Before reading the full details of Bruce's configuration options, you will probably want an overview of Bruce's design, which is avaliable here.
Bruce's config file is an XML document that specifies settings for batching, compression, and per-topic message rate limiting. It also specifies a list of initial brokers to try contacting for metadata when Bruce is starting. Below is an example config file. It is well commented, and should be self- explanatory once the reader is familiar with the information provided in the above-mentioned design section.
<?xml version="1.0" encoding="US-ASCII"?>
<!-- example Bruce configuration -->
<bruceConfig>
<batching>
<namedConfigs>
<config name="low_latency">
<!-- Here we define thresholds that trigger completion of a
batch when exceeded. The "value" attribute of any of the
3 elements below can be set to "disable". However, at
least one of them must have its value set to something
other than "disable". -->
<!-- Set 500 milliseconds max batching delay. Here, you must
specify integer values directly rather than using syntax
such as "10k". Specifying "disable" for this setting is
not recommended, since it can cause Bruce to batch a
message indefinitely if not enough other messages arrive
to exceed one of the other two batching thresholds below.
-->
<time value="500" />
<!-- No limit on message count. You can specify a value here
such as "10" or "10k". A value of "10k" is interpreted as
(10 * 1024) messages. -->
<messages value="disable" />
<!-- Somewhat arbitrary upper bound on batch data size. As
above, "256k" is interpreted as (256 * 1024) bytes. A
simple integer value such as "10000" can also be
specified. This value applies to the actual message
content (keys and values). The total size of a batch will
be a bit larger due to header overhead. The size of an
empty message is counted as 1 byte. This prevents
batching an unbounded number of empty messages in the case
where no limit is placed on batching delay or message
count. -->
<bytes value="256k" />
</config>
<config name="default_latency">
<time value="10000" />
<messages value="disable" />
<bytes value="256k" />
</config>
</namedConfigs>
<!-- Somewhat arbitrary upper bound on produce request size. As above,
a value such as "100k" is interpreted as (100 * 1024) bytes. You
can also specify simple integer values such as "100000" directly.
Here, you can not specify "disable". A nonnegative integer value
must be specified. This value applies to the actual message
content (keys and values). The total size of a produce request
will be a bit larger due to header overhead. -->
<produceRequestDataLimit value="1024k" />
<!-- This value should be exactly the same as the message.max.bytes
value in the Kafka broker configuration. A larger value will
cause Kafka to send MessageSizeTooLarge error ACKs to Bruce for
large compressed message sets, which will cause Bruce to discard
them. A smaller value will not cause data loss, but will
unnecessarily restrict the size of a compressed message set. As
above, you can supply a value such as "1024k". Specifying
"disable" here is not permitted. -->
<messageMaxBytes value="1000000" />
<!-- This specifies the configuration for combined topics batching
(where a single batch may contain multiple topics). Setting
"enable" to false is strongly discouraged due to performance
considerations.
-->
<combinedTopics enable="true" config="default_latency" />
<!-- This specifies how batching is handled for topics not specified in
"topicConfigs" below. Allowed values for the "action" attribute
are as follows:
"perTopic": This setting will cause each topic not listed in
"topicConfigs" to be batched individually on a per-topic
basis, with the "config" attribute specifying a batching
configuration from "namedConfigs" above.
"combinedTopics": This setting will cause all topics not
listed in "topicConfigs" to be batched together in mixed
topic batches. In this case, "config" must be either
missing or set to the empty string, and the batching
configuration is determined by the "combinedTopics"
element above. This is the setting that most people will
want.
"disable": This setting will cause batching to be disabled for
all topics not listed in "topicConfigs". In this case,
the "config" attribute is optional and ignored. This
setting is strongly discouraged due to performance
considerations.
-->
<defaultTopic action="combinedTopics" config="" />
<topicConfigs>
<!-- Uncomment and customize the settings in here if you wish to
have batching configurations that differ on a per-topic basis.
As above, allowed settings for the "action" attribute are
"perTopic", "combinedTopics", and "disable". The "disable"
setting is strongly discouraged due to performance
considerations.
<topic name="low_latency_topic_1" action="perTopic"
config="low_latency" />
<topic name="low_latency_topic_2" action="perTopic"
config="low_latency" />
-->
</topicConfigs>
</batching>
<compression>
<namedConfigs>
<!-- Don't bother to compress a message set whose total size is
less than minSize bytes. As above, a value such as "1k" is
interpreted as (1 * 1024) bytes. Here, a value of "disable"
is not recognized, but you can specify "0". The value of 128
below is somewhat arbitrary, and not based on experimental
data. Currently the only allowed values for "type" are
"snappy" and "none".
-->
<config name="snappy_config" type="snappy" minSize="128" />
<!-- "minSize" is ignored (and optional) if type in "none". -->
<config name="no_compression" type="none" />
</namedConfigs>
<!-- This must be an integer value at least 0 and at most 100. If the
compressed size of a message set is greater than this percentage
of the uncompressed size, then Bruce sends the data uncompressed,
so the Kafka brokers don't waste CPU cycles dealing with the
compression. The value below is somewhat arbitrary, and not based
on experimental data. -->
<sizeThresholdPercent value="75" />
<!-- This specifies the compression configuration for all topics not
listed in "topicConfigs" below.
