...
This page is meant as a template for writing a KIP. To create a KIP choose Tools->Copy on this page and modify with your content and replace the heading with the next KIP number and a description of your issue. Replace anything in italics with your own description.
Status
Current state: Under Discussion Rejected [One of "Under Discussion", "Accepted", "Rejected"]
Discussion thread: here
JIRA:
Jira | ||||||
---|---|---|---|---|---|---|
|
Planned release: 1.1.0
Please keep the discussion on the mailing list rather than commenting on the wiki (wiki discussions get unwieldy fast).
...
The options accepted by kafka-configs.sh
command will change:
--zookeeper
will be deprecated, which means it will display a warning message saying that it's ignored.- a new
--bootstrap-server
option will be added - a new
--config.properties
option will be added
Protocol Changes
KIP-133 introduced the describe and alter admin protocols and KIP-140 a wire format representation for ResourceType. We will modify these to accommodate the new requirements.
Wire Format Types
ResourceType
0: Unknown
1: Any
2: Topic
3: Group
4: Broker
5: User (new)
6: Client (new)
List Quotas
To create a robust API for the admin client we should be able to list quotas for clients and users. This is needed to be able to create finer grained authorization if needed (for instance listing would be more restrictive than simply describing) and to provide a query functionality so that multiple quota configs could be described or altered in a batch.
Code Block | ||||
---|---|---|---|---|
| ||||
ListQuotas Request (Version: 1) => quota_config_resource
quota_config_resource => type name
type => INT8
name => NULLABLE_STRING |
Request semantics:
- Can be sent to any broker
- The type of
quota_config_resource
can either be "Client" or "User". - The
name
ofquota_config_resource
is nullable.- If
name
is not provided, we list the given type of resources. - If
name
is provided, we list the children of the given user or client resource.
- If
- Authorization: "ListQuotas" can only be interpreted on the "Cluster" resource and represented by the Describe ACL due to the similarity with other listing protocols like ListOffsets or ListGroups. Unauthorized requests will receive an appropriate AuthorizationFailed error code.
Code Block | ||||
---|---|---|---|---|
| ||||
ListQuotas Response (Version: 1) => throttle_time_ms error_code error_message [quota_config_entity_name]
throttle_time_ms => INT32
error_code => INT16
error_message => NULLABLE_STRING
quota_config_entity_name => STRING |
Response semantics:
- Only the quota config names will be returned because the request explicitly specifies the config's type information ("User" and/or "Client") therefore there is no need to specify it here.
- The error returned means that the requestor doesn't have access to the specified resource.
Describe Quotas
To be able to implement the use cases of kafka-configs.sh where a quota is modified, like user, client or (user,client) we have to create a protocol to handle quota listings. The justification for a new protocol is that a quota is quite different from a broker or topic config because a quota can sometimes be identified a simple user, client or even a (user,client) tuple while a topic or a broker config can be identified only by the topic's name or the broker's ID.
- removed, it's functionality won't be available. The design decision behind is that the ConfigCommand tool will be rewritten in the tools module which doesn't depend on the core module. This makes it hard to provide backward compatibility with the current ConfigCommand.
--bootstrap-server
was added in
and will be used here.Jira server ASF JIRA columns key,summary,type,created,updated,due,assignee,reporter,priority,status,resolution serverId 5aa69414-a9e9-3523-82ec-879b028fb15b key KAFKA-6494 --adminclient.config
option will be added and should be used similarly to other tools, such as --producer.config in the console-producer. This parses a config file and initializes the admin client used internally.--adminclient-property
option will be added. It is a key=value list that will be parsed by the command. It initializes the internal admin client.
