7 @@ future<> snapshot_ctl::run_snapshot_modify_operation(noncopyable_function<future
});
}
future<> snapshot_ctl::take_snapshot(sstring tag, std::vector<sstring> keyspace_names, snapshot_options opts) {
future<> snapshot_ctl::take_snapshot(sstring tag, std::vector<sstring> keyspace_names, skip_flush sf) {
if (tag.empty()) {
throw std::runtime_error("You must supply a snapshot name.");
}
@@ -74,21 +74,21 @@ future<> snapshot_ctl::take_snapshot(sstring tag, std::vector<sstring> keyspace_
std::ranges::copy(_db.local().get_keyspaces() | std::views::keys, std::back_inserter(keyspace_names));
};
return run_snapshot_modify_operation([tag = std::move(tag), keyspace_names = std::move(keyspace_names), opts, this] () mutable {
return do_take_snapshot(std::move(tag), std::move(keyspace_names), opts);
return run_snapshot_modify_operation([tag = std::move(tag), keyspace_names = std::move(keyspace_names), sf, this] () mutable {
return do_take_snapshot(std::move(tag), std::move(keyspace_names), sf);
});
}
future<> snapshot_ctl::do_take_snapshot(sstring tag, std::vector<sstring> keyspace_names, snapshot_options opts) {
future<> snapshot_ctl::do_take_snapshot(sstring tag, std::vector<sstring> keyspace_names, skip_flush sf) {
co_await coroutine::parallel_for_each(keyspace_names, [tag, this] (const auto& ks_name) {
return check_snapshot_not_exist(ks_name, tag);
});
co_await coroutine::parallel_for_each(keyspace_names, [this, tag = std::move(tag), opts] (const auto& ks_name) {
return replica::database::snapshot_keyspace_on_all_shards(_db, ks_name, tag, opts);
co_await coroutine::parallel_for_each(keyspace_names, [this, tag = std::move(tag), sf] (const auto& ks_name) {
return replica::database::snapshot_keyspace_on_all_shards(_db, ks_name, tag, bool(sf));
});
}
future<> snapshot_ctl::take_column_family_snapshot(sstring ks_name, std::vector<sstring> tables, sstring tag, snapshot_options opts) {
future<> snapshot_ctl::take_column_family_snapshot(sstring ks_name, std::vector<sstring> tables, sstring tag, skip_flush sf) {
if (ks_name.empty()) {
throw std::runtime_error("You must supply a keyspace name");
}
@@ -99,14 +99,14 @@ future<> snapshot_ctl::take_column_family_snapshot(sstring ks_name, std::vector<
throw std::runtime_error("You must supply a snapshot name.");
}
return run_snapshot_modify_operation([this, ks_name = std::move(ks_name), tables = std::move(tables), tag = std::move(tag), opts] () mutable {
return do_take_column_family_snapshot(std::move(ks_name), std::move(tables), std::move(tag), opts);
return run_snapshot_modify_operation([this, ks_name = std::move(ks_name), tables = std::move(tables), tag = std::move(tag), sf] () mutable {
return do_take_column_family_snapshot(std::move(ks_name), std::move(tables), std::move(tag), sf);
});
}
future<> snapshot_ctl::do_take_column_family_snapshot(sstring ks_name, std::vector<sstring> tables, sstring tag, snapshot_options opts) {
future<> snapshot_ctl::do_take_column_family_snapshot(sstring ks_name, std::vector<sstring> tables, sstring tag, skip_flush sf) {
co_await check_snapshot_not_exist(ks_name, tag, tables);
co_await replica::database::snapshot_tables_on_all_shards(_db, ks_name, std::move(tables), std::move(tag), opts);
co_await replica::database::snapshot_tables_on_all_shards(_db, ks_name, std::move(tables), std::move(tag), bool(sf));
}
future<> snapshot_ctl::clear_snapshot(sstring tag, std::vector<sstring> keyspace_names, sstring cf_name) {

