netty5/resolver-dns/src/main/java/io/netty/resolver/dns/DnsNameResolverContext.java

561 lines
21 KiB
Java
Raw Normal View History

/*
* Copyright 2014 The Netty Project
*
* The Netty Project licenses this file to you under the Apache License,
* version 2.0 (the "License"); you may not use this file except in compliance
* with the License. You may obtain a copy of the License at:
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
* WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
* License for the specific language governing permissions and limitations
* under the License.
*/
package io.netty.resolver.dns;
import io.netty.buffer.ByteBuf;
Revamp DNS codec Motivation: There are various known issues in netty-codec-dns: - Message types are not interfaces, which can make it difficult for a user to implement his/her own message implementation. - Some class names and field names do not match with the terms in the RFC. - The support for decoding a DNS record was limited. A user had to encode and decode by him/herself. - The separation of DnsHeader from DnsMessage was unnecessary, although it is fine conceptually. - Buffer leak caused by DnsMessage was difficult to analyze, because the leak detector tracks down the underlying ByteBuf rather than the DnsMessage itself. - DnsMessage assumes DNS-over-UDP. - To send an EDNS message, a user have to create a new DNS record class instance unnecessarily. Modifications: - Make all message types interfaces and add default implementations - Rename some classes, properties, and constants to match the RFCs - DnsResource -> DnsRecord - DnsType -> DnsRecordType - and many more - Remove DnsClass and use an integer to support EDNS better - Add DnsRecordEncoder/DnsRecordDecoder and their default implementations - DnsRecord does not require RDATA to be ByteBuf anymore. - Add DnsRawRecord as the catch-all record type - Merge DnsHeader into DnsMessage - Make ResourceLeakDetector track AbstractDnsMessage - Remove DnsMessage.sender/recipient properties - Wrap DnsMessage with AddressedEnvelope - Add DatagramDnsQuest and DatagramDnsResponse for ease of use - Rename DnsQueryEncoder to DatagramDnsQueryEncoder - Rename DnsResponseDecoder to DatagramDnsResponseDecoder - Miscellaneous changes - Add StringUtil.TAB Result: - Cleaner APi - Can support DNS-over-TCP more easily in the future - Reduced memory footprint in the default DnsQuery/Response implementations - Better leak tracking for DnsMessages - Possibility to introduce new DnsRecord types in the future and provide full record encoder/decoder implementation. - No unnecessary instantiation for an EDNS pseudo resource record
2015-03-16 07:46:14 +01:00
import io.netty.buffer.ByteBufHolder;
import io.netty.channel.AddressedEnvelope;
import io.netty.channel.socket.InternetProtocolFamily;
import io.netty.handler.codec.CorruptedFrameException;
Revamp DNS codec Motivation: There are various known issues in netty-codec-dns: - Message types are not interfaces, which can make it difficult for a user to implement his/her own message implementation. - Some class names and field names do not match with the terms in the RFC. - The support for decoding a DNS record was limited. A user had to encode and decode by him/herself. - The separation of DnsHeader from DnsMessage was unnecessary, although it is fine conceptually. - Buffer leak caused by DnsMessage was difficult to analyze, because the leak detector tracks down the underlying ByteBuf rather than the DnsMessage itself. - DnsMessage assumes DNS-over-UDP. - To send an EDNS message, a user have to create a new DNS record class instance unnecessarily. Modifications: - Make all message types interfaces and add default implementations - Rename some classes, properties, and constants to match the RFCs - DnsResource -> DnsRecord - DnsType -> DnsRecordType - and many more - Remove DnsClass and use an integer to support EDNS better - Add DnsRecordEncoder/DnsRecordDecoder and their default implementations - DnsRecord does not require RDATA to be ByteBuf anymore. - Add DnsRawRecord as the catch-all record type - Merge DnsHeader into DnsMessage - Make ResourceLeakDetector track AbstractDnsMessage - Remove DnsMessage.sender/recipient properties - Wrap DnsMessage with AddressedEnvelope - Add DatagramDnsQuest and DatagramDnsResponse for ease of use - Rename DnsQueryEncoder to DatagramDnsQueryEncoder - Rename DnsResponseDecoder to DatagramDnsResponseDecoder - Miscellaneous changes - Add StringUtil.TAB Result: - Cleaner APi - Can support DNS-over-TCP more easily in the future - Reduced memory footprint in the default DnsQuery/Response implementations - Better leak tracking for DnsMessages - Possibility to introduce new DnsRecord types in the future and provide full record encoder/decoder implementation. - No unnecessary instantiation for an EDNS pseudo resource record
2015-03-16 07:46:14 +01:00
import io.netty.handler.codec.dns.DefaultDnsQuestion;
import io.netty.handler.codec.dns.DefaultDnsRecordDecoder;
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
import io.netty.handler.codec.dns.DnsResponseCode;
Revamp DNS codec Motivation: There are various known issues in netty-codec-dns: - Message types are not interfaces, which can make it difficult for a user to implement his/her own message implementation. - Some class names and field names do not match with the terms in the RFC. - The support for decoding a DNS record was limited. A user had to encode and decode by him/herself. - The separation of DnsHeader from DnsMessage was unnecessary, although it is fine conceptually. - Buffer leak caused by DnsMessage was difficult to analyze, because the leak detector tracks down the underlying ByteBuf rather than the DnsMessage itself. - DnsMessage assumes DNS-over-UDP. - To send an EDNS message, a user have to create a new DNS record class instance unnecessarily. Modifications: - Make all message types interfaces and add default implementations - Rename some classes, properties, and constants to match the RFCs - DnsResource -> DnsRecord - DnsType -> DnsRecordType - and many more - Remove DnsClass and use an integer to support EDNS better - Add DnsRecordEncoder/DnsRecordDecoder and their default implementations - DnsRecord does not require RDATA to be ByteBuf anymore. - Add DnsRawRecord as the catch-all record type - Merge DnsHeader into DnsMessage - Make ResourceLeakDetector track AbstractDnsMessage - Remove DnsMessage.sender/recipient properties - Wrap DnsMessage with AddressedEnvelope - Add DatagramDnsQuest and DatagramDnsResponse for ease of use - Rename DnsQueryEncoder to DatagramDnsQueryEncoder - Rename DnsResponseDecoder to DatagramDnsResponseDecoder - Miscellaneous changes - Add StringUtil.TAB Result: - Cleaner APi - Can support DNS-over-TCP more easily in the future - Reduced memory footprint in the default DnsQuery/Response implementations - Better leak tracking for DnsMessages - Possibility to introduce new DnsRecord types in the future and provide full record encoder/decoder implementation. - No unnecessary instantiation for an EDNS pseudo resource record
2015-03-16 07:46:14 +01:00
import io.netty.handler.codec.dns.DnsSection;
import io.netty.handler.codec.dns.DnsQuestion;
