Self-certification
Submitted to ICANN by: <Applicant>
String: <TLD>
Application ID: <ID>
Template Version: 1.1, Date: 5 Mars 2013
About this document
This document provides a template for the self-certification documentation
for the Pre Delegation Testing, PDT.
The self-certification documentation shall comply with
- the requirements stated in Module 5 of the AGB
- the requirements stated in the registry agreement including the specifications 2,4,6 and 10
- assertions made by the applicant in the gTLD application
The applicant shall provide the information as outlined below. The
template is based on the requirements in the AGB.
1. DNS infrastructure
AGB, chapter 5.2.2:
All tests shall be done both over IPv4 and IPv6, with
reports providing results according to both protocols.
AGB, chapter 5.2.2:
Self-certification documentation shall include data on load
capacity, latency and network reachability.
1.1. Load capacity
1.1.1. Expected load
Instruction: Describe the expected load on the name servers during
normal operation. If it is a shared name server, then the load of
other zones must be taken into consideration. Do this for both IPv4 and
IPv6.
[A short motivation for figures is expected and a description of the
shared name server, if applicable.
Provide references to the gTLD application, if applicable.]
Applicant Self Certification:
1.1.2. Statistical population
AGB, chapter 5.2.2:
The load capacity test shall be performed against a randomly selected subset of
servers within the applicant’s DNS infrastructure.
Instruction: Describe the selected subset, how it has been selected and
what percentage of the total set of servers it represents.
This applies to both IPv4 and IPv6.
[A short description and motivation is expected.]
Applicant self certification:
1.1.3. Method
AGB, chapter 5.2.2:
Responses must either contain zone data or be NXDOMAIN or NODATA
responses to be considered valid.
Instruction: Describe the method of measurement and what type of
queries and responses are used.
Applicant self certification:
1.1.4 DDoS
Instruction:
Describe mitigation of Distributed Denial of Service (DDoS) attacks.
A description of strategies and controls is expected. Provide references
to the gTLD application, if applicable].
Applicant self certification:
1.1.5. Result
AGB, chapter 5.2.2:
Load capacity shall be reported using a table and a
corresponding graph, showing percentage of queries
responded against an increasing number of queries per
second generated from local (to the servers) traffic
generators. The table shall include at least 20 data points
and loads of UDP-based [and TCP-based] queries that will
cause up to 10% query loss against a randomly selected
subset of servers within the applicant’s DNS infrastructure.
AGB, chapter 5.2.2:
Load capacity [...] [for DNSSEC support] shall be documented
as for UDP and TCP above.
AGB, chapter 5.2.2:
The documentation provided by the applicant must include the
results from a system performance test indicating available network
and server capacity and an estimate of expected capacity during normal
operation to ensure stable service as well as to adequately address
Distributed Denial of Service (DDoS) attacks.
Instruction:
- Describe available network and server capacity.
[Provide a brief description and a table.]
- Describe an estimate of expected capacity during normal operation.
[Provide a brief motivation for the figures.]
- Create a table and graph according to the requirements.
- The table shall include at least 20 data points.
- The table shall include loads that cause up to 10% query loss
against a randomly selected subset of servers.
- The table shall include both UDP and TCP.
- The table shall include DNSSEC.
- Examples can be found below.
- Repeat for both IPv4 and IPv6.
