The previous SSL enablement patches do make uses of these
options but they are still marked as Unused.
Change this and also update the db/config.hh documentation
accordingly.
Syntax is now:
client_encryption_options:
enabled: true
certificate: <path-to-PEM-x509-cert> (default conf/scylla.crt)
keyfile: <path-to-PEM-x509-key> (default conf/scylla.key)
Fixes: #756.
Signed-off-by: Benoît Canet <benoit@scylladb.com>
Message-Id: <1452032073-6933-1-git-send-email-benoit@scylladb.com>
Compaction fixes from Raphael:
There were two problems causing issue 676:
1) max_purgeable was being miscalculated (fixed by b7d36af).
2) empty row not being removed by mutation_partition::do_compact
Testcase is added to make sure that a tombstone will be purged under
certain conditions.
do_compact() wasn't removing an empty row that is covered by a
tombstone. As a result, an empty partition could be written to a
sstable. To solve this problem, let's make trim_rows remove a
row that is considered to be empty. A row is empty if it has no
tombstone, no marker and no cells.
Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>
"This is another version of the repair overhaul, to avoid streaming *all* the
data between nodes by sending checksums of token ranges and only streaming
ranges which contain differing data."
Support the "hosts" and "dataCenters" parameters of repair. The first
specifies the known good hosts to repair this host from (plus this host),
and the second asks to restrict the repair to the local data center (you
must issue the repair to a node in the data center you want to repair -
issuing the command to a data center other than the named one returns
an error).
For example these options are used by nodetool commands like:
nodetool repair -hosts 127.0.0.1,127.0.0.2 keyspace
nodetool repair -dc datacenter1
Signed-off-by: Nadav Har'El <nyh@scylladb.com>
The existing repair code always streamed the entire content of the
database. In this overhaul, we send "repair_checksum_range" messages to
the other nodes to verify whether they have exactly the same data as
this node, and if they do, we avoid streaming the identical code.
We make an attempt to split the token ranges up to contain an estimated
100 keys each, and send these ranges' checksums. Future versions of this
code will need to improve this estimation (and make this "100" a parameter)
Signed-off-by: Nadav Har'El <nyh@scylladb.com>
This patch adds a function sync_range() for synchronizing all partitions
in a given token range between a set of replicas (this node and a list of
neighbors).
Repair will call this function once it has decided that the data the
replicas hold in this range is not identical.
The implementation streams all the data in the given range, from each of
the neighbors to this node - so now this node contains the most up-to-date
data. It then streams the resulting data back to all the neighbors.
Signed-off-by: Nadav Har'El <nyh@scylladb.com>
This patch adds a new type of message, "REPAIR_CHECKSUM_RANGE" to scylla's
"messaging_service" RPC mechanism, for the use of repair:
With this message the repair's master host tells a slave host to calculate
the checksum of a column-family's partitions in a given token range, and
return that checksum.
The implementation of this message uses the checksum_range() function
defined in the previous patch.
Signed-off-by: Nadav Har'El <nyh@scylladb.com>
This patch adds functions for calculating the checksum of all the
partitions in a given token range in the given column-family - either
in the current shard, or across all shards in this node.
Signed-off-by: Nadav Har'El <nyh@scylladb.com>
This patch adds a mechanism for calculating a checksum for a set of
partitions. The repair process will use these checksums to compare the
data held by different replicas.
We use a strong checksum (SHA-256) for each individual partition in the set,
and then a simple XOR of those checksums to produce a checksum for the
entire set. XOR is good enough for merging strong checksums, and allows us
to independently calculate the checksums of different subsets of the
original sets - e.g., each shard can calculate its own checksum and we
can XOR the resulting checksums to get the final checksum.
Apache Cassandra uses a very similar checksum scheme, also using SHA-256
and XOR. One small difference in the implementation is that we include the
partition key in its checksum, while Cassandra don't, which I believe to
have no real justification (although it is very unlikely to cause problems
in practice). See further discussion on this in
https://issues.apache.org/jira/browse/CASSANDRA-10728.
