When using boost --output-format=XML flag for jenkins the output file is
garbled - it seems that some output is lost. To try and overcome this
changed impl to use a temporay file instead of a PIPE.
Signed-off-by: Shlomi Livne <shlomi@cloudius-systems.com>
"only lz4 is supported so far, but other algorithms should be fairly easy to
add, given that the infrastructure for compression is introduced here.
NOTE:
our sstable read code already reads compressed data file, so a test was added
to guarantee that data from a compressed sstable generated by our write code
can be successfully loaded."
lz4 is the unique compressor algorithm supported so far.
missing deflate and snappy algorithms.
Adding them should be relatively easy though.
Signed-off-by: Raphael S. Carvalho <raphaelsc@cloudius-systems.com>
Move raw_statement implementation out of the header file. This makes
life easier when we modify the class for clustering key support.
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>
The thread switching code assumed that we will always switch out of a
thread due to being blocked on an unavailable future. This allows
the core to store the blocked thread's context in the synthetic
continuation chained to that future (which switched back to that thread).
That assumption failed in one place: when we create a thread from within a
thread. In that case we switch to the new thread immediately, but forget
all about the old thread. We never come back to the old thread, and anything
that depends on it hangs.
Fix by creating a linked list of active thread contexts. These are all
threads that have been "preempted" by the act of creating a new thread,
terminated by the main, unthreaded, reactor context. This gives us a place
to store those threads and we come back to them and continue where we left
off.
Reported by Pekka.
"This patch series adds counters to the rpc clients then it expose them in the
messaging service with a rest API, it was tested with the messaging_service
test app that was modified to run the API."
The "mutation_reader" defined in database.cc is a convenient mechanism
for iterating over mutations. It can be useful for more than just
database.cc (I want to use it in the compaction code), so this patch moves
the type's definition to mutation.hh, and the make_memtable_reader()
function to memtable::make_reader() (in memtable.hh).
Signed-off-by: Nadav Har'El <nyh@cloudius-systems.com>
To test the messaging service API it needs to be included in the
messaging service test.
To test it, start the server with --stay-alive true
then you can use the API to get the messges information:
http://localhost:10001/messaging_service/messages/exception
will return the number of exception per connection
http://localhost:10001/messaging_service/command/completed
will return the number of completed command per connection.
Note that because the two servers are running on the same machine, to
prevent port conflict, the server port will be increment by one (i.e
10001)
Signed-off-by: Amnon Heiman <amnon@cloudius-systems.com>
This API gother information about messages passing in the system, in
this patch the following API will be supported:
/messaging_service/messages/replied
/messaging_service/messages/sent
/messaging_service/messages/pending
/messaging_service/messages/exception
/messaging_service/messages/respond_pending
/messaging_service/messages/dropped
The swagger defintion can be retrieved from:
/api-doc/messaging_service
In a typical scenario the API collect values from multiple distributed
instances. Sometimes it is needed to reduce multiple maps by merging
their keys and accumulating their values.
This is a helper function that can be used in map_reduce to perform maps
sum.
Signed-off-by: Amnon Heiman <amnon@cloudius-systems.com>
This adds a drop messages counter per verb type to the messaging
service. It will be used by the API to return the number of dropped
messages.
Signed-off-by: Amnon Heiman <amnon@cloudius-systems.com>
The messaging service holds a table of clients which the API needs
information from. This adds a foreach_client method that recieve a
functions and itererate over all the clients calling the given function
on each of them.
This implementation support the current table that holds unique ptr.
Signed-off-by: Amnon Heiman <amnon@cloudius-systems.com>
This adds a stub implementation of the storage service, to simplify
future implementation, variables that should be used in the
implementation are taken and stored.
Implementation return the currect type, but with stub values.
Signed-off-by: Amnon Heiman <amnon@cloudius-systems.com>
This patch complete the definition of the storage service.
It is a mapping of the StorageServiceMBean. When possible the following
convention where used:
API that does not return value will be POST or DELETE depends on the
context.
When possible, POST, DELETE and GET uses the same path
The following commands where added:
get_leaving_nodes
get_moving_nodes
get_joining_nodes
get_release_version
get_schema_version
get_all_data_file_locations
get_saved_caches_location
get_range_to_endpoint_map
get_pending_range_to_endpoint_map
describe_ring_jmx
get_host_id_map
get_load
get_load_map
get_current_generation_number
get_natural_endpoints
get_snapshot_details
take_snapshot
del_snapshot
true_snapshots_size
force_keyspace_compaction
force_keyspace_cleanup
scrub
upgrade_sstables
force_keyspace_flush
repair_async
force_terminate_all_repair_sessions
decommission
move
remove_node
get_removal_status
force_remove_completion
set_logging_level
get_logging_levels
get_operation_mode
is_starting
get_drain_progress
drain
truncate
get_keyspaces
update_snitch
stop_gossiping
start_gossiping
is_gossip_running
stop_daemon
is_initialized
stop_rpc_server
start_rpc_server
is_rpc_server_running
start_native_transport
stop_native_transport
is_native_transport_running
join_ring
is_joined
set_stream_throughput_mb_per_sec
get_stream_throughput_mb_per_sec
get_compaction_throughput_mb_per_sec
set_compaction_throughput_mb_per_sec
is_incremental_backups_enabled
set_incremental_backups_enabled
rebuild
bulk_load
bulk_load_async
reschedule_failed_deletions
load_new_ss_tables
sample_key_range
reset_local_schema
set_trace_probability
get_trace_probability
enable_auto_compaction
disable_auto_compaction
deliver_hints
get_cluster_name
get_partitioner_name
get_tombstone_warn_threshold
set_tombstone_warn_threshold
get_tombstone_failure_threshold
set_tombstone_failure_threshold
get_batch_size_failure_threshold
set_batch_size_failure_threshold
set_hinted_handoff_throttle_in_kb
Signed-off-by: Amnon Heiman <amnon@cloudius-systems.com>
Add attributes that we didn't support when the code was first
translated. Please note that we still don't cover all of the so somebody
needs to revisit this later.
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>
In sstables the paritioner name is store for validation. To allow Origin
to process our files we need to comply with Origin's paritioner name or
else Origin's SSTableReader::open fails on paritioner comparison.
