richardschneider / net-dns Goto Github PK

All DNS strings are USASCII.

WireWriter.WriteString and WireReader.ReadString must enforce this.

Reading a TXT record with no strings is treated as malformed. It should be accepted.

2018/07/09 13:48:59:489 [WARN]  Makaretu.Dns.MulticastService - Received malformed message
=======================================================(inner most exception)===
 (1) System.IO.EndOfStreamException
================================================================================
Method        :  ReadBytes
Type          :  Makaretu.Dns.DnsReader
Assembly      :  Makaretu.Dns, Version=0.11.0.0, Culture=neutral, PublicKeyToken=null
Assembly Path :  C:\Users\Owner\Documents\GitHub\net-mdns\test\bin\Debug\net461\Makaretu.Dns.dll
Source        :  Makaretu.Dns
Thread        :  6 ''
Helplink      :  

Message:
"Attempted to read past the end of the stream."

Stack Trace:
   at Makaretu.Dns.DnsReader.ReadBytes(Int32 length)
   at Makaretu.Dns.DnsReader.ReadString()
   at Makaretu.Dns.TXTRecord.ReadData(DnsReader reader, Int32 length)
   at Makaretu.Dns.ResourceRecord.Read(DnsReader reader)
   at Makaretu.Dns.Message.Read(DnsReader reader)
   at Makaretu.Dns.DnsObject.Read(Byte[] buffer, Int32 offset, Int32 count)
   at Makaretu.Dns.MulticastService.OnDnsMessage(Byte[] datagram, Int32 length) in C:\Users\Owner\Documents\GitHub\net-mdns\src\MulticastService.cs:line 403```

An exception (InvalidateDataException maybe) should be thrown by Read(DnsReader reader)after calling ReadData if the position and RDLEN do not agree. This is the inverse of Buffer Overflow

Make the record more user friendly.

Instead of byte[] TypeBitmaps use List<DnsType> Types.

See https://tools.ietf.org/html/rfc3845 for details.

The following unecrypted Message (in base-64) can not be decoded. It came from cloudflare using DNS over TLS (DOT).

EjSBgAABAAEAAAAABGlwZnMCaW8AABAAAcAMABAAAQAAADwAPTxkbnNsaW5rPS9pcGZzL1FtWU5RSm9LR05IVHBQeENCUGg5S2tEcGFFeGdkMmR1TWEzYUY2eXRNcEhkYW8=

Wildcard queries are not supported.

Also #28 does not allow "*" in a domain name.

Define the Notify message in https://tools.ietf.org/html/rfc1996

Add the resources records for https://tools.ietf.org/html/rfc2065 https://tools.ietf.org/html/rfc4035

See also Important Additions to DNSSEC and RRs for DNSSEC.

• DNSKEY
• RRSIG
• NSEC
• NSEC3
• NSEC3PARAM
• DS

ReadToken should handle escaped octal digits, such as \1 or \123.

See https://tools.ietf.org/html/rfc3425

Implement the OPT resource record from https://tools.ietf.org/html/rfc6891

Add the resource records for https://tools.ietf.org/html/rfc1183

• AFSDB - Andrew File System Database
• RP - Responsible Person
• X25 - X.25 (Not in current use)
• ISDN - Integrated Service Digital Network (Not in current use)
• RT - Route Through (Not in current use)

DnsWriter.WriteString must quote the string if it contains a quote, backslash or space.

See wikipedia and ICANN

Add the TSIG resource record from https://tools.ietf.org/html/rfc2845

Add the LOC resource record defined in https://tools.ietf.org/html/rfc1876

Hi @richardschneider , thank you for the repo!
When I update the simplebase to latest version of 3.0.2 I am getting below error:
Error MT2101: Can't resolve the reference 'System.Byte[] SimpleBase.Base16::Decode(System.String)', referenced from the method 'System.Void Makaretu.Dns.DSRecord::ReadData(Makaretu.Dns.PresentationReader)' in 'SimpleBase, Version=1.3.1.0, Culture=neutral, PublicKeyToken=null'.

Make the NameServer security aware; see https://tools.ietf.org/html/rfc4035#section-3.1

See https://tools.ietf.org/html/rfc4398

This should likely check the UTF-8 encoded version of the string or something similar, not the default C# representation of string. Which is, at least on windows, a UTF-16 encoded string.

Since development has ended here I have continued the project at:
https://github.com/jdomnitz/net-dns

NuGet Packages will continue to be published to:
Makaretu.Dns.New

This includes new features, bug fixes and modern framework targets. Tickets and PRs are welcome.

Do you have plans to support this project?
Will be good to have a possibility to update DNS records

See the build

Appveyor is using docfx v2.38.1. Locally 2.36.1.0 works.

Value equality, hashing and deep cloning can be implemented because of object serialisation.

From RFC 1035

\DDD where each D is a digit is the octet corresponding to
the decimal number described by DDD. The resulting
octet is assumed to be text and is not checked for
special meaning.

For some reason the Presention Reader/Writer is treating escaped numbers as octal. They should be treated as decimal.

• rename DnsReader/DnsWriter => WireReader/WireWriter
• rename Class to DnsClass
• rename to master to presentation
• add PresentationReader
• implement TSIG, TKey
• Add a SecurityAlgoithm registry
• Fix TODOs, or create an issue
• More articles
  • describe serialisation
  • describe the registries
  • DSNSEC

The SignatureExpiration and SignatureExpiration should use the DateTime type.

Presentation format should always YYYYMMDDHHmmSS format. But accept number seconds since Unix epoch.

Currently the NameServer will add most likely needed resources to the AdditionalRecords. For example a query for a PTR will return the SRV and TXT as additional records and a SRV query will return the target host address (A/AAAA) records.

A query for a PTR should also return the address for the target host.

This is needed by some peer-to-peer systems, where they expect the PTR response to contain all the information needed to contact the peer.

The standard requires that domain name labels consist of only USASCII letters, digits or '-'. This is somewhat mitigated by Internationalized Domain Names .

WireFormat.ReadDomainName should throw InvalidDataExecption for characters outside this range. For the time being, WireFormat.WriteDomainName show throw InvalidArgumentException.

PresentationFormat is not affected, because it can be UNICODE.

From https://tools.ietf.org/html/rfc1035#section-2.3.1

<label> ::= <letter> [ [ <ldh-str> ] <let-dig> ]

<ldh-str> ::= <let-dig-hyp> | <let-dig-hyp> <ldh-str>

<let-dig-hyp> ::= <let-dig> | "-"

<let-dig> ::= <letter> | <digit>

<letter> ::= any one of the 52 alphabetic characters A through Z in
upper case and a through z in lower case

<digit> ::= any one of the ten digits 0 through 9

Note that while upper and lower case letters are allowed in domain
names, no significance is attached to the case.  That is, two names with
the same spelling but different case are to be treated as if identical.

The labels must follow the rules for ARPANET host names.  They must
start with a letter, end with a letter or digit, and have as interior
characters only letters, digits, and hyphen.  There are also some
restrictions on the length.  Labels must be 63 characters or less.