A new tool, called scram-credentials.sh will be added. The need for this broker is when people use zookeeper as a credentials store for SCRAM (and currently users have no other option), then direct interaction with zookeeper is required to set up the initial credentials with inter-broker communication. The functionality of the tool will cover the following:
- Add, remove and list SCRAM credentials directly with zookeeper
- Will continue to use the
--zookeeper
option to specify the zookeeper host - Works similarly to the old config command. Examples:
- Add new credentials:
bin/scram-credentials.sh --zookeeper localhost:2181 --add 'SCRAM-SHA-256=[iterations=8192,password=alice-secret],SCRAM-SHA-512=[password=alice-secret]' --username alice
- Describe credentials:
bin/scram-credentials.sh --zookeeper localhost:2181 --describe --username alice
- Remove credentials:
bin/scram-credentials.sh --zookeeper localhost:2181 --delete 'SCRAM-SHA-512' --username alice
- Add new credentials:
Protocol Changes
KIP-133 introduced the describe and alter admin protocols and KIP-140 a wire format representation for ResourceType. We will modify these to accommodate the new requirements.
Wire Format Types
ResourceType
0: Unknown
1: Any
2: Topic
3: Group
4: Broker
5: User (new)
6: Client (new)
QuotaType (new)
0: ProducerByteRate
1: ConsumerByteRate
2: RequestPercentage
QuotaSource (new)
0: ClientInUser
1: DefaultClientInUser
2: User
3: ClientInDefaultUser
4: DefaultClientInDefaultUser
5: DefaultUser
6: Client
7: DefaultClient
Field Types
Double
A new type needs to be added to transfer quota values. Since the protocol classes in Kafka already uses ByteBuffers it is logical to use their functionality for serializing doubles. The serialization is basically a representation of the specified floating-point value according to the IEEE 754 floating-point "double format" bit layout. The ByteBuffer serializer writes eight bytes containing the given double value, in Big Endian byte order, into this buffer at the current position, and then increments the position by eight.
The implementation will be defined in org.apache.kafka.common.protocol.types
with the other protocol types and it will have no default value much like the other types available in the protocol.
Describe Quotas
The justification for a new protocol is that a quota is quite different from a broker or topic config because a quota can sometimes be identified a simple user, client or even a (user,client) tuple while a topic or a broker config can be identified only by the topic's name or the broker's ID. Moreover quotas have their own well defined types.
Code Block | ||||
---|---|---|---|---|
| ||||
DescribeQuotas Request (Version: 0) => [resource]
resource => [quota_resource] [quota_type]
quota | ||||
Code Block | ||||
| ||||
DescribeQuotas Request (Version: 1) => [resource] resource => quota_config_resource child_quota_config_resource [config_name] quota_config_resource => type name type => INT8 name => NULLABLE_STRING child_quota_config_resource => type name type => INT8 name => NULLABLE_STRING configquota_nametype => STRINGINT8 |
Request semantics:
- Can be sent to any broker
- If the
name
is <default> empty it means that listing the default quota is asked. Responses will be returned the same way for defaults. - If the configthe
quota_nametype
array isnull
is empty, all configs quotas are returned. Otherwise, configs quotas with the provided names types are returned. - Authorization: "DescribeQuotas" can only be interpreted on the "Cluster" resource and represented by the DescribeConfigs ACL due to the similarity in use cases. Unauthorized requests will receive an appropriate AuthorizationFailed error code.
...
