Traffic Ops - Using

The Traffic Ops Menu

The following tabs are available in the menu at the top of the Traffic Ops user interface.

• Health

  Information on the health of the system. Hover over this tab to get to the following options:

  Option	Description
  Table View	A real time view into the main performance indicators of the CDNs managed by Traffic Control. This view is sourced directly by the Traffic Monitor data and is updated every 10 seconds. This is the default screen of Traffic Ops. See The Health Table for details.
  Graph View	A real graphical time view into the main performance indicators of the CDNs managed by Traffic Control. This view is sourced by the Traffic Monitor data and is updated every 10 seconds. On loading, this screen will show a history of 24 hours of data from Traffic Stats See Graph View for details.
  Server Checks	A table showing the results of the periodic check extension scripts that are run. See Server Checks
  Daily Summary	A graph displaying the daily peaks of bandwidth, overall bytes served per day, and overall bytes served since initial installation per CDN.

• Delivery Services

  The main Delivery Service table. This is where you Create/Read/Update/Delete Delivery Services of all types. Hover over to get the following sub option:

  Option	Description
  Federations	Add/Edit/Delete Federation Mappings.

• Servers

  The main Servers table. This is where you Create/Read/Update/Delete servers of all types. Click the main tab to get to the main table, and hover over to get these sub options:

  Option	Description
  Upload Server CSV	Bulk add of servers from a csv file. See Bulk Upload Server

• Parameters

  Parameters and Profiles can be edited here. Hover over the tab to get the following options:

  Option	Description
  Global Profile	The table of global parameters. See Profile Parameters. This is where you Create/Read/Update/Delete parameters in the Global profile
  All Cache Groups	The table of all parameters that are assigned to a cachegroup - this may be slow to pull up, as there can be thousands of parameters.
  All Profiles	The table of all parameters that are assigned to a profile - this may be slow to pull up, as there can be thousands of parameters.
  Select Profile	Select the parameter list by profile first, then get a table of just the parameters for that profile.
  Export Profile	Profiles can be exported from one Traffic Ops instance to another using ‘Select Profile’ and under the “Profile Details” dialog for the desired profile
  Import Profile	Profiles can be imported from one Traffic Ops instance to another using the button “Import Profile” after using the “Export Profile” feature
  Orphaned Parameters	A table of parameters that are not associated to any profile of cache group. These parameters either should be deleted or associated with a profile of cache group.

• Tools

  Tools for working with Traffic Ops and it’s servers. Hover over this tab to get the following options:

  Option	Description
  Generate ISO	Generate a bootable image for any of the servers in the Servers table (or any server for that matter). See Generate ISO
  Queue Updates	Send Updates to the caches. See Queue Updates and Snapshot CRConfig
  DB Dump	Backup the Database to a .sql file.
  Snapshot CRConfig	Send updates to the Traffic Monitor / Traffic Router servers. See Queue Updates and Snapshot CRConfig
  Invalidate Content	Invalidate or purge content from all caches in the CDN. See Invalidate Content
  Manage DNSSEC keys	Manage DNSSEC Keys for a chosen CDN.

• Misc

  Miscellaneous editing options. Hover over this tab to get the following options:

  Option	Description
  CDNs	Create/Read/Update/Delete CDNs
  Cache Groups	Create/Read/Update/Delete cache groups
  Users	Create/Read/Update/Delete users
  Profiles	Create/Read/Update/Delete profiles. See Parameters and Profiles
  Networks(ASNs)	Create/Read/Update/Delete Autonomous System Numbers See The Coverage Zone File and ASN Table
  Hardware	Get detailed hardware information (note: this should be moved to a Traffic Ops Extension)
  Data Types	Create/Read/Update/Delete data types
  Divisions	Create/Read/Update/Delete divisions
  Regions	Create/Read/Update/Delete regions
  Physical Locations	Create/Read/Update/Delete locations

• ChangeLog

  The Changelog table displays the changes that are being made to the Traffic Ops database through the Traffic Ops user interface. This tab will show the number of changes since you last visited this tab in (brackets) since the last time you visited this tab. There are currently no sub menus for this tab.

• Help

  Help for Traffic Ops and Traffic Control. Hover over this tab to get the following options:

  Option	Description
  About	Traffic Ops information, such as version, database information, etc
  Release Notes	Release notes for the most recent releases of Traffic Ops
  Logout	Logout from Traffic Ops

Health

The Health Table

The Health table is the default landing screen for Traffic Ops, it displays the status of the EDGE caches in a table form directly from Traffic Monitor (bypassing Traffic Stats), sorted by Mbps Out. The columns in this table are:

• Profile: the Profile of this server or ALL, meaning this row shows data for multiple servers, and the row shows the sum of all values.
• Host Name: the host name of the server or ALL, meaning this row shows data for multiple servers, and the row shows the sum of all values.
• Edge Cache Group: the edge cache group short name or ALL, meaning this row shows data for multiple servers, and the row shows the sum of all values.
• Healthy: indicates if this cache is healthy according to the Health Protocol. A row with ALL in any of the columns will always show a , this column is valid only for individual EDGE caches.
• Admin: shows the administrative status of the server.
• Connections: the number of connections this cache (or group of caches) has open (ats.proxy.process.http.current_client_connections from ATS).
• Mbps Out: the bandwidth being served out if this cache (or group of caches)

Since the top line has ALL, ALL, ALL, it shows the total connections and bandwidth for all caches managed by this instance of Traffic Ops.