Applicant self certification:
Example table showing percentage of queries responded against an
increasing number of queries per second:
|
UDP |
TCP |
UDP +DO |
TCP +DO |
# |
QPS |
% |
QPS |
% |
QPS |
% |
QPS |
% |
1 |
500 |
100 |
500 |
100 |
500 |
100 |
500 |
100 |
2 |
1000 |
100 |
1000 |
100 |
1000 |
100 |
1000 |
100 |
3 |
1500 |
100 |
1500 |
100 |
1500 |
100 |
1500 |
100 |
4 |
2000 |
100 |
2000 |
99 |
2000 |
100 |
2000 |
99 |
5 |
2500 |
100 |
2500 |
98 |
2500 |
100 |
2500 |
98 |
6 |
3000 |
100 |
3000 |
97 |
3000 |
99.5 |
3000 |
96 |
7 |
3500 |
99 |
3300 |
96 |
3400 |
99 |
3200 |
95.5 |
8 |
4000 |
98.5 |
3700 |
95 |
3800 |
98.5 |
3600 |
95 |
9 |
4500 |
98 |
3800 |
95 |
4000 |
98 |
3700 |
94.5 |
10 |
5000 |
97.5 |
3900 |
94 |
4100 |
97.5 |
3800 |
94 |
11 |
5500 |
97 |
4000 |
94.5 |
4200 |
97 |
3900 |
93.5 |
12 |
6000 |
96 |
4100 |
94 |
4300 |
96 |
3950 |
93.5 |
13 |
6400 |
95 |
4200 |
93.5 |
4400 |
95 |
4000 |
93.25 |
14 |
6600 |
94.5 |
4300 |
93 |
4500 |
94 |
4050 |
93 |
15 |
6700 |
94 |
4350 |
92.5 |
4600 |
93.5 |
4100 |
92.75 |
16 |
6800 |
93 |
4400 |
92 |
4700 |
93 |
4150 |
92.5 |
17 |
6900 |
92 |
4450 |
91.5 |
4750 |
92.5 |
4200 |
91.5 |
18 |
7000 |
91 |
4500 |
91 |
4800 |
91.5 |
4225 |
91 |
19 |
7100 |
90.5 |
4525 |
90.5 |
4825 |
90 |
4250 |
90.5 |
20 |
7200 |
90 |
4550 |
90 |
4875 |
90 |
4260 |
90 |
Example graph:
1.2. Query latency
AGB, chapter 5.2.2:
Query latency shall be reported in milliseconds as
measured by DNS probes located just outside the border
routers of the physical network hosting the name servers,
from a network topology point of view.
1.2.1. Method
Registry agreement, specification 10:
3.8. Measuring DNS parameters. Every minute, every DNS probe will make an UDP or TCP “DNS
test” to each of the public-DNS registered “IP addresses” of the name servers of the domain
name being monitored. If a “DNS test” result is undefined/unanswered, the tested IP will be
considered unavailable from that probe until it is time to make a new test.
Registry agreement, specification 10:
3.9. Collating the results from DNS probes. The minimum number of active testing probes to
consider a measurement valid is 20 at any given measurement period, otherwise the
measurements will be discarded and will be considered inconclusive; during this situation no
fault will be flagged against the SLRs.
Registry agreement, specification 10:
3.11. Placement of DNS probes. Probes for measuring DNS parameters shall be placed as
near as possible to the DNS resolvers on the networks with the most users across the different
geographic regions; care shall be taken not to deploy probes behind high propagation-delay
links, such as satellite links.
Instruction: Describe how the query latency has been measured.
[A brief description showing how the requirements are met.]
Applicant self certification:
1.2.2. Result
Instruction: Report the min/avg/max values of query latency in milliseconds.
The test shall include both UDP and TCP. The tests shall also include DNSSEC.
Perform the tests on all name servers (IPv4/IPv6).
Applicant self certification:
Example table showing query latency (min/avg/max) in milliseconds:
NS |
IP |
UDP |
TCP |
UDP +DO |
TCP +DO |
a.ns.TLD |
192.0.2.1 |
120/125/134 |
150/151/152 |
122/140/200 |
152/160/171 |
a.ns.TLD |
2001:DB8::1 |
100/103/109 |
125/126/127 |
110/115/116 |
135/141/145 |
b.ns.TLD |
192.0.2.2 |
90/91/92 |
90/93/94 |
91/91/92 |
91/97/100 |
b.ns.TLD |
2001:DB8::2 |
98/99/103 |
110/115/118 |
105/125/190 |
107/130/200 |
c.ns.TLD |
192.0.2.3 |
50/55/60 |
53/54/55 |
60/69/81 |
70/71/72 |
1.3. Reachability
1.3.1. Peering
AGB, chapter 5.2.2:
Reachability will be documented by providing information
on the transit and peering arrangements for the DNS server
locations, listing the AS numbers of the transit providers or
peers at each point of presence and available bandwidth
at those points of presence.
Instruction: Provide the information as specified by the
requirements.
Applicant self certification:
Example table showing transit and peering arrangements:
NS |
IP |
AS |
Bandwidth |
a.ns.TLD |
192.0.2.1 |
AS1 AS2 AS3 |
100 Mbit/s |
a.ns.TLD |
2001::DB8:1 |
AS1 AS4 AS5 |
100 Mbit/s |
b.ns.TLD |
192.0.2.2 |
AS3 AS4 AS5 |
1 Gbit/s |
b.ns.TLD |
2001::DB8:2 |
AS6 AS7 AS8 |
1 Gbit/s |
c.ns.TLD |
192.0.2.3 |
AS9 |
2 Gbit/s |
1.3.2. Reachability TCP-based DNS queries
AGB, chapter 5.2.2:
Reachability will be documented by providing records of
TCP-based DNS queries from nodes external to the network
hosting the servers. These locations may be the same as
those used for measuring latency above.