View File

@@ -38,13 +38,10 @@ class backup_task_impl;
} // snapshot namespace
struct snapshot_options {
bool skip_flush = false;
bool use_sstable_identifier = false;
};
class snapshot_ctl : public peering_sharded_service<snapshot_ctl> {
public:
using skip_flush = bool_class<class skip_flush_tag>;
struct table_snapshot_details {
int64_t total;
int64_t live;
@@ -73,8 +70,8 @@ public:
*
* @param tag the tag given to the snapshot; may not be null or empty
*/
future<> take_snapshot(sstring tag, snapshot_options opts = {}) {
return take_snapshot(tag, {}, opts);
future<> take_snapshot(sstring tag, skip_flush sf = skip_flush::no) {
return take_snapshot(tag, {}, sf);
}
/**
@@ -83,7 +80,7 @@ public:
* @param tag the tag given to the snapshot; may not be null or empty
* @param keyspace_names the names of the keyspaces to snapshot; empty means "all"
*/
future<> take_snapshot(sstring tag, std::vector<sstring> keyspace_names, snapshot_options opts = {});
future<> take_snapshot(sstring tag, std::vector<sstring> keyspace_names, skip_flush sf = skip_flush::no);
/**
* Takes the snapshot of multiple tables. A snapshot name must be specified.
@@ -92,7 +89,7 @@ public:
* @param tables a vector of tables names to snapshot
* @param tag the tag given to the snapshot; may not be null or empty
*/
future<> take_column_family_snapshot(sstring ks_name, std::vector<sstring> tables, sstring tag, snapshot_options opts = {});
future<> take_column_family_snapshot(sstring ks_name, std::vector<sstring> tables, sstring tag, skip_flush sf = skip_flush::no);
/**
* Remove the snapshot with the given name from the given keyspaces.
@@ -130,8 +127,8 @@ private:
friend class snapshot::backup_task_impl;
future<> do_take_snapshot(sstring tag, std::vector<sstring> keyspace_names, snapshot_options opts = {} );
future<> do_take_column_family_snapshot(sstring ks_name, std::vector<sstring> tables, sstring tag, snapshot_options opts = {});
future<> do_take_snapshot(sstring tag, std::vector<sstring> keyspace_names, skip_flush sf = skip_flush::no);
future<> do_take_column_family_snapshot(sstring ks_name, std::vector<sstring> tables, sstring tag, skip_flush sf = skip_flush::no);
};
}

View File

@@ -17,7 +17,7 @@ SYNOPSIS
[(-u <username> | --username <username>)] snapshot
[(-cf <table> | --column-family <table> | --table <table>)]
[(-kc <kclist> | --kc.list <kclist>)]
[(-sf | --skip-flush)] [--use-sstable-identifier] [(-t <tag> | --tag <tag>)] [--] [<keyspaces...>]
[(-sf | --skip-flush)] [(-t <tag> | --tag <tag>)] [--] [<keyspaces...>]
OPTIONS
.......
@@ -37,8 +37,6 @@ Parameter Descriptio
-------------------------------------------------------------------- -------------------------------------------------------------------------------------
-sf / --skip-flush Do not flush memtables before snapshotting (snapshot will not contain unflushed data)
-------------------------------------------------------------------- -------------------------------------------------------------------------------------
--use-sstable-identifier Use the sstable identifier UUID, if available, rather than the sstable generation.
-------------------------------------------------------------------- -------------------------------------------------------------------------------------
-t <tag> / --tag <tag> The name of the snapshot
==================================================================== =====================================================================================