Code Block | ||||
---|---|---|---|---|
| ||||
DescribeQuotas Response (Version: 10) => throttle_time_ms [resource] throttle_time_ms => INT32 resource => [quota_config_resource] child_quota_config_resource [config[quota] quota_config_resource => type name type => INT8 name => NULLABLE_STRING child_quota_config_resourcecollection => type name type => INT8 name => NULLABLE_STRING config => error_code error_message [configerror_code error_message [quota_entry] error_code => INT16 error_message => NULLABLE_STRING configquota_entry => quota_type config_name => STRINGquota_value quota_source configquota_valuetype => STRINGINT8 readquota_onlyvalue => BOOLEANDOUBLE isquota_defaultsource => BOOLEAN is_sensitive => BOOLEANINT8 |
Alter Quotas
Code Block | ||||
---|---|---|---|---|
| ||||
AlterQuota Request (Version: 0) => validate_only [resource] validate_only => BOOLEAN resource => [quota_config_resource] child_quota_config_resource [config[quota] quota_config_resource => type name type => INT8 name => NULLABLE_STRING child_quota_config_resource => quota_type namequota_value quota_type => INT8 name => NULLABLE_STRING config => config_name config_value config_name => STRING config_value => STRING |
Request Semantics
quota_value => DOUBLE |
Request Semantics
- Can be sent to any broker
- If
name
is empty it - Can be sent to any broker
- If
name
is <default>
it means that altering a default quota is asked. - If an
Alter
operation is attempted on a read-only config, anInvalidRequestException
error will be returned for the relevant resource. - Authorization: "AlterQuotas" can only be interpreted on the "Cluster" resource and represented by the AlterConfigs ACL due to the similarity in use cases. Unauthorized requests will receive an appropriate AuthorizationFailed error code.
- For tools that allow users to alter quota configs, a validation/dry-run mode where validation errors are reported but no creation is attempted is available via the
validate_only
parameter.
- The AlterQuotas protocol has an incremental semantics. By this we mean that the request will update only those quotas which are sent in the request.
- Removing quotas will be done by sending a NaN as the value.
Code Block | ||||
---|---|---|---|---|
| ||||
AlterQuotas | ||||
Code Block | ||||
| ||||
AlterQuotas Response (Version: 0) => throttle_time_ms [resource] throttle_time_ms => INT32 resource => quota_config_resource child_[quota_config_resource] [configquota] quota_config_resource => type name type => INT8 name => NULLABLE_STRING child_quota_config_resource => type name type => INT8 error_code error_message quota_type name => NULLABLE_STRING config => error_code => INT16 error_message => NULLABLE_STRING configquota_nametype => STRINGINT8 |
AdminClient APIs
Code Block | ||||
---|---|---|---|---|
| ||||
public class AdminClient {
public DescribeQuotasResult describeQuotas(String userId, String clientId, Collection<String> configs, final DescribeQuotasOptions options);
public AlterQuotasResult alterQuotas(Map<QuotaEntityTuple, Config> configs, AlterQuotasOptions options);
}
public class DescribeQuotasOptions {
public DescribeQuotasOptions timeoutMs(Integer timeout);
}
public class DescribeQuotasResult {
public Map<QuotaConfigResourceTuple, KafkaFuture<Config>> values();
}
public class AlterQuotasOptions {
public AlterQuotasOptions timeoutMs(Integer timeout);
public AlterQuotasOptions validateOnly(boolean validateOnly);
}
public class AlterQuotasResult {
public Map<QuotaConfigResourceTuple, KafkaFuture<Void>> results();
} |
Request API
Code Block | ||||
---|---|---|---|---|
| ||||
public class DescribeQuotasRequest extends AbstractRequest {
public static Schema[] schemaVersions();
public static DescribeQuotasRequest parse(ByteBuffer buffer, short version);
public static class Builder extends AbstractRequest.Builder {
public Builder(Map<QuotaConfigResourceTuple, Collection<String>> quotaConfigSettings);
public DescribeQuotasRequest build(short version);
}
public DescribeQuotasRequest(short version, Map<QuotaConfigResourceTuple, Collection<String>> quotaConfigSettings);
public DescribeQuotasRequest(Struct struct, short version);
public Map<QuotaConfigResourceTuple, Collection<String>> quotaConfigSettings();
}
public class DescribeQuotasResponse extends AbstractResponse {
public static Schema[] schemaVersions();
public DescribeQuotasResponse(int throttleTimeMs, Map<QuotaConfigResourceTuple, DescribeConfigsResponse.Config> configs);
public DescribeQuotasResponse(Struct struct);
public Map<QuotaConfigResourceTuple, DescribeConfigsResponse.Config> quotaConfigSettings();
}
/**
* Stores a child and its parent resource. This is used for representing a (user, client) resource tuple where
* the user part considered to be the parent resource and the client is the child.