Instruction: Provide the information as specified by the
requirements. Repeat for both IPv4 and IPv6.
Applicant self certification:
1.4 DNSSEC
AGB, chapter 5.2.2:
DNSSEC support -- Applicant must demonstrate support for
EDNS(0) in its server infrastructure, the ability to return
correct DNSSEC-related resource records such as DNSKEY,
RRSIG, and NSEC/NSEC3 for the signed zone, and the
ability to accept and publish DS resource records from
second-level domain administrators. In particular, the
applicant must demonstrate its ability to support the full life
cycle of KSK and ZSK keys.
Instruction:
Provide information regarding
- support for EDNS(0) in the server infrastructure
- the ability to return correct DNSSEC-related resource records
(e.g. DNSKEY, RRSIG, NEC/NSEC3) for the signed zone
- the ability to accept and publish DS resource records
from second-level domain administrators.
- the applicants ability to support the full life cycle of
cryptographic keys
Applicant self certification:
1.5. SLA, DNS Infrastructure
Specification:
The DNS infrastructure must meet all service level requirements
described in specification 10 of the registry agreement.
The infrastructure must also meet all service level assertions
made in the gTLD application in relation to system performance as
described in specification 10 of the registry agreement.
The infrastructure must furthermore meet all other performance
assertions made in the gTLD application.
Instruction:
- Provide a table describing the adherence to the SLA Service Level
Requirements as given in specification 10 of the registry agreement.
- Provide references to all assertions in the gTLD application relating
to system performance as given in specification 10 of the registry agreement.
- Provide a table describing the adherence to the above mentioned assertions.
- Provide references to all other assertions in the gTLD application relating to
system performance.
- Provide a description of the adherence to these assertions.
[All non adherence must be stated.]
Applicant self certification:
Example table showing adherence to the SLA:
Parameter |
Self-certification (monthly basis) |
DNS service availability |
____ min downtime |
DNS name server availability |
≤ ____ min of downtime |
TCP DNS resolution RTT |
≤ ____ ms, for at least 95% of the queries |
UDP DNS resolution RTT |
≤ ____ ms, for at least 95% of the queries |
DNS update time |
≤ ____ min, for at least 95% of the probes |
2. Whois
AGB, chapter 5.2.3:
Applicant must provide Whois services for
the anticipated load. ICANN will verify that Whois data is
accessible over IPv4 and IPv6 via both TCP port 43 and via
a web interface and review self-certification
documentation regarding Whois transaction capacity.
2.1. Load capacity
AGB, chapter 5.2.3:
Self-certification documents shall describe the maximum
number of queries per second successfully handled by
both the port 43 servers as well as the web interface,
together with an applicant-provided load expectation.
2.1.1. Expected load
Instruction: Describe the expected load on the Whois service
(port 43 and HTTP) during normal operation. If it is a Shared Registry
System, then the load of other TLD:s must be taken into consideration.
Do this for both IPv4 and IPv6.
[A short motivation for the figures is expected and a short description of the
shared registry system, if applicable.
Provide references to the gTLD application, if applicable.]
Applicant self certification:
2.1.2. Port 43
2.1.2.1. Method
Instruction: Describe how the performance has been measured.
[A brief description is expected.]
Applicant self certification:
2.1.2.2. Result
Instruction: Describe the maximum number of queries per second
successfully handled by the port 43 servers over IPv4 and IPv6.
Applicant self certification:
Example table showing maximum queries per second:
2.1.3. HTTP
2.1.3.1. Method
Instruction: Describe how the performance has been measured.
[A brief description is expected.]
Applicant self certification:
2.1.3.2. Result
Instruction: Describe the maximum number of queries per second
successfully handled by the web interface over IPv4 and IPv6.
Applicant self certification:
Example table showing maximum queries per second:
(QPS) |
HTTP |
HTTPS |
IPv4 |
500 |
400 |
IPv6 |
505 |
396 |
2.2. Data mining
AGB, chapter 5.2.3:
Additionally, a description of deployed control functions to
detect and mitigate data mining of the Whois database
shall be documented.
Instruction: Describe the control functions for detection
and mitigation of datamining of the Whois database.
This is applicable for both port 43 and HTTP.
[The description should cover both the strategy used and implemented controls.
Provide references to the gTLD application, if applicable.]
