nixpkgs-suyu/nixos/modules/services/databases/cassandra.nix

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

565 lines
18 KiB
Nix
Raw Normal View History

{ config, lib, pkgs, ... }:
let
2021-04-29 08:39:35 +02:00
inherit (lib)
concatStringsSep
flip
literalMD
literalExpression
2021-04-29 08:39:35 +02:00
optionalAttrs
optionals
recursiveUpdate
mdDoc
2021-04-29 08:39:35 +02:00
mkEnableOption
mkIf
mkOption
types
versionAtLeast
;
cfg = config.services.cassandra;
2021-04-29 08:22:39 +02:00
defaultUser = "cassandra";
2021-04-29 08:22:39 +02:00
cassandraConfig = flip recursiveUpdate cfg.extraConfig (
{
commitlog_sync = "batch";
commitlog_sync_batch_window_in_ms = 2;
start_native_transport = cfg.allowClients;
cluster_name = cfg.clusterName;
partitioner = "org.apache.cassandra.dht.Murmur3Partitioner";
endpoint_snitch = "SimpleSnitch";
data_file_directories = [ "${cfg.homeDir}/data" ];
commitlog_directory = "${cfg.homeDir}/commitlog";
saved_caches_directory = "${cfg.homeDir}/saved_caches";
2021-04-29 08:39:35 +02:00
} // optionalAttrs (cfg.seedAddresses != [ ]) {
2021-04-29 08:22:39 +02:00
seed_provider = [
{
class_name = "org.apache.cassandra.locator.SimpleSeedProvider";
parameters = [{ seeds = concatStringsSep "," cfg.seedAddresses; }];
}
];
2021-04-29 08:39:35 +02:00
} // optionalAttrs (versionAtLeast cfg.package.version "3") {
2021-04-29 08:22:39 +02:00
hints_directory = "${cfg.homeDir}/hints";
}
);
cassandraConfigWithAddresses = cassandraConfig // (
if cfg.listenAddress == null
then { listen_interface = cfg.listenInterface; }
else { listen_address = cfg.listenAddress; }
) // (
if cfg.rpcAddress == null
then { rpc_interface = cfg.rpcInterface; }
else { rpc_address = cfg.rpcAddress; }
);
cassandraEtc = pkgs.stdenv.mkDerivation {
name = "cassandra-etc";
cassandraYaml = builtins.toJSON cassandraConfigWithAddresses;
cassandraEnvPkg = "${cfg.package}/conf/cassandra-env.sh";
cassandraLogbackConfig = pkgs.writeText "logback.xml" cfg.logbackConfig;
passAsFile = [ "extraEnvSh" ];
inherit (cfg) extraEnvSh;
buildCommand = ''
mkdir -p "$out"
echo "$cassandraYaml" > "$out/cassandra.yaml"
ln -s "$cassandraLogbackConfig" "$out/logback.xml"
( cat "$cassandraEnvPkg"
echo "# lines from services.cassandra.extraEnvSh: "
cat "$extraEnvShPath"
) > "$out/cassandra-env.sh"
# Delete default JMX Port, otherwise we can't set it using env variable
sed -i '/JMX_PORT="7199"/d' "$out/cassandra-env.sh"
# Delete default password file
sed -i '/-Dcom.sun.management.jmxremote.password.file=\/etc\/cassandra\/jmxremote.password/d' "$out/cassandra-env.sh"
'';
};
defaultJmxRolesFile =
builtins.foldl'
(left: right: left + right) ""
(map (role: "${role.username} ${role.password}") cfg.jmxRoles);
fullJvmOptions =
cfg.jvmOpts
2021-04-29 08:39:35 +02:00
++ optionals (cfg.jmxRoles != [ ]) [
"-Dcom.sun.management.jmxremote.authenticate=true"
"-Dcom.sun.management.jmxremote.password.file=${cfg.jmxRolesFile}"
2021-04-29 08:39:35 +02:00
] ++ optionals cfg.remoteJmx [
"-Djava.rmi.server.hostname=${cfg.rpcAddress}"
];
2021-04-29 08:22:39 +02:00
in
{
options.services.cassandra = {
2021-04-29 08:22:39 +02:00
enable = mkEnableOption ''
Apache Cassandra Scalable and highly available database.