View File

@@ -2810,26 +2810,26 @@ future<> database::drop_cache_for_keyspace_on_all_shards(sharded<database>& shar
});
}
future<> database::snapshot_table_on_all_shards(sharded<database>& sharded_db, table_id uuid, sstring tag, db::snapshot_options opts) {
if (!opts.skip_flush) {
future<> database::snapshot_table_on_all_shards(sharded<database>& sharded_db, table_id uuid, sstring tag, bool skip_flush) {
if (!skip_flush) {
co_await flush_table_on_all_shards(sharded_db, uuid);
}
auto table_shards = co_await get_table_on_all_shards(sharded_db, uuid);
co_await table::snapshot_on_all_shards(sharded_db, table_shards, tag, opts);
co_await table::snapshot_on_all_shards(sharded_db, table_shards, tag);
}
future<> database::snapshot_tables_on_all_shards(sharded<database>& sharded_db, std::string_view ks_name, std::vector<sstring> table_names, sstring tag, db::snapshot_options opts) {
return parallel_for_each(table_names, [&sharded_db, ks_name, tag = std::move(tag), opts] (auto& table_name) {
future<> database::snapshot_tables_on_all_shards(sharded<database>& sharded_db, std::string_view ks_name, std::vector<sstring> table_names, sstring tag, bool skip_flush) {
return parallel_for_each(table_names, [&sharded_db, ks_name, tag = std::move(tag), skip_flush] (auto& table_name) {
auto uuid = sharded_db.local().find_uuid(ks_name, table_name);
return snapshot_table_on_all_shards(sharded_db, uuid, tag, opts);
return snapshot_table_on_all_shards(sharded_db, uuid, tag, skip_flush);
});
}
future<> database::snapshot_keyspace_on_all_shards(sharded<database>& sharded_db, std::string_view ks_name, sstring tag, db::snapshot_options opts) {
future<> database::snapshot_keyspace_on_all_shards(sharded<database>& sharded_db, std::string_view ks_name, sstring tag, bool skip_flush) {
auto& ks = sharded_db.local().find_keyspace(ks_name);
co_await coroutine::parallel_for_each(ks.metadata()->cf_meta_data(), [&, tag = std::move(tag), opts] (const auto& pair) -> future<> {
co_await coroutine::parallel_for_each(ks.metadata()->cf_meta_data(), [&, tag = std::move(tag), skip_flush] (const auto& pair) -> future<> {
auto uuid = pair.second->id();
co_await snapshot_table_on_all_shards(sharded_db, uuid, tag, opts);
co_await snapshot_table_on_all_shards(sharded_db, uuid, tag, skip_flush);
});
}
@@ -2951,12 +2951,7 @@ future<> database::truncate_table_on_all_shards(sharded<database>& sharded_db, s
auto truncated_at = truncated_at_opt.value_or(db_clock::now());
auto name = snapshot_name_opt.value_or(
format("{:d}-{}", truncated_at.time_since_epoch().count(), cf.schema()->cf_name()));
// Use the sstable identifier in snapshot names to allow de-duplication of sstables
// at backup time even if they were migrated across shards or nodes and were renamed a given a new generation.
// We hard-code that here since we have no way to pass this option to auto-snapshot and
// it is always safe to use the sstable identifier for the sstable generation.
auto opts = db::snapshot_options{.use_sstable_identifier = true};
co_await table::snapshot_on_all_shards(sharded_db, table_shards, name, opts);
co_await table::snapshot_on_all_shards(sharded_db, table_shards, name);
}
co_await sharded_db.invoke_on_all([&] (database& db) {

View File

@@ -1040,12 +1040,12 @@ public:
private:
using snapshot_file_set = foreign_ptr<std::unique_ptr<std::unordered_set<sstring>>>;
future<snapshot_file_set> take_snapshot(sstring jsondir, db::snapshot_options opts);
future<snapshot_file_set> take_snapshot(sstring jsondir);
// Writes the table schema and the manifest of all files in the snapshot directory.
future<> finalize_snapshot(const global_table_ptr& table_shards, sstring jsondir, std::vector<snapshot_file_set> file_sets);
static future<> seal_snapshot(sstring jsondir, std::vector<snapshot_file_set> file_sets);
public:
static future<> snapshot_on_all_shards(sharded<database>& sharded_db, const global_table_ptr& table_shards, sstring name, db::snapshot_options opts);
static future<> snapshot_on_all_shards(sharded<database>& sharded_db, const global_table_ptr& table_shards, sstring name);
future<std::unordered_map<sstring, snapshot_details>> get_snapshot_details();
static future<snapshot_details> get_snapshot_details(std::filesystem::path snapshot_dir, std::filesystem::path datadir);
@@ -2009,9 +2009,9 @@ public:
static future<> drop_cache_for_table_on_all_shards(sharded<database>& sharded_db, table_id id);
static future<> drop_cache_for_keyspace_on_all_shards(sharded<database>& sharded_db, std::string_view ks_name);
static future<> snapshot_table_on_all_shards(sharded<database>& sharded_db, table_id id, sstring tag, db::snapshot_options opts);
static future<> snapshot_tables_on_all_shards(sharded<database>& sharded_db, std::string_view ks_name, std::vector<sstring> table_names, sstring tag, db::snapshot_options opts);
static future<> snapshot_keyspace_on_all_shards(sharded<database>& sharded_db, std::string_view ks_name, sstring tag, db::snapshot_options opts);
static future<> snapshot_table_on_all_shards(sharded<database>& sharded_db, table_id id, sstring tag, bool skip_flush);
static future<> snapshot_tables_on_all_shards(sharded<database>& sharded_db, std::string_view ks_name, std::vector<sstring> table_names, sstring tag, bool skip_flush);
static future<> snapshot_keyspace_on_all_shards(sharded<database>& sharded_db, std::string_view ks_name, sstring tag, bool skip_flush);
public:
bool update_column_family(schema_ptr s);