Applicant self certification:
2.3. Searchable whois
Instruction: Describe any provision made for searchable Whois lookup services for
names registered in the TLD.
[Provide references to the gTLD application, if applicable.]
Applicant self certification:
2.4. SLA, Whois
AGB, chapter 5.2.3:
System performance -- The registry system must scale to
meet the performance requirements described in
specification 10 of the registry agreement.
Specification:
The registry data directory services must meet all service level
requirements described in specification 10 of the registry agreement.
The services must also meet all service level assertions made in the
gTLD application in relation to system performance as described in
specification 10 of the registry agreement.
The services must furthermore meet all other performance assertions
made in the gTLD application.
Instruction:
- Provide a table describing the adherence to the Service Level Requirements as
given in specification 10 of the registry agreement.
- Provide references to all assertions in the gTLD application relating to system
performance as given in Specification 10 of the registry agreement.
- Provide a table describing the adherence to the above mentioned assertions.
- Provide references to all other assertions in the gTLD application relating to system performance.
- Provide a description of the adherence to these assertions.
Applicant self certification:
Example table showing adherence to the SLA:
Parameter |
Self-certification (monthly basis) |
RDDS availability |
____ min downtime |
RDDS query RTT |
≤ ____ ms, for at least 95% of the queries |
RDDS update time |
≤ ____ min, for at least 95% of the probes |
3. EPP
3.1. Load capacity
AGB, chapter 5.2.3:
Documentation shall provide a maximum Transaction per
Second rate for the EPP interface with 10 data points
corresponding to registry database sizes from 0 (empty) to
the expected size after one year of operation, as
determined by applicant.
Instruction:
- Describe the expected registry database size after one year of operation.
A brief motivation is expected.]
- Describe the transaction capacity as a function of the
registry database. The description shall contain at least 10 data points
ranging from an empty database to the size after one year.
[A motivation should be given.
Provide references to the gTLD application, if applicable.]
Applicant self certification:
Example table showing the load capacity:
# |
Number of domains |
TPS |
1 |
0 |
2000 |
2 |
200000 |
1900 |
3 |
250000 |
1850 |
4 |
300000 |
1825 |
5 |
350000 |
1800 |
6 |
400000 |
1775 |
7 |
500000 |
1750 |
8 |
650000 |
1700 |
9 |
800000 |
1625 |
10 |
1000000 |
1550 |
3.2. Initial load
AGB, chapter 5.2.3:
Documentation shall also describe measures taken to
handle load during initial registry operations, such as a
land-rush period.
Instruction: Describe how the load is handled for the initial
registry operations, such as a land-rush period.
[A brief description is expected. A justification of the chosen solution should be provided.]
Applicant self certification:
3.3. EPP Extensions
Instruction:
- Describe all provided EPP extentions.
[Brief descriptions are expected.]
- Describe how these extensions comply with RFC 3735.
- Provide references to assertions made in gTLD
application relating to EPP extensions.
- Provide a description of the adherence to these assertions.
Applicant self certification:
3.4. EPP IPv6
Instruction:
If EPP services are available over IPv6, give a short description of IPv6 usage.
[Provide references to the gTLD application, if applicable.]
Applicant self certification:
3.5. SLA, EPP
AGB, chapter 5.2.3:
System performance -- The registry system must scale to
meet the performance requirements described in
specification 10 of the registry agreement.
Specification:
The EPP services must meet all service level requirements
described in specification 10 of the registry agreement.
The services must also meet all service level assertions made in the gTLD
application in relation to system performance as
described in specification 10 of the registry agreement.
The services must furthermore meet all other performance assertions
made in the gTLD application.
Instruction:
- Provide a table describing the adherence to the Service Level Requirements as
given in specification 10 of the registry agreement.
- Provide references to all assertions in the gTLD application relating
to system performance as given in Specification 10 of the registry agreement.
- Provide a table describing the adherence to the above mentioned assertions.
- Provide references to all other assertions in the
gTLD application relating to system performance.
- Provide a description of the adherence to these assertions.
Applicant self certification:
Example table showing adherence to the SLA:
Parameter |
Self-certification (monthly basis) |
EPP service availability |
____ min downtime |
EPP session-command RTT |
≤ ____ ms, for at least 90% of the commands |
EPP query-command RTT |
≤ ____ ms, for at least 90% of the commands |
EPP transform-command RTT |
≤ ____ ms, for at least 90% of the commands |
© Internet Corporation For Assigned Names and Numbers.