'';
2021-04-29 08:22:39 +02:00
2019-04-07 07:22:41 +02:00
clusterName = mkOption {
type = types.str;
default = "Test Cluster";
description = mdDoc ''
2019-04-07 07:22:41 +02:00
The name of the cluster.
This setting prevents nodes in one logical cluster from joining
another. All nodes in a cluster must have the same value.
'';
};
2021-04-29 08:22:39 +02:00
user = mkOption {
type = types.str;
default = defaultUser;
description = mdDoc "Run Apache Cassandra under this user.";
};
2021-04-29 08:22:39 +02:00
group = mkOption {
type = types.str;
default = defaultUser;
description = mdDoc "Run Apache Cassandra under this group.";
};
2021-04-29 08:22:39 +02:00
homeDir = mkOption {
type = types.path;
default = "/var/lib/cassandra";
description = mdDoc ''
Home directory for Apache Cassandra.
'';
};
2021-04-29 08:22:39 +02:00
package = mkOption {
type = types.package;
default = pkgs.cassandra;
defaultText = literalExpression "pkgs.cassandra";
example = literalExpression "pkgs.cassandra_3_11";
description = mdDoc ''
The Apache Cassandra package to use.
'';
};
2021-04-29 08:22:39 +02:00
jvmOpts = mkOption {
type = types.listOf types.str;
2021-04-29 08:22:39 +02:00
default = [ ];
description = mdDoc ''
Populate the `JVM_OPT` environment variable.
'';
};
2021-04-29 08:22:39 +02:00
listenAddress = mkOption {
type = types.nullOr types.str;
default = "127.0.0.1";
example = null;
description = mdDoc ''
Address or interface to bind to and tell other Cassandra nodes
to connect to. You _must_ change this if you want multiple
nodes to be able to communicate!
Set {option}`listenAddress` OR {option}`listenInterface`, not both.
Leaving it blank leaves it up to
`InetAddress.getLocalHost()`. This will always do the "Right
Thing" _if_ the node is properly configured (hostname, name
resolution, etc), and the Right Thing is to use the address
associated with the hostname (it might not be).
Setting {option}`listenAddress` to `0.0.0.0` is always wrong.
'';
};
2021-04-29 08:22:39 +02:00
listenInterface = mkOption {
type = types.nullOr types.str;
default = null;
example = "eth1";
description = mdDoc ''
Set `listenAddress` OR `listenInterface`, not both. Interfaces
must correspond to a single address, IP aliasing is not
supported.
'';
};
2021-04-29 08:22:39 +02:00
rpcAddress = mkOption {
type = types.nullOr types.str;
default = "127.0.0.1";
example = null;
description = mdDoc ''
The address or interface to bind the native transport server to.
Set {option}`rpcAddress` OR {option}`rpcInterface`, not both.
Leaving {option}`rpcAddress` blank has the same effect as on
{option}`listenAddress` (i.e. it will be based on the configured hostname
of the node).
Note that unlike {option}`listenAddress`, you can specify `"0.0.0.0"`, but you
must also set `extraConfig.broadcast_rpc_address` to a value other
than `"0.0.0.0"`.
For security reasons, you should not expose this port to the
internet. Firewall it if needed.
'';
};
2021-04-29 08:22:39 +02:00
rpcInterface = mkOption {
type = types.nullOr types.str;
default = null;
example = "eth1";
description = mdDoc ''
Set {option}`rpcAddress` OR {option}`rpcInterface`, not both. Interfaces must
correspond to a single address, IP aliasing is not supported.