View File

@@ -3268,7 +3268,7 @@ future<> table::write_schema_as_cql(const global_table_ptr& table_shards, sstrin
}
// Runs the orchestration code on an arbitrary shard to balance the load.
future<> table::snapshot_on_all_shards(sharded<database>& sharded_db, const global_table_ptr& table_shards, sstring name, db::snapshot_options opts) {
future<> table::snapshot_on_all_shards(sharded<database>& sharded_db, const global_table_ptr& table_shards, sstring name) {
auto* so = std::get_if<storage_options::local>(&table_shards->get_storage_options().value);
if (so == nullptr) {
throw std::runtime_error("Snapshotting non-local tables is not implemented");
@@ -3291,7 +3291,7 @@ future<> table::snapshot_on_all_shards(sharded<database>& sharded_db, const glob
co_await io_check([&jsondir] { return recursive_touch_directory(jsondir); });
co_await coroutine::parallel_for_each(smp::all_cpus(), [&] (unsigned shard) -> future<> {
file_sets.emplace_back(co_await smp::submit_to(shard, [&] {
return table_shards->take_snapshot(jsondir, opts);
return table_shards->take_snapshot(jsondir);
}));
});
co_await io_check(sync_directory, jsondir);
@@ -3300,22 +3300,19 @@ future<> table::snapshot_on_all_shards(sharded<database>& sharded_db, const glob
});
}
future<table::snapshot_file_set> table::take_snapshot(sstring jsondir, db::snapshot_options opts) {
tlogger.trace("take_snapshot {}: use_sstable_identifier={}", jsondir, opts.use_sstable_identifier);
future<table::snapshot_file_set> table::take_snapshot(sstring jsondir) {
tlogger.trace("take_snapshot {}", jsondir);
auto sstable_deletion_guard = co_await get_sstable_list_permit();
auto tables = *_sstables->all() | std::ranges::to<std::vector<sstables::shared_sstable>>();
auto table_names = std::make_unique<std::unordered_set<sstring>>();
auto& ks_name = schema()->ks_name();
auto& cf_name = schema()->cf_name();
co_await _sstables_manager.dir_semaphore().parallel_for_each(tables, [&, opts] (sstables::shared_sstable sstable) -> future<> {
auto gen = co_await io_check([sstable, &dir = jsondir, opts] {
return sstable->snapshot(dir, opts.use_sstable_identifier);
co_await _sstables_manager.dir_semaphore().parallel_for_each(tables, [&jsondir, &table_names] (sstables::shared_sstable sstable) {
table_names->insert(sstable->component_basename(sstables::component_type::Data));
return io_check([sstable, &dir = jsondir] {
return sstable->snapshot(dir);
});
auto fname = sstable->component_basename(ks_name, cf_name, sstable->get_version(), gen, sstable->get_format(), sstables::component_type::Data);
table_names->insert(fname);
});
co_return make_foreign(std::move(table_names));
}

View File

@@ -2117,14 +2117,11 @@ sstable::write_scylla_metadata(shard_id shard, struct run_identifier identifier,
}
sstable_id sid;
// Force a random sstable_id for testing purposes
bool random_sstable_identifier = utils::get_local_injector().is_enabled("random_sstable_identifier");
if (!random_sstable_identifier && generation().is_uuid_based()) {
if (generation().is_uuid_based()) {
sid = sstable_id(generation().as_uuid());
} else {
sid = sstable_id(utils::UUID_gen::get_time_UUID());
auto msg = random_sstable_identifier ? "forced random sstable_id" : "has numerical generation";
sstlog.info("SSTable {} {}. SSTable identifier in scylla_metadata set to {}", get_filename(), msg, sid);
sstlog.info("SSTable {} has numerical generation. SSTable identifier in scylla_metadata set to {}", get_filename(), sid);
}
_components->scylla_metadata->data.set<scylla_metadata_type::SSTableIdentifier>(scylla_metadata::sstable_identifier{sid});
@@ -2538,11 +2535,8 @@ std::vector<std::pair<component_type, sstring>> sstable::all_components() const
return all;
}
future<generation_type> sstable::snapshot(const sstring& dir, bool use_sstable_identifier) const {
// Use the sstable identifier UUID if available to enable global de-duplication of sstables in backup.
generation_type gen = (use_sstable_identifier && _sstable_identifier) ? generation_type(_sstable_identifier->uuid()) : _generation;
co_await _storage->snapshot(*this, dir, storage::absolute_path::yes, gen);
co_return gen;
future<> sstable::snapshot(const sstring& dir) const {
return _storage->snapshot(*this, dir, storage::absolute_path::yes);
}
future<> sstable::change_state(sstable_state to, delayed_commit_changes* delay_commit) {