Revamp DNS codec Motivation: There are various known issues in netty-codec-dns: - Message types are not interfaces, which can make it difficult for a user to implement his/her own message implementation. - Some class names and field names do not match with the terms in the RFC. - The support for decoding a DNS record was limited. A user had to encode and decode by him/herself. - The separation of DnsHeader from DnsMessage was unnecessary, although it is fine conceptually. - Buffer leak caused by DnsMessage was difficult to analyze, because the leak detector tracks down the underlying ByteBuf rather than the DnsMessage itself. - DnsMessage assumes DNS-over-UDP. - To send an EDNS message, a user have to create a new DNS record class instance unnecessarily. Modifications: - Make all message types interfaces and add default implementations - Rename some classes, properties, and constants to match the RFCs - DnsResource -> DnsRecord - DnsType -> DnsRecordType - and many more - Remove DnsClass and use an integer to support EDNS better - Add DnsRecordEncoder/DnsRecordDecoder and their default implementations - DnsRecord does not require RDATA to be ByteBuf anymore. - Add DnsRawRecord as the catch-all record type - Merge DnsHeader into DnsMessage - Make ResourceLeakDetector track AbstractDnsMessage - Remove DnsMessage.sender/recipient properties - Wrap DnsMessage with AddressedEnvelope - Add DatagramDnsQuest and DatagramDnsResponse for ease of use - Rename DnsQueryEncoder to DatagramDnsQueryEncoder - Rename DnsResponseDecoder to DatagramDnsResponseDecoder - Miscellaneous changes - Add StringUtil.TAB Result: - Cleaner APi - Can support DNS-over-TCP more easily in the future - Reduced memory footprint in the default DnsQuery/Response implementations - Better leak tracking for DnsMessages - Possibility to introduce new DnsRecord types in the future and provide full record encoder/decoder implementation. - No unnecessary instantiation for an EDNS pseudo resource record
2015-03-16 07:46:14 +01:00
import io.netty.handler.codec.dns.DnsRawRecord;
import io.netty.handler.codec.dns.DnsRecord;
import io.netty.handler.codec.dns.DnsRecordType;
import io.netty.handler.codec.dns.DnsResponse;
import io.netty.util.ReferenceCountUtil;
import io.netty.util.concurrent.Future;
import io.netty.util.concurrent.FutureListener;
import io.netty.util.concurrent.Promise;
import io.netty.util.internal.StringUtil;
import java.net.Inet4Address;
import java.net.Inet6Address;
import java.net.InetAddress;
import java.net.InetSocketAddress;
import java.net.UnknownHostException;
import java.util.ArrayList;
import java.util.Collections;
import java.util.HashMap;
import java.util.IdentityHashMap;
import java.util.Iterator;
import java.util.List;
import java.util.Locale;
import java.util.Map;
import java.util.Set;
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
abstract class DnsNameResolverContext<T> {
private static final int INADDRSZ4 = 4;
private static final int INADDRSZ6 = 16;
Revamp DNS codec Motivation: There are various known issues in netty-codec-dns: - Message types are not interfaces, which can make it difficult for a user to implement his/her own message implementation. - Some class names and field names do not match with the terms in the RFC. - The support for decoding a DNS record was limited. A user had to encode and decode by him/herself. - The separation of DnsHeader from DnsMessage was unnecessary, although it is fine conceptually. - Buffer leak caused by DnsMessage was difficult to analyze, because the leak detector tracks down the underlying ByteBuf rather than the DnsMessage itself. - DnsMessage assumes DNS-over-UDP. - To send an EDNS message, a user have to create a new DNS record class instance unnecessarily. Modifications: - Make all message types interfaces and add default implementations - Rename some classes, properties, and constants to match the RFCs - DnsResource -> DnsRecord - DnsType -> DnsRecordType - and many more - Remove DnsClass and use an integer to support EDNS better - Add DnsRecordEncoder/DnsRecordDecoder and their default implementations - DnsRecord does not require RDATA to be ByteBuf anymore. - Add DnsRawRecord as the catch-all record type - Merge DnsHeader into DnsMessage - Make ResourceLeakDetector track AbstractDnsMessage - Remove DnsMessage.sender/recipient properties - Wrap DnsMessage with AddressedEnvelope - Add DatagramDnsQuest and DatagramDnsResponse for ease of use - Rename DnsQueryEncoder to DatagramDnsQueryEncoder - Rename DnsResponseDecoder to DatagramDnsResponseDecoder - Miscellaneous changes - Add StringUtil.TAB Result: - Cleaner APi - Can support DNS-over-TCP more easily in the future - Reduced memory footprint in the default DnsQuery/Response implementations - Better leak tracking for DnsMessages - Possibility to introduce new DnsRecord types in the future and provide full record encoder/decoder implementation. - No unnecessary instantiation for an EDNS pseudo resource record
2015-03-16 07:46:14 +01:00
private static final FutureListener<AddressedEnvelope<DnsResponse, InetSocketAddress>> RELEASE_RESPONSE =
new FutureListener<AddressedEnvelope<DnsResponse, InetSocketAddress>>() {
@Override
public void operationComplete(Future<AddressedEnvelope<DnsResponse, InetSocketAddress>> future) {
if (future.isSuccess()) {
future.getNow().release();
}
}
};
private final DnsNameResolver parent;
private final DnsServerAddressStream nameServerAddrs;
private final String hostname;
protected String pristineHostname;
private final DnsCache resolveCache;
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
private final boolean traceEnabled;
private final int maxAllowedQueries;
private final InternetProtocolFamily[] resolveAddressTypes;
private final DnsRecord[] additionals;
Revamp DNS codec Motivation: There are various known issues in netty-codec-dns: - Message types are not interfaces, which can make it difficult for a user to implement his/her own message implementation. - Some class names and field names do not match with the terms in the RFC. - The support for decoding a DNS record was limited. A user had to encode and decode by him/herself. - The separation of DnsHeader from DnsMessage was unnecessary, although it is fine conceptually. - Buffer leak caused by DnsMessage was difficult to analyze, because the leak detector tracks down the underlying ByteBuf rather than the DnsMessage itself. - DnsMessage assumes DNS-over-UDP. - To send an EDNS message, a user have to create a new DNS record class instance unnecessarily. Modifications: - Make all message types interfaces and add default implementations - Rename some classes, properties, and constants to match the RFCs - DnsResource -> DnsRecord - DnsType -> DnsRecordType - and many more - Remove DnsClass and use an integer to support EDNS better - Add DnsRecordEncoder/DnsRecordDecoder and their default implementations - DnsRecord does not require RDATA to be ByteBuf anymore. - Add DnsRawRecord as the catch-all record type - Merge DnsHeader into DnsMessage - Make ResourceLeakDetector track AbstractDnsMessage - Remove DnsMessage.sender/recipient properties - Wrap DnsMessage with AddressedEnvelope - Add DatagramDnsQuest and DatagramDnsResponse for ease of use - Rename DnsQueryEncoder to DatagramDnsQueryEncoder - Rename DnsResponseDecoder to DatagramDnsResponseDecoder - Miscellaneous changes - Add StringUtil.TAB Result: - Cleaner APi - Can support DNS-over-TCP more easily in the future - Reduced memory footprint in the default DnsQuery/Response implementations - Better leak tracking for DnsMessages - Possibility to introduce new DnsRecord types in the future and provide full record encoder/decoder implementation. - No unnecessary instantiation for an EDNS pseudo resource record