'';
};
2021-04-29 08:22:39 +02:00
logbackConfig = mkOption {
type = types.lines;
default = ''
<configuration scan="false">
<appender name="STDOUT" class="ch.qos.logback.core.ConsoleAppender">
<encoder>
<pattern>%-5level %date{HH:mm:ss,SSS} %msg%n</pattern>
</encoder>
</appender>
<root level="INFO">
<appender-ref ref="STDOUT" />
</root>
<logger name="com.thinkaurelius.thrift" level="ERROR"/>
</configuration>
'';
description = mdDoc ''
XML logback configuration for cassandra
'';
};
2021-04-29 08:22:39 +02:00
seedAddresses = mkOption {
type = types.listOf types.str;
default = [ "127.0.0.1" ];
description = mdDoc ''
The addresses of hosts designated as contact points in the cluster. A
joining node contacts one of the nodes in the seeds list to learn the
topology of the ring.
Set to `[ "127.0.0.1" ]` for a single node cluster.
'';
};
2021-04-29 08:22:39 +02:00
allowClients = mkOption {
type = types.bool;
default = true;
description = mdDoc ''
Enables or disables the native transport server (CQL binary protocol).
This server uses the same address as the {option}`rpcAddress`,
but the port it uses is not `rpc_port` but
`native_transport_port`. See the official Cassandra
docs for more information on these variables and set them using
{option}`extraConfig`.
'';
};
2021-04-29 08:22:39 +02:00
extraConfig = mkOption {
type = types.attrs;
2021-04-29 08:22:39 +02:00
default = { };
example =
2021-04-29 08:22:39 +02:00
{
commitlog_sync_batch_window_in_ms = 3;
};
description = mdDoc ''
Extra options to be merged into {file}`cassandra.yaml` as nix attribute set.
'';
};
2021-04-29 08:22:39 +02:00
2020-09-16 17:19:39 +02:00
extraEnvSh = mkOption {
type = types.lines;
default = "";
example = literalExpression ''"CLASSPATH=$CLASSPATH:''${extraJar}"'';
description = mdDoc ''
Extra shell lines to be appended onto {file}`cassandra-env.sh`.
2020-09-16 17:19:39 +02:00
'';
};
2021-04-29 08:22:39 +02:00
fullRepairInterval = mkOption {
type = types.nullOr types.str;
default = "3w";
example = null;
description = mdDoc ''
2021-04-29 08:22:39 +02:00
Set the interval how often full repairs are run, i.e.
{command}`nodetool repair --full` is executed. See
<https://cassandra.apache.org/doc/latest/operating/repair.html>
2021-04-29 08:22:39 +02:00
for more information.
Set to `null` to disable full repairs.
2021-04-29 08:22:39 +02:00
'';
};
2021-04-29 08:22:39 +02:00
fullRepairOptions = mkOption {
type = types.listOf types.str;
2021-04-29 08:22:39 +02:00
default = [ ];
example = [ "--partitioner-range" ];
description = mdDoc ''
2021-04-29 08:22:39 +02:00
Options passed through to the full repair command.
'';
};
2021-04-29 08:22:39 +02:00
incrementalRepairInterval = mkOption {
type = types.nullOr types.str;
default = "3d";
example = null;
description = mdDoc ''
2021-04-29 08:22:39 +02:00
Set the interval how often incremental repairs are run, i.e.
{command}`nodetool repair` is executed. See
<https://cassandra.apache.org/doc/latest/operating/repair.html>
2021-04-29 08:22:39 +02:00
for more information.
Set to `null` to disable incremental repairs.
2021-04-29 08:22:39 +02:00
'';
};
2021-04-29 08:22:39 +02:00
incrementalRepairOptions = mkOption {
type = types.listOf types.str;
2021-04-29 08:22:39 +02:00
default = [ ];
example = [ "--partitioner-range" ];
description = mdDoc ''
2021-04-29 08:22:39 +02:00
Options passed through to the incremental repair command.
'';
};
2021-04-29 08:22:39 +02:00
maxHeapSize = mkOption {
type = types.nullOr types.str;
default = null;
example = "4G";
description = mdDoc ''
Must be left blank or set together with {option}`heapNewSize`.