*/
public final class QuotaConfigResourceTuple {
private final Resource quotaConfigResource;
private final Resource childQuotaConfigResource;
public Resource quotaConfigResource();
public Resource childQuotaConfigResource();
public QuotaConfigResourceTuple(Resource quotaConfigResource);
public QuotaConfigResourceTuple(Resource quotaConfigResource, Resource childQuotaConfigResource);
}
|
New Command Line Interface
The kafka-config.sh command line interface will change a little bit in terms of help message and response format as we will use argparse4j for parsing arguments.
Help Message
AlterConfigs
This request needs some change as currently the --add-config operation of ConfigCommand would do incremental operations in Zookeeper but the AlterConfigs protocol sets the whole properties object. The purpose of this change to add an boolean parameter to the request so that it can specify the behavior (incremental or set) which needs to be executed.
Code Block | ||||
---|---|---|---|---|
| ||||
AlterConfigs Request (Version: 1) => [resources] validate_only incremental_update
validate_only => BOOLEAN
incremental_update => BOOLEAN // new addition
resources => resource_type resource_name [configs]
resource_type => INT8
resource_name => STRING
configs => config_name config_value
config_name => STRING
config_value => STRING |
Request Semantics:
- The default value of
incremental_update
is false. That means that the request will wipe the node's data and sets what is sent in the request. - Setting the
incremental_update
flag to true makes sure that existing configs are not deleted. - Deleting a config in incremental mode is done by sending an empty string as value.
- Other existing semantics aren't changed.
AdminClient APIs
Code Block | ||||
---|---|---|---|---|
| ||||
public static class Quota {
public QuotaType type();
public double value();
public QuotaSource source();
}
public enum QuotaType {
PRODUCER_BYTE_RATE((byte) 0, "producer_byte_rate"),
CONSUMER_BYTE_RATE((byte) 1, "consumer_byte_rate"),
REQUEST_PERCENTAGE((byte) 2, "request_percentage");
QuotaType(byte id, String name);
public byte id();
public String quotaName();
}
public enum QuotaSource {
CLIENT_OF_USER((byte) 0, "Client of user"),
DEFAULT_CLIENT_OF_USER((byte) 1, "Default client of user"),
USER((byte) 2, "User"),
CLIENT_OF_DEFAULT_USER((byte) 3, "Client of default user"),
DEFAULT_CLIENT_OF_DEFAULT_USER((byte) 4, "Default client of default user"),
DEFAULT_USER((byte) 5, "Default user"), CLIENT((byte) 6, "Client"),
DEFAULT_CLIENT((byte) 7, "Default client");
QuotaSource(byte id, String description);
public byte id();
public String description();
}
/**
* Makes sure that the list of resources that is used as key in a hashmap is immutable and has a fixed implementation for the hashCode.