View File

@@ -397,10 +397,6 @@ public:
return _version;
}
format_types get_format() const {
return _format;
}
// Returns the total bytes of all components.
uint64_t bytes_on_disk() const;
file_size_stats get_file_size_stats() const;
@@ -442,10 +438,7 @@ public:
std::vector<std::pair<component_type, sstring>> all_components() const;
// When use_sstable_identifier is true and the sstable identifier is available,
// use it to name the sstable in the snapshot, rather than the sstable generation.
// Returns the generation used for snapshot.
future<generation_type> snapshot(const sstring& dir, bool use_sstable_identifier = false) const;
future<> snapshot(const sstring& dir) const;
// Delete the sstable by unlinking all sstable files
// Ignores all errors.

View File

@@ -31,7 +31,6 @@
#include "replica/database.hh"
#include "utils/assert.hh"
#include "utils/lister.hh"
#include "utils/rjson.hh"
#include "partition_slice_builder.hh"
#include "mutation/frozen_mutation.hh"
#include "test/lib/mutation_source_test.hh"
@@ -39,7 +38,6 @@
#include "service/migration_manager.hh"
#include "sstables/sstables.hh"
#include "sstables/generation_type.hh"
#include "sstables/sstable_version.hh"
#include "db/config.hh"
#include "db/commitlog/commitlog_replayer.hh"
#include "db/commitlog/commitlog.hh"
@@ -53,7 +51,6 @@
#include "db/system_keyspace.hh"
#include "db/view/view_builder.hh"
#include "replica/mutation_dump.hh"
#include "utils/error_injection.hh"
using namespace std::chrono_literals;
using namespace sstables;
@@ -615,13 +612,13 @@ future<> do_with_some_data(std::vector<sstring> cf_names, std::function<future<>
});
}
future<> take_snapshot(cql_test_env& e, sstring ks_name = "ks", sstring cf_name = "cf", sstring snapshot_name = "test", db::snapshot_options opts = {}) {
future<> take_snapshot(cql_test_env& e, sstring ks_name = "ks", sstring cf_name = "cf", sstring snapshot_name = "test", bool skip_flush = false) {
try {
auto uuid = e.db().local().find_uuid(ks_name, cf_name);
co_await replica::database::snapshot_table_on_all_shards(e.db(), uuid, snapshot_name, opts);
co_await replica::database::snapshot_table_on_all_shards(e.db(), uuid, snapshot_name, skip_flush);
} catch (...) {
testlog.error("Could not take snapshot for {}.{} snapshot_name={} skip_flush={} use_sstable_identifier={}: {}",
ks_name, cf_name, snapshot_name, opts.skip_flush, opts.use_sstable_identifier, std::current_exception());
testlog.error("Could not take snapshot for {}.{} snapshot_name={} skip_flush={}: {}",
ks_name, cf_name, snapshot_name, skip_flush, std::current_exception());
throw;
}
}
@@ -635,37 +632,6 @@ future<std::set<sstring>> collect_files(fs::path path) {
co_return ret;
}
static bool is_component(const sstring& fname, const sstring& suffix) {
return fname.ends_with(suffix);
}
static std::set<sstring> collect_sstables(const std::set<sstring>& all_files, const sstring& suffix) {
// Verify manifest against the files in the snapshots dir
auto pred = [&suffix] (const sstring& fname) {
return is_component(fname, suffix);
};
return std::ranges::filter_view(all_files, pred) | std::ranges::to<std::set<sstring>>();
}
// Validate that the manifest.json lists exactly the SSTables present in the snapshot directory
static future<> validate_manifest(const fs::path& snapshot_dir, const std::set<sstring>& in_snapshot_dir) {
sstring suffix = "-Data.db";
auto sstables_in_snapshot = collect_sstables(in_snapshot_dir, suffix);
std::set<sstring> sstables_in_manifest;
auto manifest_str = co_await util::read_entire_file_contiguous(snapshot_dir / "manifest.json");
auto manifest_json = rjson::parse(manifest_str);
auto& manifest_files = manifest_json["files"];
BOOST_REQUIRE(manifest_files.IsArray());
for (auto& f : manifest_files.GetArray()) {
if (is_component(f.GetString(), suffix)) {
sstables_in_manifest.insert(f.GetString());
}
}
