Introduce a subscribers_list class that exposes 3 methods:
- push_back(s) - adds a new element s to the back of the list
- remove(s) - removes an element s from the list
- for_each(f) - invoke f on each element of the list
- make a subscriber_list store shared_ptr to a subscriber
to allow removing (currently it stores a naked pointer to the object).
subscribers_list allows push_back() and remove() to be called while
another thread (e.g. seastar::async()) is in the middle of for_each().
Signed-off-by: Vlad Zolotarov <vladz@cloudius-systems.com>
New in v2:
- Simplify subscribers_list::remove() method.
- load_broadcaster: inherit from enable_shared_from_this instead
of async_sharded_service.
* seastar 9d8913a...9ae6407 (2):
> core/memory.cc: Declare member min_free_pages from cpu_pages struct
> http: All http replies should have version set
It may happen that the user will migrate a table to Scylla which
compaction strategy isn't supported yet, such as Data tiered.
Let's handle that by falling back to size-tiered compaction
strategy and printing a warning message.
Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>
remove() is the function used to remove every reference to a cf from
the compaction manager. This function works by removing cf from the
queue, and waiting for possible ongoing compaction on cf.
However, a cf may be re-queued by compaction manager task if there
is pending compaction by the end of compaction.
If cf is still referenced by the time remove() returns, we could end
up with an use-after-free. To fix that, a task shouldn't re-queue a
cf if it was asked to stop. The stat pending_tasks was also not
being updated when a cf was removed from the task queue.
Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>
The only place local_dc is checked during mutation sending is in
send_to_live_endpoints(), but current code pass it there throw several
function call layers. Simplify the code by getting local_dc when it is
used directly.
Since 4641dfff24 "service: Copy client
state to query state" after executing a query client state needs to be
merged back. If that's not done client_state::_last_timestamp_micros
won't be advanced properly and mutations originating from the same
source may have exactly the same timestamp.
Signed-off-by: Paweł Dziepak <pdziepak@scylladb.com>
Fix some PEP8 problems found in the tester code:
* Wrong spacing around operators
* Lines between class and function definitions
* Fixed some of the larger than 80 column statements
* Removed an unused import
Signed-off-by: Lucas Meneghel Rodrigues <lmr@scylladb.com>
_unreachable_endpoints is replicated to call cores. No need to query on
core 0.
This also fixes a bug in storage_proxy::truncate_blocking
which might access _unreachable_endpoints on non-zero cores.
The current code will try to print the output of a
subprocess.Popen().communicate() call even if that
call raised an exception and that output is None.
Let's fix this problem by only printing the output
if it's not None.
Signed-off-by: Lucas Meneghel Rodrigues <lmr@scylladb.com>
Since commit 5613979a85
broadcast address has to be set before it's used for the first
time.
Signed-off-by: Vlad Zolotarov <vladz@cloudius-systems.com>
* seastar 501e4cb...9d8913a (3):
> Add mutable to with_lock and do_with
> app-template: disable collectd by default
> reactor: use fdatasync() instead of fsync()
"Currently, CQL requests are processed on the same CPU core where the
connection lives in. This series adds infrastructure for migrating CQL
processing to other cores and implements a round-robin load balancing
algorithm that can be enabled with the "--load-balance=round-robin"
command line option. Load balancing is not enabled by default because we
need to first run performance tests to determine if the simple
round-robin algorithm is sufficient, or wheter we need to implement more
sophisticated dynamic load balancing."