*/
public class ConfigResourceList {
public List<ConfigResource> getResourceList();
public class AdminClient {
public DescribeQuotasResult describeQuotas(Map<ConfigResourceList, Collection<QuotaType>>, DescribeQuotasOptions options);
public AlterQuotasResult alterQuotas(Map<ConfigResourceList, Collection<Quota>> configs, AlterQuotasOptions options);
}
public class DescribeQuotasOptions extends AbstractOptions<DescribeQuotasOptions> {
public DescribeQuotasOptions timeoutMs(Integer timeout);
}
public class DescribeQuotasResult {
public Map<List<Resource>, KafkaFuture<Collection<Quota>>> values();
}
public class AlterQuotasOptions extends AbstractOptions<AlterQuotasOptions> {
public AlterQuotasOptions timeoutMs(Integer timeout);
public AlterQuotasOptions validateOnly(boolean validateOnly);
}
public class AlterQuotasResult {
public Map<List<Resource>, KafkaFuture<Void>> results();
}
public class AlterConfigsOptions extends AbstractOptions<AlterConfigsOptions> {
public AlterConfigsOptions timeoutMs(Integer timeoutMs);
public AlterConfigsOptions validateOnly(boolean validateOnly);
public boolean shouldValidateOnly();
public AlterConfigsOptions incrementalUpdate(boolean incrementalUpdate); // new
public boolean shouldUpdateIncrementally(); // new
} |
Request API
Code Block | ||||
---|---|---|---|---|
| ||||
public class QuotaCollection {
public QuotaCollection(ApiError error, Collection<Quota> entries);
public QuotaCollection(Collection<Quota> entries);
public ApiError error();
public Collection<Quota> entries();
}
public class DescribeQuotasRequest extends AbstractRequest {
public static Schema[] schemaVersions();
public static DescribeQuotasRequest parse(ByteBuffer buffer, short version);
public static class Builder extends AbstractRequest.Builder {
public Builder(Map<List<Resource>, Collection<QuotaType>> quotaSettings);
public DescribeQuotasRequest build(short version);
}
public DescribeQuotasRequest(short version, Map<List<Resource>, Collection<QuotaType>> quotaSettings);
public DescribeQuotasRequest(Struct struct, short version);
public Map<List<Resource>, Collection<QuotaType>> quotaTypes();
}
public class DescribeQuotasResponse extends AbstractResponse {
public static Schema[] schemaVersions();
public DescribeQuotasResponse(int throttleTimeMs, Map<ConfigResourceList, KafkaFuture<Collection<Quota>>>);
public DescribeQuotasResponse(Struct struct);
public Map<List<Resource>, QuotaCollection> quotas();
}
public class AlterQuotasRequest extends AbstractRequest {
public static Schema[] schemaVersions();
public static class Builder extends AbstractRequest.Builder {
public Builder(Map<List<Resource>, QuotaCollection> quotaSettings);
public DescribeQuotasRequest build(short version);
}
public AlterQuotasRequest(short version, Map<List<Resource>, QuotaCollection> quotas, boolean validateOnly);
public AlterQuotasRequest(Struct struct, short version);
public Map<List<Resource>, QuotaCollection> quotas();
}
public class AlterQuotasResponse extends AbstractResponse {
public static Schema[] schemaVersions();
public AlterQuotasRequest(short version, Map<List<Resource>, ApiError> quotas, boolean validateOnly);
public AlterQuotasRequest(Struct struct, short version);
public Map<List<Resource>, ApiError> errors();
public int throttleTimeMs();
}
|
New Command Line Interface
The kafka-config.sh
command line interface will change a little bit in terms of help message and response format as we will use argparse4j for parsing arguments.
Help Message
No Format |
---|
usage: config-command [-h] --entity-type {topics,clients,users,brokers}
[--force FORCE] [--add-config ADDCONFIG]
[--delete-config DELETECONFIG]
(--entity-name ENTITYNAME | --entity-default)
(--describe | --alter)
(--bootstrap-server BOOTSTRAPSERVERS |
--adminclient.config CONFIGPROPERTIES |
--adminclient-property ADMINCLIENTPROPERTY)
Change configs for topics, clients, users, brokers dynamically.