2015-03-16 07:46:14 +01:00
private final Set<Future<AddressedEnvelope<DnsResponse, InetSocketAddress>>> queriesInProgress =
Collections.newSetFromMap(
new IdentityHashMap<Future<AddressedEnvelope<DnsResponse, InetSocketAddress>>, Boolean>());
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
private List<DnsCacheEntry> resolvedEntries;
private StringBuilder trace;
private int allowedQueries;
private boolean triedCNAME;
protected DnsNameResolverContext(DnsNameResolver parent,
String hostname,
DnsRecord[] additionals,
DnsCache resolveCache) {
this.parent = parent;
this.hostname = hostname;
this.additionals = additionals;
this.resolveCache = resolveCache;
nameServerAddrs = parent.nameServerAddresses.stream();
maxAllowedQueries = parent.maxQueriesPerResolve();
resolveAddressTypes = parent.resolveAddressTypesUnsafe();
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
traceEnabled = parent.isTraceEnabled();
allowedQueries = maxAllowedQueries;
}
void resolve(Promise<T> promise) {
boolean directSearch = parent.searchDomains().length == 0 || StringUtil.endsWith(hostname, '.');
if (directSearch) {
internalResolve(promise);
} else {
final Promise<T> original = promise;
promise = parent.executor().newPromise();
promise.addListener(new FutureListener<T>() {
int count;
@Override
public void operationComplete(Future<T> future) throws Exception {
if (future.isSuccess()) {
original.trySuccess(future.getNow());
} else if (count < parent.searchDomains().length) {
String searchDomain = parent.searchDomains()[count++];
Promise<T> nextPromise = parent.executor().newPromise();
String nextHostname = hostname + '.' + searchDomain;
DnsNameResolverContext<T> nextContext = newResolverContext(parent,
nextHostname, additionals, resolveCache);
nextContext.pristineHostname = hostname;
nextContext.internalResolve(nextPromise);
nextPromise.addListener(this);
} else {
original.tryFailure(future.cause());
}
}
});
if (parent.ndots() == 0) {
internalResolve(promise);
} else {
int dots = 0;
for (int idx = hostname.length() - 1; idx >= 0; idx--) {
if (hostname.charAt(idx) == '.' && ++dots >= parent.ndots()) {
internalResolve(promise);
return;
}
}
promise.tryFailure(new UnknownHostException(hostname));
}
}
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
}
private void internalResolve(Promise<T> promise) {
InetSocketAddress nameServerAddrToTry = nameServerAddrs.next();
for (DnsRecordType type: parent.resolveRecordTypes()) {
if (!query(hostname, type, nameServerAddrToTry, promise)) {
return;
}
}
}
private void query(InetSocketAddress nameServerAddr, final DnsQuestion question, final Promise<T> promise) {
if (allowedQueries == 0 || promise.isCancelled()) {
tryToFinishResolve(promise);
return;
}
allowedQueries --;
final Future<AddressedEnvelope<DnsResponse, InetSocketAddress>> f = parent.query0(
nameServerAddr, question, additionals,
parent.ch.eventLoop().<AddressedEnvelope<? extends DnsResponse, InetSocketAddress>>newPromise());
queriesInProgress.add(f);
Revamp DNS codec Motivation: There are various known issues in netty-codec-dns: - Message types are not interfaces, which can make it difficult for a user to implement his/her own message implementation. - Some class names and field names do not match with the terms in the RFC. - The support for decoding a DNS record was limited. A user had to encode and decode by him/herself. - The separation of DnsHeader from DnsMessage was unnecessary, although it is fine conceptually. - Buffer leak caused by DnsMessage was difficult to analyze, because the leak detector tracks down the underlying ByteBuf rather than the DnsMessage itself. - DnsMessage assumes DNS-over-UDP. - To send an EDNS message, a user have to create a new DNS record class instance unnecessarily. Modifications: - Make all message types interfaces and add default implementations - Rename some classes, properties, and constants to match the RFCs - DnsResource -> DnsRecord - DnsType -> DnsRecordType - and many more - Remove DnsClass and use an integer to support EDNS better - Add DnsRecordEncoder/DnsRecordDecoder and their default implementations - DnsRecord does not require RDATA to be ByteBuf anymore. - Add DnsRawRecord as the catch-all record type - Merge DnsHeader into DnsMessage - Make ResourceLeakDetector track AbstractDnsMessage - Remove DnsMessage.sender/recipient properties - Wrap DnsMessage with AddressedEnvelope - Add DatagramDnsQuest and DatagramDnsResponse for ease of use - Rename DnsQueryEncoder to DatagramDnsQueryEncoder - Rename DnsResponseDecoder to DatagramDnsResponseDecoder - Miscellaneous changes - Add StringUtil.TAB Result: - Cleaner APi - Can support DNS-over-TCP more easily in the future - Reduced memory footprint in the default DnsQuery/Response implementations - Better leak tracking for DnsMessages - Possibility to introduce new DnsRecord types in the future and provide full record encoder/decoder implementation. - No unnecessary instantiation for an EDNS pseudo resource record
2015-03-16 07:46:14 +01:00
f.addListener(new FutureListener<AddressedEnvelope<DnsResponse, InetSocketAddress>>() {
@Override
Revamp DNS codec Motivation: There are various known issues in netty-codec-dns: - Message types are not interfaces, which can make it difficult for a user to implement his/her own message implementation. - Some class names and field names do not match with the terms in the RFC. - The support for decoding a DNS record was limited. A user had to encode and decode by him/herself. - The separation of DnsHeader from DnsMessage was unnecessary, although it is fine conceptually. - Buffer leak caused by DnsMessage was difficult to analyze, because the leak detector tracks down the underlying ByteBuf rather than the DnsMessage itself. - DnsMessage assumes DNS-over-UDP. - To send an EDNS message, a user have to create a new DNS record class instance unnecessarily. Modifications: - Make all message types interfaces and add default implementations - Rename some classes, properties, and constants to match the RFCs - DnsResource -> DnsRecord - DnsType -> DnsRecordType - and many more - Remove DnsClass and use an integer to support EDNS better - Add DnsRecordEncoder/DnsRecordDecoder and their default implementations - DnsRecord does not require RDATA to be ByteBuf anymore. - Add DnsRawRecord as the catch-all record type - Merge DnsHeader into DnsMessage - Make ResourceLeakDetector track AbstractDnsMessage - Remove DnsMessage.sender/recipient properties - Wrap DnsMessage with AddressedEnvelope - Add DatagramDnsQuest and DatagramDnsResponse for ease of use - Rename DnsQueryEncoder to DatagramDnsQueryEncoder - Rename DnsResponseDecoder to DatagramDnsResponseDecoder - Miscellaneous changes - Add StringUtil.TAB Result: - Cleaner APi - Can support DNS-over-TCP more easily in the future - Reduced memory footprint in the default DnsQuery/Response implementations - Better leak tracking for DnsMessages - Possibility to introduce new DnsRecord types in the future and provide full record encoder/decoder implementation. - No unnecessary instantiation for an EDNS pseudo resource record
2015-03-16 07:46:14 +01:00
public void operationComplete(Future<AddressedEnvelope<DnsResponse, InetSocketAddress>> future) {
queriesInProgress.remove(future);
if (promise.isDone() || future.isCancelled()) {
return;
}
try {
if (future.isSuccess()) {
onResponse(question, future.getNow(), promise);
} else {
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
// Server did not respond or I/O error occurred; try again.