If left blank a sensible value for the available amount of RAM and CPU
cores is calculated.
Override to set the amount of memory to allocate to the JVM at
start-up. For production use you may wish to adjust this for your
environment. `MAX_HEAP_SIZE` is the total amount of memory dedicated
to the Java heap. `HEAP_NEWSIZE` refers to the size of the young
generation.
The main trade-off for the young generation is that the larger it
is, the longer GC pause times will be. The shorter it is, the more
expensive GC will be (usually).
'';
};
2021-04-29 08:22:39 +02:00
heapNewSize = mkOption {
type = types.nullOr types.str;
default = null;
example = "800M";
description = mdDoc ''
Must be left blank or set together with {option}`heapNewSize`.
If left blank a sensible value for the available amount of RAM and CPU
cores is calculated.
Override to set the amount of memory to allocate to the JVM at
start-up. For production use you may wish to adjust this for your
environment. `HEAP_NEWSIZE` refers to the size of the young
generation.
The main trade-off for the young generation is that the larger it
is, the longer GC pause times will be. The shorter it is, the more
expensive GC will be (usually).
The example `HEAP_NEWSIZE` assumes a modern 8-core+ machine for decent pause
times. If in doubt, and if you do not particularly want to tweak, go with
100 MB per physical CPU core.
'';
};
2021-04-29 08:22:39 +02:00
mallocArenaMax = mkOption {
type = types.nullOr types.int;
default = null;
example = 4;
description = mdDoc ''
Set this to control the amount of arenas per-thread in glibc.
'';
};
2021-04-29 08:22:39 +02:00
remoteJmx = mkOption {
type = types.bool;
default = false;
description = mdDoc ''
Cassandra ships with JMX accessible *only* from localhost.
To enable remote JMX connections set to true.
Be sure to also enable authentication and/or TLS.
See: <https://wiki.apache.org/cassandra/JmxSecurity>
'';
};
2021-04-29 08:22:39 +02:00
jmxPort = mkOption {
type = types.int;
default = 7199;
description = mdDoc ''
Specifies the default port over which Cassandra will be available for
JMX connections.
For security reasons, you should not expose this port to the internet.
Firewall it if needed.
'';
};
2021-04-29 08:22:39 +02:00
jmxRoles = mkOption {
2021-04-29 08:22:39 +02:00
default = [ ];
description = mdDoc ''
Roles that are allowed to access the JMX (e.g. {command}`nodetool`)
BEWARE: The passwords will be stored world readable in the nix store.
It's recommended to use your own protected file using
{option}`jmxRolesFile`
Doesn't work in versions older than 3.11 because they don't like that
it's world readable.
'';
type = types.listOf (types.submodule {
options = {
username = mkOption {
type = types.str;
description = lib.mdDoc "Username for JMX";
};
password = mkOption {
type = types.str;
description = lib.mdDoc "Password for JMX";
};
};
});
};
2021-04-29 08:22:39 +02:00
jmxRolesFile = mkOption {
type = types.nullOr types.path;
2021-04-29 08:22:39 +02:00
default =
2021-04-29 08:39:35 +02:00
if versionAtLeast cfg.package.version "3.11"
2021-04-29 08:22:39 +02:00
then pkgs.writeText "jmx-roles-file" defaultJmxRolesFile
else null;
defaultText = literalMD ''generated configuration file if version is at least 3.11, otherwise `null`'';
example = "/var/lib/cassandra/jmx.password";
description = lib.mdDoc ''
Specify your own jmx roles file.
Make sure the permissions forbid "others" from reading the file if
you're using Cassandra below version 3.11.