optional arguments:
-h, --help show this help message and exit
--entity-type {topics,clients,users,brokers}
REQUIRED: the type of entity
(topics/clients/users/brokers)
--force FORCE Suppresses console prompts
--add-config ADDCONFIG
|
No Format |
usage: config-command [-h] --entity-type {topics,clients,users,brokers} [--force FORCE] [--add-config ADDCONFIG] Key Value pairs of configs to add. Square [--delete-config DELETECONFIG] brackets can be used to group values which (--entity-name ENTITYNAME | --entity-default) contain commas: 'k1=v1,k2=[v1,v2,v2],k3=v3'. (---describe | --alter) delete-config DELETECONFIG Config keys to remove in (--bootstrap-servers BOOTSTRAPSERVERS |the following form: 'k1, --config.properties CONFIGPROPERTIES) Change configs for topics, clients, users, brokers dynamically. optional arguments: -h, --help k2'. You can specify only one in --entity-name and --entity-default --entity-name ENTITYNAME show this help message and exit --entity-type {topics,clients,users,brokers} Name of entity (client id/user principal name) --entity-default Default entity name for clients/users (applies REQUIRED:to the type of corresponding entity type in command line) You can specify only one in --alter, --describe --describe (topics/clients/users/brokers) --force FORCEList configs for the given Suppressesentity. console prompts(default: --add-config ADDCONFIG false) --alter Key Value pairs of Alter configsthe configuration for to the addentity. Square(default: bracketsfalse) canREQUIRED. You becan specify usedonly one to group values whichin --bootstrap-servers, --adminclient.config --bootstrap-server BOOTSTRAPSERVER containThe commas: 'k1=v1,k2=[v1,v2,v2],k3=v3'. --delete-config DELETECONFIG broker list string in the form Config keys to remove in the following formHOST1: 'k1, PORT1,HOST2:PORT2. --command-config COMMANDCONFIG k2'. You can specify only one in --entity-name and --entity-default --entity-name ENTITYNAME The config properties file for the Name of entity (client id/user principal name) --entity-default Admin Client. Process Defaultfinished entitywith nameexit for clients/users (applies to code 0 |
Output Format
No Format |
---|
corresponding entity type in command line) You can specify only one in --alter, --describe --describe CONFIGS FOR TOPIC topicA List configs for the given entity. (default: Name false) Value --alter Alter the configuration for the entity. (default: Sensitive Read-only false) YouSource can specify only one in --bootstrap-servers, --config.properties --bootstrap-servers BOOTSTRAPSERVERS compression.type = producer REQUIRED: The broker list string in the form false false HOST1:PORT1,HOST2:PORT2. --config.properties CONFIGPROPERTIES Default config message.format.version = 1.0-IV0 REQUIRED: The config properties file for the false false Admin Client. Process finished withDefault exitconfig code 0 |
Output Format
No Format |
---|
file.delete.delay.ms = 6000 CONFIGS FOR TOPIC topicA false false Dynamic topic config leader.replication.throttled.replicas = Name Value false false Sensitive Read-only Default config compression.typemax.message.bytes = producer1000012 false false true Default config message.format.version min.compaction.lag.ms = 1.0-IV00 false false true Default config filemessage.deletetimestamp.delay.mstype = 60000CreateTime false false false Default config true leader.replication.throttled.replicas = min.insync.replicas = 1 false false true false false Default config max.message.bytes = 1000012 segment.jitter.ms = 0 false false true false min.compaction.lag.ms = 0false Default config preallocate = false false true message.timestamp.type = CreateTime false false Default config false false true index.interval.bytes = 4096 min.insync.replicas = 1 false false Default config false false min.cleanable.dirty.ratio = 0.5 true segment.jitter.ms = 0 false false Default config unclean.leader.election.enable = false false false true false false preallocate = false Default config retention.bytes = 10 false false true index.interval.bytes = 4096 false false Dynamic topic config falsedelete.retention.ms = 86400000 false true min.cleanable.dirty.ratio = 0.5 false false Default config false false true cleanup.policy = delete unclean.leader.election.enable = false false false false Default config false true retentionflush.bytesms = 10 9223372036854775807 false false falseDefault config follower.replication.throttled.replicas = false false delete.retention.ms = 86400000 false false Default config false false true segment.bytes = 1073741824 cleanup.policy = delete false false Default config false false true retention.ms = 604800000 flush.ms = 9223372036854775807 false false false false Default config true follower.replication.throttled.replicas = segment.ms = 604800000 false false true false false Default config segment.bytesmessage.timestamp.difference.max.ms = 10737418249223372036854775807 false false false Default falseconfig true flush.messages = 9223372036854775807 retention.ms = 604800000 false false Default config false false true segment.index.bytes = 10485760 segment.ms = 604800000 false false Default config false false true messageproducer.timestamp.difference.max.ms = 9223372036854775807byte.rate = 1000 false false true false false flush.messages = 9223372036854775807 false false true segment.index.bytes = 10485760 false false true |
As seen above, the describe format becomes more organized and it will also return default properties (as the protocol currently supports that). In case of alter we will also do an extra describe after executing the alter and print the most fresh state.