if (traceEnabled) {
addTrace(future.cause());
}
query(nameServerAddrs.next(), question, promise);
}
} finally {
tryToFinishResolve(promise);
}
}
});
}
void onResponse(final DnsQuestion question, AddressedEnvelope<DnsResponse, InetSocketAddress> envelope,
Promise<T> promise) {
try {
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
final DnsResponse res = envelope.content();
final DnsResponseCode code = res.code();
if (code == DnsResponseCode.NOERROR) {
final DnsRecordType type = question.type();
if (type == DnsRecordType.A || type == DnsRecordType.AAAA) {
onResponseAorAAAA(type, question, envelope, promise);
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
} else if (type == DnsRecordType.CNAME) {
onResponseCNAME(question, envelope, promise);
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
}
return;
}
if (traceEnabled) {
addTrace(envelope.sender(),
"response code: " + code + " with " + res.count(DnsSection.ANSWER) + " answer(s) and " +
res.count(DnsSection.AUTHORITY) + " authority resource(s)");
}
// Retry with the next server if the server did not tell us that the domain does not exist.
if (code != DnsResponseCode.NXDOMAIN) {
query(nameServerAddrs.next(), question, promise);
}
} finally {
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
ReferenceCountUtil.safeRelease(envelope);
}
}
Revamp DNS codec Motivation: There are various known issues in netty-codec-dns: - Message types are not interfaces, which can make it difficult for a user to implement his/her own message implementation. - Some class names and field names do not match with the terms in the RFC. - The support for decoding a DNS record was limited. A user had to encode and decode by him/herself. - The separation of DnsHeader from DnsMessage was unnecessary, although it is fine conceptually. - Buffer leak caused by DnsMessage was difficult to analyze, because the leak detector tracks down the underlying ByteBuf rather than the DnsMessage itself. - DnsMessage assumes DNS-over-UDP. - To send an EDNS message, a user have to create a new DNS record class instance unnecessarily. Modifications: - Make all message types interfaces and add default implementations - Rename some classes, properties, and constants to match the RFCs - DnsResource -> DnsRecord - DnsType -> DnsRecordType - and many more - Remove DnsClass and use an integer to support EDNS better - Add DnsRecordEncoder/DnsRecordDecoder and their default implementations - DnsRecord does not require RDATA to be ByteBuf anymore. - Add DnsRawRecord as the catch-all record type - Merge DnsHeader into DnsMessage - Make ResourceLeakDetector track AbstractDnsMessage - Remove DnsMessage.sender/recipient properties - Wrap DnsMessage with AddressedEnvelope - Add DatagramDnsQuest and DatagramDnsResponse for ease of use - Rename DnsQueryEncoder to DatagramDnsQueryEncoder - Rename DnsResponseDecoder to DatagramDnsResponseDecoder - Miscellaneous changes - Add StringUtil.TAB Result: - Cleaner APi - Can support DNS-over-TCP more easily in the future - Reduced memory footprint in the default DnsQuery/Response implementations - Better leak tracking for DnsMessages - Possibility to introduce new DnsRecord types in the future and provide full record encoder/decoder implementation. - No unnecessary instantiation for an EDNS pseudo resource record
2015-03-16 07:46:14 +01:00
private void onResponseAorAAAA(
DnsRecordType qType, DnsQuestion question, AddressedEnvelope<DnsResponse, InetSocketAddress> envelope,
Promise<T> promise) {
Revamp DNS codec Motivation: There are various known issues in netty-codec-dns: - Message types are not interfaces, which can make it difficult for a user to implement his/her own message implementation. - Some class names and field names do not match with the terms in the RFC. - The support for decoding a DNS record was limited. A user had to encode and decode by him/herself. - The separation of DnsHeader from DnsMessage was unnecessary, although it is fine conceptually. - Buffer leak caused by DnsMessage was difficult to analyze, because the leak detector tracks down the underlying ByteBuf rather than the DnsMessage itself. - DnsMessage assumes DNS-over-UDP. - To send an EDNS message, a user have to create a new DNS record class instance unnecessarily. Modifications: - Make all message types interfaces and add default implementations - Rename some classes, properties, and constants to match the RFCs - DnsResource -> DnsRecord - DnsType -> DnsRecordType - and many more - Remove DnsClass and use an integer to support EDNS better - Add DnsRecordEncoder/DnsRecordDecoder and their default implementations - DnsRecord does not require RDATA to be ByteBuf anymore. - Add DnsRawRecord as the catch-all record type - Merge DnsHeader into DnsMessage - Make ResourceLeakDetector track AbstractDnsMessage - Remove DnsMessage.sender/recipient properties - Wrap DnsMessage with AddressedEnvelope - Add DatagramDnsQuest and DatagramDnsResponse for ease of use - Rename DnsQueryEncoder to DatagramDnsQueryEncoder - Rename DnsResponseDecoder to DatagramDnsResponseDecoder - Miscellaneous changes - Add StringUtil.TAB Result: - Cleaner APi - Can support DNS-over-TCP more easily in the future - Reduced memory footprint in the default DnsQuery/Response implementations - Better leak tracking for DnsMessages - Possibility to introduce new DnsRecord types in the future and provide full record encoder/decoder implementation. - No unnecessary instantiation for an EDNS pseudo resource record
2015-03-16 07:46:14 +01:00
// We often get a bunch of CNAMES as well when we asked for A/AAAA.