'';
};
};
config = mkIf cfg.enable {
2021-04-29 08:22:39 +02:00
assertions = [
{
assertion = (cfg.listenAddress == null) != (cfg.listenInterface == null);
message = "You have to set either listenAddress or listenInterface";
}
{
assertion = (cfg.rpcAddress == null) != (cfg.rpcInterface == null);
message = "You have to set either rpcAddress or rpcInterface";
}
{
assertion = (cfg.maxHeapSize == null) == (cfg.heapNewSize == null);
message = "If you set either of maxHeapSize or heapNewSize you have to set both";
}
{
assertion = cfg.remoteJmx -> cfg.jmxRolesFile != null;
message = ''
If you want JMX available remotely you need to set a password using
<literal>jmxRoles</literal> or <literal>jmxRolesFile</literal> if
using Cassandra older than v3.11.
'';
}
];
users = mkIf (cfg.user == defaultUser) {
2021-04-29 08:51:40 +02:00
users.${defaultUser} = {
2021-04-29 08:22:39 +02:00
group = cfg.group;
home = cfg.homeDir;
createHome = true;
uid = config.ids.uids.cassandra;
description = "Cassandra service user";
};
2021-04-29 08:51:40 +02:00
groups.${defaultUser}.gid = config.ids.gids.cassandra;
};
2021-04-29 08:22:39 +02:00
systemd.services.cassandra = {
description = "Apache Cassandra service";
after = [ "network.target" ];
environment = {
CASSANDRA_CONF = "${cassandraEtc}";
JVM_OPTS = builtins.concatStringsSep " " fullJvmOptions;
MAX_HEAP_SIZE = toString cfg.maxHeapSize;
HEAP_NEWSIZE = toString cfg.heapNewSize;
MALLOC_ARENA_MAX = toString cfg.mallocArenaMax;
LOCAL_JMX = if cfg.remoteJmx then "no" else "yes";
JMX_PORT = toString cfg.jmxPort;
};
wantedBy = [ "multi-user.target" ];
serviceConfig = {
User = cfg.user;
Group = cfg.group;
ExecStart = "${cfg.package}/bin/cassandra -f";
SuccessExitStatus = 143;
};
2021-04-29 08:22:39 +02:00
};
2021-04-29 08:22:39 +02:00
systemd.services.cassandra-full-repair = {
description = "Perform a full repair on this Cassandra node";
after = [ "cassandra.service" ];
requires = [ "cassandra.service" ];
serviceConfig = {
User = cfg.user;
Group = cfg.group;
ExecStart =
2021-04-29 08:39:35 +02:00
concatStringsSep " "
2021-04-29 08:22:39 +02:00
([
"${cfg.package}/bin/nodetool"
"repair"
"--full"
] ++ cfg.fullRepairOptions);
};
2021-04-29 08:22:39 +02:00
};
systemd.timers.cassandra-full-repair =
mkIf (cfg.fullRepairInterval != null) {
description = "Schedule full repairs on Cassandra";
wantedBy = [ "timers.target" ];
2021-04-29 08:22:39 +02:00
timerConfig = {
OnBootSec = cfg.fullRepairInterval;
OnUnitActiveSec = cfg.fullRepairInterval;
Persistent = true;
};
};
2021-04-29 08:22:39 +02:00
systemd.services.cassandra-incremental-repair = {
description = "Perform an incremental repair on this cassandra node.";
after = [ "cassandra.service" ];
requires = [ "cassandra.service" ];
serviceConfig = {
User = cfg.user;
Group = cfg.group;
ExecStart =
2021-04-29 08:39:35 +02:00
concatStringsSep " "
2021-04-29 08:22:39 +02:00
([
"${cfg.package}/bin/nodetool"
"repair"
] ++ cfg.incrementalRepairOptions);
};
2021-04-29 08:22:39 +02:00
};
systemd.timers.cassandra-incremental-repair =
mkIf (cfg.incrementalRepairInterval != null) {
description = "Schedule incremental repairs on Cassandra";
wantedBy = [ "timers.target" ];
2021-04-29 08:22:39 +02:00
timerConfig = {
OnBootSec = cfg.incrementalRepairInterval;
OnUnitActiveSec = cfg.incrementalRepairInterval;
Persistent = true;
};
};
};
meta.maintainers = with lib.maintainers; [ roberth ];
}