Signed-off-by: Nadav Har'El <nyh@scylladb.com>
A cut-and-paste accident in query::to_partition_range caused the wrong
end's inclusiveness to be tested.
Signed-off-by: Nadav Har'El <nyh@scylladb.com>
Everything except alter_table_statement::announce_migration() is
translated. announce_migration() has to wait for multi schema support to
be merged.
Signed-off-by: Paweł Dziepak <pdziepak@scylladb.com>
We have an API that wraps open_file_dma which we use in some places, but in
many other places we call the reactor version directly.
This patch changes the latter to match the former. It will have the added benefit
of allowing us to make easier changes to these interfaces if needed.
Signed-off-by: Glauber Costa <glauber@scylladb.com>
Message-Id: <29296e4ec6f5e84361992028fe3f27adc569f139.1451950408.git.glauber@scylladb.com>
This exception was not caught properly as a std::exception
by report_failed_future call to report_exception because the
superclass std::exception was not initialized.
Fixes#669.
Signed-off-by: Benoît Canet <benoit@scylladb.com>
Recently, Scylla was changed to mandate the use of XFS
for its data directories, unless the flag --developer-mode true
is provided. So during the AMI setup stage, if the user
did not provide extra disks for the setup scripts to prepare,
the scylla service will refuse to start. Therefore, the
message in scylla_prepare has to be changed to an actual
error message, and the file name, to be changed to
something that reflects the event that happened.
Signed-off-by: Lucas Meneghel Rodrigues <lmr@scylladb.com>
Avi Kivity (1):
Merge "rpc negotiation fixes" from Gleb
Gleb Natapov (3):
rpc: fix peer address printing during logging
rpc: make server send negotiation frame back before closing connection on error
rpc: fix documentation for negotiation procedure.
Nadav Har'El (1):
fix operator<<() for std::vector<T>
Tomasz Grabiec (1):
core/byteorder: Add missing include
scylla-ami.sh moved some ami specific files. This parts have been
dropped when converging scylla-ami into scylla_install. Fixing that.
Signed-off-by: Shlomi Livne <shlomi@scylladb.com>
The script imports the /etc/sysconfig/scylla-server for configuration
settings (NR_PAGES). The /etc/sysconfig/scylla-server iincludes an AMI
param which is of string value and called as a last step in
scylla_install (after scylla_bootparam_setup has been initated).
The AMI variable is setup in scylla_install and is used in multiple
scripts. To resolve the conflict moving the import of
/etc/sysconfig/scylla-server after the AMI variable has been compared.
Fixes: #744
Signed-off-by: Shlomi Livne <shlomi@scylladb.com>
Switch to CentOS 7 as the Docker base image. It's more stable and
updated less frequently than Fedora. As a bonus, it's Thrift package
doesn't pull the world as a dependency which reduces image size from 700
MB to 380 MB.
Suggested by Avi.
Message-Id: <1451911969-26647-1-git-send-email-penberg@scylladb.com>
If requests are delayed downstream from the cql server, and the client is
able to generate unrelated requests without limit, then the transient memory
consumed by the requests will overflow the shard's capacity.
Fix by adding a semaphore to cap the amount of transient memory occupied by
requests.
Fixes#674.
compatible: can be cast, keeps sort order
value-compatible: can be cast, may change sort order
frozen: values participate in sort order
unfrozen: only sort keys participate in sort order
Fixes#740.
"Messaging service inherits from seastar::async_sharded_service which
guaranties that sharded<>::stop() will not complete until all references
to the service will not go away. It was done specifically to avoid using
more verbose gate interface in sharded<> services since it turned out
that almost all of them need one eventually. Unfortunately patches to
add redundant gate to messaging_service sneaked pass my review. The series
reverts them."