Revamp DNS codec Motivation: There are various known issues in netty-codec-dns: - Message types are not interfaces, which can make it difficult for a user to implement his/her own message implementation. - Some class names and field names do not match with the terms in the RFC. - The support for decoding a DNS record was limited. A user had to encode and decode by him/herself. - The separation of DnsHeader from DnsMessage was unnecessary, although it is fine conceptually. - Buffer leak caused by DnsMessage was difficult to analyze, because the leak detector tracks down the underlying ByteBuf rather than the DnsMessage itself. - DnsMessage assumes DNS-over-UDP. - To send an EDNS message, a user have to create a new DNS record class instance unnecessarily. Modifications: - Make all message types interfaces and add default implementations - Rename some classes, properties, and constants to match the RFCs - DnsResource -> DnsRecord - DnsType -> DnsRecordType - and many more - Remove DnsClass and use an integer to support EDNS better - Add DnsRecordEncoder/DnsRecordDecoder and their default implementations - DnsRecord does not require RDATA to be ByteBuf anymore. - Add DnsRawRecord as the catch-all record type - Merge DnsHeader into DnsMessage - Make ResourceLeakDetector track AbstractDnsMessage - Remove DnsMessage.sender/recipient properties - Wrap DnsMessage with AddressedEnvelope - Add DatagramDnsQuest and DatagramDnsResponse for ease of use - Rename DnsQueryEncoder to DatagramDnsQueryEncoder - Rename DnsResponseDecoder to DatagramDnsResponseDecoder - Miscellaneous changes - Add StringUtil.TAB Result: - Cleaner APi - Can support DNS-over-TCP more easily in the future - Reduced memory footprint in the default DnsQuery/Response implementations - Better leak tracking for DnsMessages - Possibility to introduce new DnsRecord types in the future and provide full record encoder/decoder implementation. - No unnecessary instantiation for an EDNS pseudo resource record
2015-03-16 07:46:14 +01:00
final DnsResponse response = envelope.content();
final Map<String, String> cnames = buildAliasMap(response);
Revamp DNS codec Motivation: There are various known issues in netty-codec-dns: - Message types are not interfaces, which can make it difficult for a user to implement his/her own message implementation. - Some class names and field names do not match with the terms in the RFC. - The support for decoding a DNS record was limited. A user had to encode and decode by him/herself. - The separation of DnsHeader from DnsMessage was unnecessary, although it is fine conceptually. - Buffer leak caused by DnsMessage was difficult to analyze, because the leak detector tracks down the underlying ByteBuf rather than the DnsMessage itself. - DnsMessage assumes DNS-over-UDP. - To send an EDNS message, a user have to create a new DNS record class instance unnecessarily. Modifications: - Make all message types interfaces and add default implementations - Rename some classes, properties, and constants to match the RFCs - DnsResource -> DnsRecord - DnsType -> DnsRecordType - and many more - Remove DnsClass and use an integer to support EDNS better - Add DnsRecordEncoder/DnsRecordDecoder and their default implementations - DnsRecord does not require RDATA to be ByteBuf anymore. - Add DnsRawRecord as the catch-all record type - Merge DnsHeader into DnsMessage - Make ResourceLeakDetector track AbstractDnsMessage - Remove DnsMessage.sender/recipient properties - Wrap DnsMessage with AddressedEnvelope - Add DatagramDnsQuest and DatagramDnsResponse for ease of use - Rename DnsQueryEncoder to DatagramDnsQueryEncoder - Rename DnsResponseDecoder to DatagramDnsResponseDecoder - Miscellaneous changes - Add StringUtil.TAB Result: - Cleaner APi - Can support DNS-over-TCP more easily in the future - Reduced memory footprint in the default DnsQuery/Response implementations - Better leak tracking for DnsMessages - Possibility to introduce new DnsRecord types in the future and provide full record encoder/decoder implementation. - No unnecessary instantiation for an EDNS pseudo resource record
2015-03-16 07:46:14 +01:00
final int answerCount = response.count(DnsSection.ANSWER);
boolean found = false;
Revamp DNS codec Motivation: There are various known issues in netty-codec-dns: - Message types are not interfaces, which can make it difficult for a user to implement his/her own message implementation. - Some class names and field names do not match with the terms in the RFC. - The support for decoding a DNS record was limited. A user had to encode and decode by him/herself. - The separation of DnsHeader from DnsMessage was unnecessary, although it is fine conceptually. - Buffer leak caused by DnsMessage was difficult to analyze, because the leak detector tracks down the underlying ByteBuf rather than the DnsMessage itself. - DnsMessage assumes DNS-over-UDP. - To send an EDNS message, a user have to create a new DNS record class instance unnecessarily. Modifications: - Make all message types interfaces and add default implementations - Rename some classes, properties, and constants to match the RFCs - DnsResource -> DnsRecord - DnsType -> DnsRecordType - and many more - Remove DnsClass and use an integer to support EDNS better - Add DnsRecordEncoder/DnsRecordDecoder and their default implementations - DnsRecord does not require RDATA to be ByteBuf anymore. - Add DnsRawRecord as the catch-all record type - Merge DnsHeader into DnsMessage - Make ResourceLeakDetector track AbstractDnsMessage - Remove DnsMessage.sender/recipient properties - Wrap DnsMessage with AddressedEnvelope - Add DatagramDnsQuest and DatagramDnsResponse for ease of use - Rename DnsQueryEncoder to DatagramDnsQueryEncoder - Rename DnsResponseDecoder to DatagramDnsResponseDecoder - Miscellaneous changes - Add StringUtil.TAB Result: - Cleaner APi - Can support DNS-over-TCP more easily in the future - Reduced memory footprint in the default DnsQuery/Response implementations - Better leak tracking for DnsMessages - Possibility to introduce new DnsRecord types in the future and provide full record encoder/decoder implementation. - No unnecessary instantiation for an EDNS pseudo resource record
2015-03-16 07:46:14 +01:00
for (int i = 0; i < answerCount; i ++) {
final DnsRecord r = response.recordAt(DnsSection.ANSWER, i);
final DnsRecordType type = r.type();
if (type != DnsRecordType.A && type != DnsRecordType.AAAA) {
continue;
}
final String qName = question.name().toLowerCase(Locale.US);
final String rName = r.name().toLowerCase(Locale.US);
// Make sure the record is for the questioned domain.
if (!rName.equals(qName)) {
// Even if the record's name is not exactly same, it might be an alias defined in the CNAME records.
String resolved = qName;
do {
resolved = cnames.get(resolved);
if (rName.equals(resolved)) {
break;
}
} while (resolved != null);
if (resolved == null) {
continue;
}
}
Revamp DNS codec Motivation: There are various known issues in netty-codec-dns: - Message types are not interfaces, which can make it difficult for a user to implement his/her own message implementation. - Some class names and field names do not match with the terms in the RFC. - The support for decoding a DNS record was limited. A user had to encode and decode by him/herself. - The separation of DnsHeader from DnsMessage was unnecessary, although it is fine conceptually. - Buffer leak caused by DnsMessage was difficult to analyze, because the leak detector tracks down the underlying ByteBuf rather than the DnsMessage itself. - DnsMessage assumes DNS-over-UDP. - To send an EDNS message, a user have to create a new DNS record class instance unnecessarily. Modifications: - Make all message types interfaces and add default implementations - Rename some classes, properties, and constants to match the RFCs - DnsResource -> DnsRecord - DnsType -> DnsRecordType - and many more - Remove DnsClass and use an integer to support EDNS better - Add DnsRecordEncoder/DnsRecordDecoder and their default implementations - DnsRecord does not require RDATA to be ByteBuf anymore. - Add DnsRawRecord as the catch-all record type - Merge DnsHeader into DnsMessage - Make ResourceLeakDetector track AbstractDnsMessage - Remove DnsMessage.sender/recipient properties - Wrap DnsMessage with AddressedEnvelope - Add DatagramDnsQuest and DatagramDnsResponse for ease of use - Rename DnsQueryEncoder to DatagramDnsQueryEncoder - Rename DnsResponseDecoder to DatagramDnsResponseDecoder - Miscellaneous changes - Add StringUtil.TAB Result: - Cleaner APi - Can support DNS-over-TCP more easily in the future - Reduced memory footprint in the default DnsQuery/Response implementations - Better leak tracking for DnsMessages - Possibility to introduce new DnsRecord types in the future and provide full record encoder/decoder implementation. - No unnecessary instantiation for an EDNS pseudo resource record
2015-03-16 07:46:14 +01:00
if (!(r instanceof DnsRawRecord)) {
continue;
}
final ByteBuf content = ((ByteBufHolder) r).content();
final int contentLen = content.readableBytes();
if (contentLen != INADDRSZ4 && contentLen != INADDRSZ6) {
continue;
}
final byte[] addrBytes = new byte[contentLen];
content.getBytes(content.readerIndex(), addrBytes);
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
final InetAddress resolved;
try {
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
resolved = InetAddress.getByAddress(hostname, addrBytes);
} catch (UnknownHostException e) {
// Should never reach here.
throw new Error(e);
}
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
if (resolvedEntries == null) {
resolvedEntries = new ArrayList<DnsCacheEntry>(8);
}
final DnsCacheEntry e = new DnsCacheEntry(hostname, resolved);
resolveCache.cache(hostname, additionals, resolved, r.timeToLive(), parent.ch.eventLoop());
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
resolvedEntries.add(e);
found = true;
// Note that we do not break from the loop here, so we decode/cache all A/AAAA records.