testlog.debug("SSTables in manifest.json: {}", fmt::join(sstables_in_manifest, ", "));
BOOST_REQUIRE_EQUAL(sstables_in_snapshot, sstables_in_manifest);
}
static future<> snapshot_works(const std::string& table_name) {
return do_with_some_data({"cf"}, [table_name] (cql_test_env& e) {
take_snapshot(e, "ks", table_name).get();
@@ -685,8 +651,6 @@ static future<> snapshot_works(const std::string& table_name) {
// all files were copied and manifest was generated
BOOST_REQUIRE_EQUAL(in_table_dir, in_snapshot_dir);
validate_manifest(snapshot_dir, in_snapshot_dir).get();
return make_ready_future<>();
}, true);
}
@@ -705,8 +669,7 @@ SEASTAR_TEST_CASE(index_snapshot_works) {
SEASTAR_TEST_CASE(snapshot_skip_flush_works) {
return do_with_some_data({"cf"}, [] (cql_test_env& e) {
db::snapshot_options opts = {.skip_flush = true};
take_snapshot(e, "ks", "cf", "test", opts).get();
take_snapshot(e, "ks", "cf", "test", true /* skip_flush */).get();
auto& cf = e.local_db().find_column_family("ks", "cf");
@@ -719,41 +682,6 @@ SEASTAR_TEST_CASE(snapshot_skip_flush_works) {
});
}
SEASTAR_TEST_CASE(snapshot_use_sstable_identifier_works) {
#ifndef SCYLLA_ENABLE_ERROR_INJECTION
fmt::print("Skipping test as it depends on error injection. Please run in mode where it's enabled (debug,dev).\n");
return make_ready_future<>();
#endif
sstring table_name = "cf";
// Force random sstable identifiers, otherwise the initial sstable_id is equal
// to the sstable generation and the test can't distinguish between them.
utils::get_local_injector().enable("random_sstable_identifier", false);
return do_with_some_data({table_name}, [table_name] (cql_test_env& e) -> future<> {
sstring tag = "test";
db::snapshot_options opts = {.use_sstable_identifier = true};
co_await take_snapshot(e, "ks", table_name, tag, opts);
auto& cf = e.local_db().find_column_family("ks", table_name);
auto table_directory = table_dir(cf);
auto snapshot_dir = table_directory / sstables::snapshots_dir / tag;
auto in_table_dir = co_await collect_files(table_directory);
// snapshot triggered a flush and wrote the data down.
BOOST_REQUIRE_GE(in_table_dir.size(), 9);
testlog.info("Files in table dir: {}", fmt::join(in_table_dir, ", "));
auto in_snapshot_dir = co_await collect_files(snapshot_dir);
testlog.info("Files in snapshot dir: {}", fmt::join(in_snapshot_dir, ", "));
in_table_dir.insert("manifest.json");
in_table_dir.insert("schema.cql");
// all files were copied and manifest was generated
BOOST_REQUIRE_EQUAL(in_table_dir.size(), in_snapshot_dir.size());
BOOST_REQUIRE_NE(in_table_dir, in_snapshot_dir);
co_await validate_manifest(snapshot_dir, in_snapshot_dir);
}, true);
}
SEASTAR_TEST_CASE(snapshot_list_okay) {
return do_with_some_data({"cf"}, [] (cql_test_env& e) {
auto& cf = e.local_db().find_column_family("ks", "cf");
@@ -1528,7 +1456,7 @@ SEASTAR_TEST_CASE(snapshot_with_quarantine_works) {
}
BOOST_REQUIRE(found);
co_await take_snapshot(e, "ks", "cf", "test", db::snapshot_options{.skip_flush = true});
co_await take_snapshot(e, "ks", "cf", "test", true /* skip_flush */);
testlog.debug("Expected: {}", expected);

View File

@@ -115,7 +115,7 @@ def compact_keyspace(cql, ks, flush_memtables=True):
args.extend([ks, cf])
run_nodetool(cql, "compact", *args)
def take_snapshot(cql, table, tag, skip_flush, use_sstable_identifier=False):
def take_snapshot(cql, table, tag, skip_flush):
ks, cf = table.split('.')
if has_rest_api(cql):
requests.post(f'{rest_api_url(cql)}/storage_service/snapshots/', params={'kn': ks, 'cf' : cf, 'tag': tag, 'sf': skip_flush})
@@ -123,8 +123,6 @@ def take_snapshot(cql, table, tag, skip_flush, use_sstable_identifier=False):
args = ['--tag', tag, '--table', cf]
if skip_flush:
args.append('--skip-flush')
if use_sstable_identifier:
args.append('--use-sstable-identifier')
args.append(ks)
run_nodetool(cql, "snapshot", *args)