-->
<defaultTopic config="snappy_config" />
<topicConfigs>
<!-- Uncomment and customize the settings in here if you wish to
configure compression on a per-topic basis.
<topic name="no_compression_topic_1" config="no_compression" />
<topic name="no_compression_topic_2" config="no_compression" />
-->
</topicConfigs>
</compression>
<topicRateLimiting>
<namedConfigs>
<!-- This configuration specifies that all messages should be
discarded. -->
<config name="zero" interval="1" maxCount="0" />
<!-- This configuration specifies no rate limit (i.e. don't discard
any messages regardless of their arrival rate). -->
<config name="infinity" interval="1" maxCount="unlimited" />
<!-- This configuration specifies a limit of at most 1000 messages
every 10000 milliseconds. Messages that would exceed this
limit are discarded. -->
<config name="config1" interval="10000" maxCount="1000" />
<!-- This configuration specifies a limit of at most (4 * 1024)
messages every 15000 milliseconds. Messages that would exceed
this limit are discarded. -->
<config name="config2" interval="15000" maxCount="4k" />
</namedConfigs>
<!-- This specifies a default configuration for topics not listed in
<topicConfigs> below. Each such topic is rate-limited
individually. In other words, with this configuration, topic
"topic_a" would be allowed 1000 messages every 10000 milliseconds,
and "topic_b" would also be allowed 1000 messages every 10000
milliseconds. -->
<defaultTopic config="config1" />
<topicConfigs>
<!-- Rate limit configurations for individual topics go here. -->
<topic name="topic1" config="zero" />
<topic name="topic2" config="infinity" />
<topic name="topic3" config="config1" />
<topic name="topic4" config="config2" />
</topicConfigs>
</topicRateLimiting>
<initialBrokers>
<!-- When Bruce starts, it chooses a broker in this list to contact for
metadata. If Bruce cannot get metadata from the host it chooses,
it tries other hosts until it succeeds. Once Bruce successfully
gets metadata, the broker list in the metadata determines which
brokers Bruce will connect to for message transmission and future
metadata requests. Specifying a single host is ok, but multiple
hosts are recommended to guard against the case where a single
specified host is down.
-->
<broker host="broker_host_1" port="9092" />
<broker host="broker_host_2" port="9092" />
</initialBrokers>
</bruceConfig>
Bruce's required command line arguments are summarized below:
--config_path PATH
: This specifies the location of the config file.--msg_buffer_max MAX_KB
: This specifies the amount of memory in kbytes Bruce reserves for message data. If this buffer space is exhausted, Bruce starts discarding messages.--receive_socket_name PATH
: This specifies the pathname of Bruce's UNIX domain datagram socket that clients write messages to.
Bruce's optional command line arguments are summarized below:
-h --help
: Display help message.--version
: Display version information and exit.--receive_socket_mode MODE
: This specifies the file permissions for Bruce's UNIX domain datagram socket. Octal values are prefixed with 0. For instance,--receive_socket_mode 0777
specifies unrestricted access.--log_level LEVEL
: This specifies the maximum enabled log level for syslog messages. Allowed values are { LOG_ERR, LOG_WARNING, LOG_NOTICE, LOG_INFO, LOG_DEBUG }. The default value is LOG_NOTICE.--log_echo
: Echo syslog messages to standard error.--protocol_version VERSION
: This specifies the protocol version to use when communicating with Kafka, as specified here. Currently 0 is the only allowed value. Note: There should actually be two separate options: a metadata protocol version and a producer protocol version, since Kafka versions these parts of the protocol separately. Some minor refactoring needs to be done to make Bruce behave in this manner.--status_loopback_only
: This specifies that Bruce's web interface should only be available on the loopback interface.--status_port PORT
: This specifies the port Bruce uses for its web interface. The default value is 9090.--max_input_msg_size N
: This specifies the maximum input message size in bytes expected from clients. Messages larger than this value will be discarded. Here, "size" means the size of the entire datagram. The default value is 65536.--allow_large_unix_datagrams
: Allow large enough values for max_input_msg_size that a client sending a UNIX domain datagram of the maximum allowed size will need to increase its SO_SNDBUF socket option above the default value.--max_failed_delivery_attempts N
: Each time Bruce receives an error ACK causing it to initiate a "pause without discard" or "resend" action as documented here, Bruce increments the failed delivery attempt account for each message in the message set that the ACK applies to. Once a message's failed delivery attempt count exceeds this value, the message is discarded. The default vaule is 5.--daemon
: Causes Bruce to run as a daemon.--client_id ID
: This specifies the client ID string to send in produce requests, as documented here. If unspecified, the client ID will be empty.--required_acks N
: This specifies the requires ACKs value to send in produce requests, as documented here. The default value is -1.--replication_timeout N
: specifies the time in milliseconds the broker will wait for successful replication to occur, as described here, before returning an error. The default value is 10000.--shutdown_max_delay N
: This specifies the maximum time in milliseconds Bruce will spend trying to send queued messages and receive ACKs before shutting down once it receives a shutdown signal. If the time limit expires and Bruce still has queued messages, they will be discarded. The recommended way to shut down Bruce is to stop all clients and let Bruce empty its queues before sending Bruce a shutdown signal. The default value is 30000.--dispatcher_restart_max_delay N