}
if (found) {
return;
}
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
if (traceEnabled) {
addTrace(envelope.sender(), "no matching " + qType + " record found");
}
// We aked for A/AAAA but we got only CNAME.
if (!cnames.isEmpty()) {
onResponseCNAME(question, envelope, cnames, false, promise);
}
}
private void onResponseCNAME(DnsQuestion question, AddressedEnvelope<DnsResponse, InetSocketAddress> envelope,
Promise<T> promise) {
onResponseCNAME(question, envelope, buildAliasMap(envelope.content()), true, promise);
}
private void onResponseCNAME(
Revamp DNS codec Motivation: There are various known issues in netty-codec-dns: - Message types are not interfaces, which can make it difficult for a user to implement his/her own message implementation. - Some class names and field names do not match with the terms in the RFC. - The support for decoding a DNS record was limited. A user had to encode and decode by him/herself. - The separation of DnsHeader from DnsMessage was unnecessary, although it is fine conceptually. - Buffer leak caused by DnsMessage was difficult to analyze, because the leak detector tracks down the underlying ByteBuf rather than the DnsMessage itself. - DnsMessage assumes DNS-over-UDP. - To send an EDNS message, a user have to create a new DNS record class instance unnecessarily. Modifications: - Make all message types interfaces and add default implementations - Rename some classes, properties, and constants to match the RFCs - DnsResource -> DnsRecord - DnsType -> DnsRecordType - and many more - Remove DnsClass and use an integer to support EDNS better - Add DnsRecordEncoder/DnsRecordDecoder and their default implementations - DnsRecord does not require RDATA to be ByteBuf anymore. - Add DnsRawRecord as the catch-all record type - Merge DnsHeader into DnsMessage - Make ResourceLeakDetector track AbstractDnsMessage - Remove DnsMessage.sender/recipient properties - Wrap DnsMessage with AddressedEnvelope - Add DatagramDnsQuest and DatagramDnsResponse for ease of use - Rename DnsQueryEncoder to DatagramDnsQueryEncoder - Rename DnsResponseDecoder to DatagramDnsResponseDecoder - Miscellaneous changes - Add StringUtil.TAB Result: - Cleaner APi - Can support DNS-over-TCP more easily in the future - Reduced memory footprint in the default DnsQuery/Response implementations - Better leak tracking for DnsMessages - Possibility to introduce new DnsRecord types in the future and provide full record encoder/decoder implementation. - No unnecessary instantiation for an EDNS pseudo resource record
2015-03-16 07:46:14 +01:00
DnsQuestion question, AddressedEnvelope<DnsResponse, InetSocketAddress> response,
Map<String, String> cnames, boolean trace, Promise<T> promise) {
// Resolve the host name in the question into the real host name.
final String name = question.name().toLowerCase(Locale.US);
String resolved = name;
boolean found = false;
while (!cnames.isEmpty()) { // Do not attempt to call Map.remove() when the Map is empty
// because it can be Collections.emptyMap()
// whose remove() throws a UnsupportedOperationException.
final String next = cnames.remove(resolved);
if (next != null) {
found = true;
resolved = next;
} else {
break;
}
}
if (found) {
followCname(response.sender(), name, resolved, promise);
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
} else if (trace && traceEnabled) {
addTrace(response.sender(), "no matching CNAME record found");
}
}
private static Map<String, String> buildAliasMap(DnsResponse response) {
Revamp DNS codec Motivation: There are various known issues in netty-codec-dns: - Message types are not interfaces, which can make it difficult for a user to implement his/her own message implementation. - Some class names and field names do not match with the terms in the RFC. - The support for decoding a DNS record was limited. A user had to encode and decode by him/herself. - The separation of DnsHeader from DnsMessage was unnecessary, although it is fine conceptually. - Buffer leak caused by DnsMessage was difficult to analyze, because the leak detector tracks down the underlying ByteBuf rather than the DnsMessage itself. - DnsMessage assumes DNS-over-UDP. - To send an EDNS message, a user have to create a new DNS record class instance unnecessarily. Modifications: - Make all message types interfaces and add default implementations - Rename some classes, properties, and constants to match the RFCs - DnsResource -> DnsRecord - DnsType -> DnsRecordType - and many more - Remove DnsClass and use an integer to support EDNS better - Add DnsRecordEncoder/DnsRecordDecoder and their default implementations - DnsRecord does not require RDATA to be ByteBuf anymore. - Add DnsRawRecord as the catch-all record type - Merge DnsHeader into DnsMessage - Make ResourceLeakDetector track AbstractDnsMessage - Remove DnsMessage.sender/recipient properties - Wrap DnsMessage with AddressedEnvelope - Add DatagramDnsQuest and DatagramDnsResponse for ease of use - Rename DnsQueryEncoder to DatagramDnsQueryEncoder - Rename DnsResponseDecoder to DatagramDnsResponseDecoder - Miscellaneous changes - Add StringUtil.TAB Result: - Cleaner APi - Can support DNS-over-TCP more easily in the future - Reduced memory footprint in the default DnsQuery/Response implementations - Better leak tracking for DnsMessages - Possibility to introduce new DnsRecord types in the future and provide full record encoder/decoder implementation. - No unnecessary instantiation for an EDNS pseudo resource record
2015-03-16 07:46:14 +01:00
final int answerCount = response.count(DnsSection.ANSWER);
Map<String, String> cnames = null;
Revamp DNS codec Motivation: There are various known issues in netty-codec-dns: - Message types are not interfaces, which can make it difficult for a user to implement his/her own message implementation. - Some class names and field names do not match with the terms in the RFC. - The support for decoding a DNS record was limited. A user had to encode and decode by him/herself. - The separation of DnsHeader from DnsMessage was unnecessary, although it is fine conceptually. - Buffer leak caused by DnsMessage was difficult to analyze, because the leak detector tracks down the underlying ByteBuf rather than the DnsMessage itself. - DnsMessage assumes DNS-over-UDP. - To send an EDNS message, a user have to create a new DNS record class instance unnecessarily. Modifications: - Make all message types interfaces and add default implementations - Rename some classes, properties, and constants to match the RFCs - DnsResource -> DnsRecord - DnsType -> DnsRecordType - and many more - Remove DnsClass and use an integer to support EDNS better - Add DnsRecordEncoder/DnsRecordDecoder and their default implementations - DnsRecord does not require RDATA to be ByteBuf anymore. - Add DnsRawRecord as the catch-all record type - Merge DnsHeader into DnsMessage - Make ResourceLeakDetector track AbstractDnsMessage - Remove DnsMessage.sender/recipient properties - Wrap DnsMessage with AddressedEnvelope - Add DatagramDnsQuest and DatagramDnsResponse for ease of use - Rename DnsQueryEncoder to DatagramDnsQueryEncoder - Rename DnsResponseDecoder to DatagramDnsResponseDecoder - Miscellaneous changes - Add StringUtil.TAB Result: - Cleaner APi - Can support DNS-over-TCP more easily in the future - Reduced memory footprint in the default DnsQuery/Response implementations - Better leak tracking for DnsMessages - Possibility to introduce new DnsRecord types in the future and provide full record encoder/decoder implementation. - No unnecessary instantiation for an EDNS pseudo resource record