View File

@@ -99,7 +99,7 @@ def test_listsnapshots_no_snapshots(nodetool, request):
assert res.stdout == "Snapshot Details: \nThere are no snapshots\n"
def check_snapshot_out(res, tag, ktlist, skip_flush, use_sstable_identifier=False):
def check_snapshot_out(res, tag, ktlist, skip_flush):
"""Check that the output of nodetool snapshot contains the expected messages"""
if len(ktlist) == 0:
@@ -110,7 +110,7 @@ def check_snapshot_out(res, tag, ktlist, skip_flush, use_sstable_identifier=Fals
pattern = re.compile("Requested creating snapshot\\(s\\)"
f" for \\[{keyspaces}\\]"
f" with snapshot name \\[(.+)\\]"
f" and options \\{{skip_flush={str(skip_flush).lower()}, use_sstable_identifier={str(use_sstable_identifier).lower()}\\}}")
f" and options \\{{skipFlush={str(skip_flush).lower()}\\}}")
print(res)
print(pattern)
@@ -138,13 +138,13 @@ def test_snapshot_keyspace(nodetool):
res = nodetool("snapshot", "--tag", tag, "ks1", expected_requests=[
expected_request("POST", "/storage_service/snapshots",
params={"tag": tag, "sf": "false", "use_sstable_identifier": "false", "kn": "ks1"})
params={"tag": tag, "sf": "false", "kn": "ks1"})
])
check_snapshot_out(res.stdout, tag, ["ks1"], False)
res = nodetool("snapshot", "--tag", tag, "ks1", "ks2", expected_requests=[
expected_request("POST", "/storage_service/snapshots",
params={"tag": tag, "sf": "false", "use_sstable_identifier": "false", "kn": "ks1,ks2"})
params={"tag": tag, "sf": "false", "kn": "ks1,ks2"})
])
check_snapshot_out(res.stdout, tag, ["ks1", "ks2"], False)
@@ -155,13 +155,13 @@ def test_snapshot_keyspace_with_table(nodetool, option_name):
res = nodetool("snapshot", "--tag", tag, "ks1", option_name, "tbl", expected_requests=[
expected_request("POST", "/storage_service/snapshots",
params={"tag": tag, "sf": "false", "use_sstable_identifier": "false", "kn": "ks1", "cf": "tbl"})
params={"tag": tag, "sf": "false", "kn": "ks1", "cf": "tbl"})
])
check_snapshot_out(res.stdout, tag, ["ks1"], False)
res = nodetool("snapshot", "--tag", tag, "ks1", option_name, "tbl1,tbl2", expected_requests=[
expected_request("POST", "/storage_service/snapshots",
params={"tag": tag, "sf": "false", "use_sstable_identifier": "false", "kn": "ks1", "cf": "tbl1,tbl2"})
params={"tag": tag, "sf": "false", "kn": "ks1", "cf": "tbl1,tbl2"})
])
check_snapshot_out(res.stdout, tag, ["ks1"], False)
@@ -186,7 +186,7 @@ class kn_param(NamedTuple):
def test_snapshot_keyspace_table_single_arg(nodetool, param, scylla_only):
tag = "my_snapshot"
req_params = {"tag": tag, "sf": "false", "use_sstable_identifier": "false", "kn": param.kn}
req_params = {"tag": tag, "sf": "false", "kn": param.kn}
if param.cf:
req_params["cf"] = param.cf
@@ -202,19 +202,19 @@ def test_snapshot_ktlist(nodetool, option_name):
res = nodetool("snapshot", "--tag", tag, option_name, "ks1.tbl1", expected_requests=[
expected_request("POST", "/storage_service/snapshots",
params={"tag": tag, "sf": "false", "use_sstable_identifier": "false", "kn": "ks1", "cf": "tbl1"})
params={"tag": tag, "sf": "false", "kn": "ks1", "cf": "tbl1"})
])
check_snapshot_out(res.stdout, tag, ["ks1.tbl1"], False)
res = nodetool("snapshot", "--tag", tag, option_name, "ks1.tbl1,ks2.tbl2", expected_requests=[
expected_request("POST", "/storage_service/snapshots",
params={"tag": tag, "sf": "false", "use_sstable_identifier": "false", "kn": "ks1.tbl1,ks2.tbl2"})
params={"tag": tag, "sf": "false", "kn": "ks1.tbl1,ks2.tbl2"})
])
check_snapshot_out(res.stdout, tag, ["ks1.tbl1", "ks2.tbl2"], False)
res = nodetool("snapshot", "--tag", tag, option_name, "ks1,ks2", expected_requests=[
expected_request("POST", "/storage_service/snapshots",
params={"tag": tag, "sf": "false", "use_sstable_identifier": "false", "kn": "ks1,ks2"})
params={"tag": tag, "sf": "false", "kn": "ks1,ks2"})
])
check_snapshot_out(res.stdout, tag, ["ks1" ,"ks2"], False)
@@ -229,8 +229,7 @@ def test_snapshot_ktlist(nodetool, option_name):
{"ks": ["ks1", "ks2"], "tbl": []},
])
@pytest.mark.parametrize("skip_flush", [False, True])
@pytest.mark.parametrize("use_sstable_identifier", [False, True])
def test_snapshot_options_matrix(nodetool, tag, ktlist, skip_flush, use_sstable_identifier):
def test_snapshot_options_matrix(nodetool, tag, ktlist, skip_flush):
cmd = ["snapshot"]
params = {}
@@ -243,11 +242,8 @@ def test_snapshot_options_matrix(nodetool, tag, ktlist, skip_flush, use_sstable_
if skip_flush:
cmd.append("--skip-flush")
if use_sstable_identifier:
cmd.append("--use-sstable-identifier")
params["sf"] = str(skip_flush).lower()
params["use_sstable_identifier"] = str(use_sstable_identifier).lower()
if ktlist:
if "tbl" in ktlist:
@@ -277,7 +273,7 @@ def test_snapshot_options_matrix(nodetool, tag, ktlist, skip_flush, use_sstable_
expected_request("POST", "/storage_service/snapshots", params=params)
])
check_snapshot_out(res.stdout, tag, keyspaces, skip_flush, use_sstable_identifier)
check_snapshot_out(res.stdout, tag, keyspaces, skip_flush)
def test_snapshot_multiple_keyspace_with_table(nodetool):