: This specifies the maximum allowed delay in milliseconds for dispatcher shutdown during a pause or metadata update event. During this time period, each dispatcher thread will attempt to finish sending any message currently being sent before shutting down. Each dispatcher thread will also attempt to drain its queue of sent produce requests waiting for responses from Kafka. Any unsent messages in remaining in a dispatcher thread's queue after it shuts down will be sent after the dispatcher is restarted with new metadata. The messages contained in any remaining sent produce requests in a dispatcher thread's "ACK waiting" queue on dispatcher shutdown will be resent after the dispatcher is restarted with new metadata. These messages will be reported as possible duplicates in Bruce's discard reporting interface. The default value is 5000.--metadata_refresh_interval N
: This specifies Bruce's metadata refresh interval in minutes. The actual interval will vary somewhat due to added randomness. This will spread out the metadata requests of multiple Bruce instances to prevent them from all requesting metadata at the same time. The default value is 15.--kafka_socket_timeout N
: This specifies the socket timeout in seconds that Bruce uses when communicating with the Kafka brokers. The default value is 60.--min_pause_delay N
: This specifies a lower bound on the initial time period in milliseconds Bruce will wait before sending a metadata request in response to a pause event or retrying a failed metadata request. The default value is 5000. See also --pause_rate_limit_initial and --pause_rate_limit_max_double below.--pause_rate_limit_initial N
: This specifies an initial delay in milliseconds that Bruce will wait before sending a metadata request in response to a pause event or retrying a failed metadata request. The --min_pause_delay option described above places a lower bound on the actual value, which has a bit of randomness added to it. The default value is 5000.--pause_rate_limit_max_double N
: On repeated failed metadata requests or errors that cause pause events, the delay before requesting metadata (as specified by --pause_rate_limit_initial) is doubled each time up to a maximum number of times specified here. The actual delay has some randomness added to it. The default value is 4.--discard_report_interval N
: This specifies the discard report interval in seconds. The default value is 600.--no_log_discard
: This prevents Bruce from writing syslog messages when discards occur. Discards will still be reported through Bruce's web interface.--debug_dir DIR
: This specifies a directory for debug instrumentation files, as described here. If unspecified, the debug instrumentation file option is disabled.--msg_debug_time_limit N
: This specifies a message debugging time limit in seconds, as described here. The default value is 3600.--msg_debug_byte_limit N
: This specifies a message debugging byte limit, as described here. The default value is (2 * 1024 8 1024 * 1024).--skip_compare_metadata_on_refresh
: On metadata refresh, don't compare new metadata to old metadata. Always replace the metadata even if it is unchanged. This should be disabled for normal operation, but enabling it may be useful for testing.--discard_log_path PATH
: Absolute pathname of local file where discards will be logged. This is intended for debugging. If unspecified, logging of discards to a file will be disabled.--discard_log_bad_msg_prefix_size N
: Maximum bad message prefix size in bytes to write to discard logfile when discarding. The default value is 256.--discard_log_max_file_size N
: Maximum size (in Kb) of discard logfile. When the next log entry e would exceed the maximum, the logfile (with name f) is renamed to f.N wnere N is the current time in milliseconds since the epoch. Then a new file f is opened, and e is written to f. The default value is 1024. See also --discard_log_max_archive_size.--discard_log_max_archive_size N
: See description of --discard_log_max_file_size. Once a discard logfile is renamed from f to f.N due to the size restriction imposed by discard_log_max_file_size, the directory containing f.N is scanned for all old discard logfiles. If their combined size exceeds discard_log_max_archive_size (specified in Kb), then old logfiles are deleted, starting with the oldest, until their combined size no longer exceeds the maximum. The default value is 8192.--discard_report_bad_msg_prefix_size N
: Maximum bad message prefix size in bytes to write to discard report available from Bruce's web interface. The default value is 256.--topic_autocreate
: Enable automatic topic creation. For this to work, the brokers must be configured withauto.create.topics.enable=true
.--omit_timestamp
: Do not use this option, since it will soon be removed. Its purpose is to provide compatibility with legacy infrastructure at if(we).--use_old_input_format
: Do not use this option, since it will soon be removed. Its purpose is to provide compatibility with legacy infrastructure at if(we).--use_old_output_format
: Do not use this option, since it will soon be removed. Its purpose is to provide compatibility with legacy infrastructure at if(we).
Now that you are familiar with all of Bruce's configuration options, you may find information on troubleshooting helpful.
detailed_config.md: Copyright 2014 if(we), Inc.
detailed_config.md is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
You should have received a copy of the license along with this work. If not, see http://creativecommons.org/licenses/by-sa/4.0/.