2015-03-16 07:46:14 +01:00
for (int i = 0; i < answerCount; i ++) {
final DnsRecord r = response.recordAt(DnsSection.ANSWER, i);
final DnsRecordType type = r.type();
if (type != DnsRecordType.CNAME) {
continue;
}
if (!(r instanceof DnsRawRecord)) {
continue;
}
Revamp DNS codec Motivation: There are various known issues in netty-codec-dns: - Message types are not interfaces, which can make it difficult for a user to implement his/her own message implementation. - Some class names and field names do not match with the terms in the RFC. - The support for decoding a DNS record was limited. A user had to encode and decode by him/herself. - The separation of DnsHeader from DnsMessage was unnecessary, although it is fine conceptually. - Buffer leak caused by DnsMessage was difficult to analyze, because the leak detector tracks down the underlying ByteBuf rather than the DnsMessage itself. - DnsMessage assumes DNS-over-UDP. - To send an EDNS message, a user have to create a new DNS record class instance unnecessarily. Modifications: - Make all message types interfaces and add default implementations - Rename some classes, properties, and constants to match the RFCs - DnsResource -> DnsRecord - DnsType -> DnsRecordType - and many more - Remove DnsClass and use an integer to support EDNS better - Add DnsRecordEncoder/DnsRecordDecoder and their default implementations - DnsRecord does not require RDATA to be ByteBuf anymore. - Add DnsRawRecord as the catch-all record type - Merge DnsHeader into DnsMessage - Make ResourceLeakDetector track AbstractDnsMessage - Remove DnsMessage.sender/recipient properties - Wrap DnsMessage with AddressedEnvelope - Add DatagramDnsQuest and DatagramDnsResponse for ease of use - Rename DnsQueryEncoder to DatagramDnsQueryEncoder - Rename DnsResponseDecoder to DatagramDnsResponseDecoder - Miscellaneous changes - Add StringUtil.TAB Result: - Cleaner APi - Can support DNS-over-TCP more easily in the future - Reduced memory footprint in the default DnsQuery/Response implementations - Better leak tracking for DnsMessages - Possibility to introduce new DnsRecord types in the future and provide full record encoder/decoder implementation. - No unnecessary instantiation for an EDNS pseudo resource record
2015-03-16 07:46:14 +01:00
final ByteBuf recordContent = ((ByteBufHolder) r).content();
final String domainName = decodeDomainName(recordContent);
if (domainName == null) {
continue;
}
if (cnames == null) {
cnames = new HashMap<String, String>();
}
Revamp DNS codec Motivation: There are various known issues in netty-codec-dns: - Message types are not interfaces, which can make it difficult for a user to implement his/her own message implementation. - Some class names and field names do not match with the terms in the RFC. - The support for decoding a DNS record was limited. A user had to encode and decode by him/herself. - The separation of DnsHeader from DnsMessage was unnecessary, although it is fine conceptually. - Buffer leak caused by DnsMessage was difficult to analyze, because the leak detector tracks down the underlying ByteBuf rather than the DnsMessage itself. - DnsMessage assumes DNS-over-UDP. - To send an EDNS message, a user have to create a new DNS record class instance unnecessarily. Modifications: - Make all message types interfaces and add default implementations - Rename some classes, properties, and constants to match the RFCs - DnsResource -> DnsRecord - DnsType -> DnsRecordType - and many more - Remove DnsClass and use an integer to support EDNS better - Add DnsRecordEncoder/DnsRecordDecoder and their default implementations - DnsRecord does not require RDATA to be ByteBuf anymore. - Add DnsRawRecord as the catch-all record type - Merge DnsHeader into DnsMessage - Make ResourceLeakDetector track AbstractDnsMessage - Remove DnsMessage.sender/recipient properties - Wrap DnsMessage with AddressedEnvelope - Add DatagramDnsQuest and DatagramDnsResponse for ease of use - Rename DnsQueryEncoder to DatagramDnsQueryEncoder - Rename DnsResponseDecoder to DatagramDnsResponseDecoder - Miscellaneous changes - Add StringUtil.TAB Result: - Cleaner APi - Can support DNS-over-TCP more easily in the future - Reduced memory footprint in the default DnsQuery/Response implementations - Better leak tracking for DnsMessages - Possibility to introduce new DnsRecord types in the future and provide full record encoder/decoder implementation. - No unnecessary instantiation for an EDNS pseudo resource record
2015-03-16 07:46:14 +01:00
cnames.put(r.name().toLowerCase(Locale.US), domainName.toLowerCase(Locale.US));
}
return cnames != null? cnames : Collections.<String, String>emptyMap();
}
void tryToFinishResolve(Promise<T> promise) {
if (!queriesInProgress.isEmpty()) {
// There are still some queries we did not receive responses for.
if (gotPreferredAddress()) {
// But it's OK to finish the resolution process if we got a resolved address of the preferred type.
finishResolve(promise);
}
// We did not get any resolved address of the preferred type, so we can't finish the resolution process.
return;
}
// There are no queries left to try.
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
if (resolvedEntries == null) {
// .. and we could not find any A/AAAA records.
if (!triedCNAME) {
// As the last resort, try to query CNAME, just in case the name server has it.
triedCNAME = true;
query(hostname, DnsRecordType.CNAME, nameServerAddrs.next(), promise);
return;
}
}
// We have at least one resolved address or tried CNAME as the last resort..
finishResolve(promise);
}
private boolean gotPreferredAddress() {
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
if (resolvedEntries == null) {
return false;
}
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
final int size = resolvedEntries.size();
switch (parent.preferredAddressType()) {
case IPv4:
for (int i = 0; i < size; i ++) {
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
if (resolvedEntries.get(i).address() instanceof Inet4Address) {
return true;
}
}
break;
case IPv6:
for (int i = 0; i < size; i ++) {
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
if (resolvedEntries.get(i).address() instanceof Inet6Address) {
return true;
}
}
break;
default:
throw new Error();
}
return false;
}
private void finishResolve(Promise<T> promise) {
if (!queriesInProgress.isEmpty()) {
// If there are queries in progress, we should cancel it because we already finished the resolution.