View File

@@ -2362,23 +2362,16 @@ void snapshot_operation(scylla_rest_client& client, const bpo::variables_map& vm
params["sf"] = "false";
}
if (vm.contains("use-sstable-identifier")) {
params["use_sstable_identifier"] = "true";
} else {
params["use_sstable_identifier"] = "false";
}
client.post("/storage_service/snapshots", params);
if (kn_msg.empty()) {
kn_msg = params["kn"];
}
fmt::print(std::cout, "Requested creating snapshot(s) for [{}] with snapshot name [{}] and options {{skip_flush={}, use_sstable_identifier={}}}\n",
fmt::print(std::cout, "Requested creating snapshot(s) for [{}] with snapshot name [{}] and options {{skipFlush={}}}\n",
kn_msg,
params["tag"],
params["sf"],
params["use_sstable_identifier"]);
params["sf"]);
fmt::print(std::cout, "Snapshot directory: {}\n", params["tag"]);
}
@@ -4605,7 +4598,6 @@ For more information, see: {}
typed_option<sstring>("keyspace-table-list", "The keyspace.table pair(s) to snapshot, multiple ones can be joined with ','"),
typed_option<sstring>("tag,t", "The name of the snapshot"),
typed_option<>("skip-flush", "Do not flush memtables before snapshotting (snapshot will not contain unflushed data)"),
typed_option<>("use-sstable-identifier", "Use the sstable identifier UUID, if available, rather than the sstable generation for the sstable file names within the snapshot dir and the manifest file"),
},
{
typed_option<std::vector<sstring>>("keyspaces", "The keyspaces to snapshot", -1),

View File

@@ -123,7 +123,7 @@ _get_distribution_components() {
continue
;;
esac
echo $target
echo "$target"
done
}
@@ -164,7 +164,7 @@ if [[ "${CLANG_BUILD}" = "INSTALL" ]]; then
echo "[clang-stage3] build the compiler applied CSPGO profile"
cd "${CLANG_BUILD_DIR}"
llvm-profdata merge build/csprofiles/default_*.profraw -output=csir.prof
llvm-profdata merge build/profiles/csir-*.profraw -output=csir.prof
llvm-profdata merge ir.prof csir.prof -output=combined.prof
rm -rf build
# linker flags are needed for BOLT