Compatibility, Deprecation, And Migration Plan
Compatibility
The behavior of the --zookeeper
command line parameter will change. After this change it will print a warning message saying its ignored. Therefore every user will need to change --zookeeper
to --bootstrap-servers
.
Default user
|
As seen above, the describe format becomes more organized and it will also return default properties (as the protocol currently supports that). In case of alter we will also do an extra describe after executing the alter and print the most fresh state.
Compatibility, Deprecation, And Migration Plan
Compatibility
Firstly, the --zookeeper option will be removed from kafka-configs.sh and the backing code will be replaced. Therefore every user will need to change --zookeeper
to --bootstrap-servers
, --adminclient-property
or --adminclient.config
. SCRAM update will be done by the newly introduced scram-credentials.sh
tool. Other existing behavior will be kept.
Secondly, users as of this KIP would be able to describe all topics or brokers in one step but can't do it for clients and users. For those who have this use case would still need to use the old command for a while (see below). The reason for this change is currently MetadataRequest provides enough information about topics and brokers so it's possible to describe all of them in one step but there's no such information about clients and users.
Finally, backward compatibilty (for instance a 2.0 Also backward compatibilty (saying an 1.1 client wants to admin a 1.0 server) will be impacted as some of the protocols are newly created and doesn't exist in old servers. In these cases users should continue using the scala version of the ConfigCommand by putting the core jar on their classpath and defining the USE_OLD_COMMAND=true
environment variable. This variable will set the main class to the old command in the config and invokes that. This way the environment variable ensures that users who aren't able to use the new command currently would need to make minimal changes in order to continue using it. the protocols are newly created and doesn't exist in old servers. In these cases users should continue using the scala version of the ConfigCommand by putting the core jar on their classpath.
The old Alternatively though the command could be launched through kafka-run-class.sh like this:
...
From the compatibility point of view there might be a bigger impact as mentioned above. Since the command now uses the wire protocols (including some newly introduced ones) the backward compatibility will be impacted. That means that a user can't use a 12.1 0 client to administrate a 1.0 broker as in the older broker some of the wire protocols don't exist. This again should be acceptable most of the users as most of the admin commands require the core jar on the classpath which means that most of the time the commands are executed from an environment with the same version of the brokers. In the remaining cases users will have to change to use kafka-run-class or the USE_OLD_COMMAND
environment variableof the brokers. Therefore the old tool should still be available.
Deprecation
kafka.admin.ConfigCommand
will print a warning message saying it is deprecated and will be removed in version 2.0.To ease the migration for users who are stuck with this command, the USE_OLD_COMMAND
will be introduceda future version.
Special Migration Tools
There are no tools required.
...
The current --zookeeper
option will be disabled removed with this change as it has minimal impact on the current usersusers.
Listing multiple users' and clients' quotas at once won't be possible. If this is required, users would need to use the old tool.
Test Plan
Most of the functionality can be covered with end-to-end tests. There will be unit tests also to verify the protocol and the broker side logic.
Rejected Alternatives
...
the protocol and the broker side logic.
Future Considerations
At this moment this ConfigCommand can describe all the topics and all the brokers with one command but can't describe all clients or users. The reason for this is that we can gain the required information for topics and brokers by a MetadataRequest but we have no such protocols for getting a list of users or clients.