Revamp DNS codec Motivation: There are various known issues in netty-codec-dns: - Message types are not interfaces, which can make it difficult for a user to implement his/her own message implementation. - Some class names and field names do not match with the terms in the RFC. - The support for decoding a DNS record was limited. A user had to encode and decode by him/herself. - The separation of DnsHeader from DnsMessage was unnecessary, although it is fine conceptually. - Buffer leak caused by DnsMessage was difficult to analyze, because the leak detector tracks down the underlying ByteBuf rather than the DnsMessage itself. - DnsMessage assumes DNS-over-UDP. - To send an EDNS message, a user have to create a new DNS record class instance unnecessarily. Modifications: - Make all message types interfaces and add default implementations - Rename some classes, properties, and constants to match the RFCs - DnsResource -> DnsRecord - DnsType -> DnsRecordType - and many more - Remove DnsClass and use an integer to support EDNS better - Add DnsRecordEncoder/DnsRecordDecoder and their default implementations - DnsRecord does not require RDATA to be ByteBuf anymore. - Add DnsRawRecord as the catch-all record type - Merge DnsHeader into DnsMessage - Make ResourceLeakDetector track AbstractDnsMessage - Remove DnsMessage.sender/recipient properties - Wrap DnsMessage with AddressedEnvelope - Add DatagramDnsQuest and DatagramDnsResponse for ease of use - Rename DnsQueryEncoder to DatagramDnsQueryEncoder - Rename DnsResponseDecoder to DatagramDnsResponseDecoder - Miscellaneous changes - Add StringUtil.TAB Result: - Cleaner APi - Can support DNS-over-TCP more easily in the future - Reduced memory footprint in the default DnsQuery/Response implementations - Better leak tracking for DnsMessages - Possibility to introduce new DnsRecord types in the future and provide full record encoder/decoder implementation. - No unnecessary instantiation for an EDNS pseudo resource record
2015-03-16 07:46:14 +01:00
for (Iterator<Future<AddressedEnvelope<DnsResponse, InetSocketAddress>>> i = queriesInProgress.iterator();
i.hasNext();) {
Future<AddressedEnvelope<DnsResponse, InetSocketAddress>> f = i.next();
i.remove();
if (!f.cancel(false)) {
f.addListener(RELEASE_RESPONSE);
}
}
}
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
if (resolvedEntries != null) {
// Found at least one resolved address.
for (InternetProtocolFamily f: resolveAddressTypes) {
if (finishResolve(f.addressType(), resolvedEntries, promise)) {
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
return;
}
}
}
// No resolved address found.
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
final int tries = maxAllowedQueries - allowedQueries;
final StringBuilder buf = new StringBuilder(64);
buf.append("failed to resolve '");
if (pristineHostname != null) {
buf.append(pristineHostname);
} else {
buf.append(hostname);
}
buf.append('\'');
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
if (tries > 1) {
if (tries < maxAllowedQueries) {
buf.append(" after ")
.append(tries)
.append(" queries ");
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
} else {
buf.append(". Exceeded max queries per resolve ")
.append(maxAllowedQueries)
.append(' ');
}
}
if (trace != null) {
buf.append(':')
.append(trace);
}
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
final UnknownHostException cause = new UnknownHostException(buf.toString());
resolveCache.cache(hostname, additionals, cause, parent.ch.eventLoop());
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
promise.tryFailure(cause);
}
abstract boolean finishResolve(Class<? extends InetAddress> addressType, List<DnsCacheEntry> resolvedEntries,
Promise<T> promise);
abstract DnsNameResolverContext<T> newResolverContext(DnsNameResolver parent, String hostname,
DnsRecord[] additionals, DnsCache resolveCache);
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
static String decodeDomainName(ByteBuf in) {
in.markReaderIndex();
try {
return DefaultDnsRecordDecoder.decodeName(in);
} catch (CorruptedFrameException e) {
// In this case we just return null.
return null;
} finally {
in.resetReaderIndex();
}
}
private void followCname(InetSocketAddress nameServerAddr, String name, String cname, Promise<T> promise) {
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
if (traceEnabled) {
if (trace == null) {
trace = new StringBuilder(128);
}
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
trace.append(StringUtil.NEWLINE);
trace.append("\tfrom ");
trace.append(nameServerAddr);
trace.append(": ");
trace.append(name);
trace.append(" CNAME ");
trace.append(cname);
}
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
final InetSocketAddress nextAddr = nameServerAddrs.next();
if (parent.isCnameFollowARecords() && !query(hostname, DnsRecordType.A, nextAddr, promise)) {
return;
}
if (parent.isCnameFollowAAAARecords()) {
query(hostname, DnsRecordType.AAAA, nextAddr, promise);
}
}
private boolean query(String hostname, DnsRecordType type, final InetSocketAddress nextAddr, Promise<T> promise) {
final DnsQuestion question;
try {
question = new DefaultDnsQuestion(hostname, type);
} catch (IllegalArgumentException e) {
// java.net.IDN.toASCII(...) may throw an IllegalArgumentException if it fails to parse the hostname
promise.tryFailure(e);
return false;
}
query(nextAddr, question, promise);
return true;
}
private void addTrace(InetSocketAddress nameServerAddr, String msg) {
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
assert traceEnabled;
if (trace == null) {
trace = new StringBuilder(128);
}
trace.append(StringUtil.NEWLINE);
trace.append("\tfrom ");
trace.append(nameServerAddr);
trace.append(": ");
trace.append(msg);
}
private void addTrace(Throwable cause) {
Provide more control over DnsNameResolver.query() / Add NameResolver.resolveAll() Related issues: - #3971 - #3973 - #3976 - #4035 Motivation: 1. Previously, DnsNameResolver.query() retried the request query by its own. It prevents a user from deciding when to retry or stop. It is also impossible to get the response object whose code is not NOERROR. 2. NameResolver does not have an operation that resolves a host name into multiple addresses, like InetAddress.getAllByName() Modifications: - Changes related with DnsNameResolver.query() - Make query() not retry - Move the retry logic to DnsNameResolver.resolve() instead. - Make query() fail the promise only when I/O error occurred or it failed to get a response - Add DnsNameResolverException and use it when query() fails so that the resolver can give more information about the failure - query() does not cache anymore. - Changes related with NameResolver.resolveAll() - Add NameResolver.resolveAll() - Add SimpleNameResolver.doResolveAll() - Changes related with DnsNameResolver.resolve() and resolveAll() - Make DnsNameResolveContext abstract so that DnsNameResolver can decide to get single or multiple addresses from it - Re-implement cache so that the cache works for resolve() and resolveAll() - Add 'traceEnabled' property to enable/disable trace information - Miscellaneous changes - Use ObjectUtil.checkNotNull() wherever possible - Add InternetProtocolFamily.addressType() to remove repetitive switch-case blocks in DnsNameResolver(Context) - Do not raise an exception when decoding a truncated DNS response Result: - Full control over query() - A user can now retrieve all addresses via (Dns)NameResolver.resolveAll() - DNS cache works only for resolve() and resolveAll() now.
2015-07-12 12:34:05 +02:00
assert traceEnabled;
if (trace == null) {
trace = new StringBuilder(128);
}
trace.append(StringUtil.NEWLINE);
trace.append("Caused by: ");
trace.append(cause);
}
}