Graph View

The Graph View shows a live view of the last 24 hours of bits per seconds served and open connections at the edge in a graph. This data is sourced from Traffic Stats. If there are 2 CDNs configured, this view will show the statistis for both, and the graphs are stacked. On the left-hand side, the totals and immediate values as well as the percentage of total possible capacity are displayed. This view is update every 10 seconds.

Server Checks

The server checks page is intended to give an overview of the Servers managed by Traffic Control as well as their status. This data comes from Traffic Ops extensions.

Name	Description
ILO	Ping the iLO interface for EDGE or MID servers
10G	Ping the IPv4 address of the EDGE or MID servers
10G6	Ping the IPv6 address of the EDGE or MID servers
MTU	Ping the EDGE or MID using the configured MTU from Traffic Ops
FQDN	DNS check that matches what the DNS servers responds with compared to what Traffic Ops has.
DSCP	Checks the DSCP value of packets from the edge server to the Traffic Ops server.
RTR	Content Router checks. Checks the health of the Content Routers. Checks the health of the caches using the Content Routers.
CHR	Cache Hit Ratio in percent.
CDU	Total Cache Disk Usage in percent.
ORT	Operational Readiness Test. Uses the ORT script on the edge and mid servers to determine if the configuration in Traffic Ops matches the configuration on the edge or mid. The user that this script runs as must have an ssh key on the edge servers.

Daily Summary

Displays daily max gbps and bytes served for all CDNs. In order for the graphs to appear, the ‘daily_bw_url’ and ‘daily_served_url’ parameters need to be be created, assigned to the global profile, and have a value of a grafana graph. For more information on configuring grafana, see the Traffic Stats section.

Server

This view shows a table of all the servers in Traffic Ops. The table columns show the most important details of the server. The IPAddrr column is clickable to launch an ssh:// link to this server. The icon will link to a Traffic Stats graph of this server for caches, and the will link to the server status pages for other server types.

Server Types

These are the types of servers that can be managed in Traffic Ops:

Name	Description
EDGE	Edge Cache
MID	Mid Tier Cache
ORG	Origin
CCR	Traffic Router
RASCAL	Rascal health polling & reporting
TOOLS_SERVER	Ops hosts for managment
RIAK	Riak keystore
SPLUNK	SPLUNK indexer search head etc
TRAFFIC_STATS	traffic_stats server
INFLUXDB	influxDb server

Bulk Upload Server

TBD

Delivery Service

The fields in the Delivery Service view are:

Name	Description
Active	When this is set to false Traffic Router will not serve DNS or HTTP responses for this delivery service.
Cache URL expression	Cache URL rule to apply to this delivery service. See ATS documentation on cacheurl. [1]
CDN	The CDN in which the delivery service belongs to.
Check Path	A path (ex: /crossdomain.xml) to verify the connection to the origin server with. This can be used by Check Extension scripts to do periodic health checks against the delivery service.
Content Routing Type	The type of content routing this delivery service will use. See Delivery Service Types.
Deep Caching	(for HTTP routed delivery services only) When to do deep caching for this delivery service (see Deep Caching for more details): NEVER ALWAYS
Delivery Service DNS TTL	The Time To Live on the DNS record for the Traffic Router A and AAAA records (<routing-name>.<deliveryservice>.<cdn-domain>).
Delivery Servics URLs	(Read Only) An example of how the delivery URL may start. This could be multiple rows if multiple HOST_REGEXP entries have been entered.
Display Name	A human-readable name for the delivery service.
DNS Bypass CNAME	(For DNS routed delivery services only) This is a CNAME to respond to DNS requests with when a Delivery Service is unavailable.
DNS Bypass Ip	(For DNS routed delivery services only) This is the address to respond to A requests with when the max Bps or Max Tps for this delivery service are exceeded.
DNS Bypass IPv6	(For DNS routed delivery services only) This is the address to respond to AAAA requests with when the max Bps or Max Tps for this delivery service are exceeded.
DNS Bypass TTL	(For DNS routed delivery services only) This is the TTL to use for DNS Bypass responses when a delivery service is unavailable.
DSCP Tag	The DSCP value to mark IP packets to the client with.
Edge Header Rewrite Rules	Header Rewrite rules to apply for this delivery service at the EDGE tier. See Header Rewrite Options and DSCP. [1]
Geo Limit?	Some services are intended to be limited by geography. The possible settings are are: None - Do not limit by geography. CZF only - If the requesting IP is not in the Coverage Zone File, do not serve the request. CZF + US - If the requesting IP is not in the Coverage Zone File or not in the United States, do not serve the request.
Geo Limit Redirect URL	(for HTTP routed delivery services only) This is the URL Traffic Router will redirect to when Geo Limit Failure. See GeoLimit Failure Redirect feature
Geo Limit Countries	A comma delimited list of Country Codes that will be allowed to make requests from the Delivery Service
Geo Miss Default Latitude	Default Latitude for this delivery service. When client localization fails for both Coverage Zone and Geo Lookup, the client will be routed as if it was at this lat.
Geo Miss Default Longitude	Default Longitude for this delivery service. When client localization fails for both Coverage Zone and Geo Lookup, the client will be routed as if it was at this long.
Geolocation Provider	An ordered list of Geolocation Providers to be used in make geo location decisions.
Global Max Mbps	The maximum bits per second this delivery service can serve across all EDGE caches before traffic will be diverted to the bypass destination. For a DNS delivery service, the Bypass Ipv4 or Ipv6 will be used (depending on whether this was a A or AAAA request), and for HTTP delivery services the Bypass FQDN will be used.
Global Max TPS	The maximum transactions this delivery service can serve across all EDGE caches before traffic will be diverted to the bypass destination. For a DNS delivery service, the Bypass Ipv4 or Ipv6 will be used (depending on whether this was a A or AAAA request), and for HTTP delivery services the Bypass FQDN will be used.
HTTP Bypass FQDN	(for HTTP routed delivery services only) This is the FQDN Traffic Router will redirect to (with the same path) when the max Bps or Max Tps for this deliveryservice are exceeded.
Info URL	Info URL for this delivery service. To be consumed from the APIs by downstream tools (Portal).
Initial Dispersion	For HTTP delivery services - the number of caches that will be used by Traffic Router. Traffic Router will consistently return the same caches as long as they are healthy.
IPv6 Routing Enabled	When set to yes, the Traffic Router will respond to AAAA DNS requests for the routed name of this delivery service. Otherwise, only A records will be served.
Key (XML ID)	A unique string that identifies this delivery service.
Logs Enabled?	Whether or not Traffic Logs is enabled for the delivery service. This feature is currently incomplete without Traffic Logs
Long Description	Long description for this delivery service. To be consumed from the APIs by downstream tools (Portal).
Long Description 2	Another Long Description
Long Description 3	Another Long Description
Max DNS Answers	For DNS delivery services - the max number of DNS answers that will be returned by Traffic Router.
Mid Header Rewrite Rules	Header Rewrite rules to apply for this delivery service at the MID tier. See Header Rewrite Options and DSCP. [1]
Multi Site Origin	Enable the Multi Site Origin feature for this delivery service. See Multi Site Origin
Origin Server Base URL	The Origin Server’s base URL. This includes the protocol (http or https). Example: http://movies.origin.com
Origin Shield (Pipe Delimited String)	Origin Shield string.
Profile	The profile for this delivery service.
Protocol	The protocol to serve this delivery service to the clients with: 0 http 1 https 2 both http and https
Query String Handling	How to treat query strings: 0 use in cache key and hand up to origin - this means each unique query string is treated as a unique URL. 1 Do not use in cache key, but pass up to origin - this means 2 URLs that are the same except for the query string will match, and cache HIT, while the origin still sees original query string in the request. 2 Drop at edge - this means 2 URLs that are the same except for the query string will match, and cache HIT, while the origin will not see original query string in the request. Note: Choosing to drop query strings at the edge will preclude the use of a Regex Remap Expression. See Regex Remap Expression. To set the qstring without the use of regex remap, or for further options, see Qstring Handling.
Range Request Handling	How to treat range requests: 0 Do not cache (ranges requested from files that are already cached due to a non range request will be a HIT) 1 Use the background_fetch plugin. 2 Use the cache_range_requests plugin.
Raw Remap Text	For HTTP and DNS deliveryservices, this will get added to the end of the remap line on the cache verbatim. For ANY_MAP deliveryservices this is the remap line. [1]
Regex Remap Expression	Regex Remap rule to apply to this delivery service at the Edge tier. See ATS documentation on regex_remap. [1] Note: you will not be able to save a Regex Remap Expression if you have Query String Handling set to drop query strings at the edge. See Regex Remap Expression.
Regional Geo-Blocking	See documentation here
Routing Name	The routing name to use for the delivery FQDN, i.e. <routing-name>.<deliveryservice>.<cdn-domain>. It must be a valid hostname without periods. [2]
Signing Algorithm
SSL Key Version	The most recent version for SSL keys, this is incremented each time new certificates or generated or pasted in traffic ops
Tenant	By selecting a tenant for the delivery service, you restrict access of that delivery service to only those users that belong to the same tenant or a parent tenant.
Traffic Router Request Headers	Additional request header to be logged by Traffic Router
Traffic Router Response Headers	Reponse headers to be returned by Traffic Router for each request

[1]	(1, 2, 3, 4, 5) These fields are not validated by Traffic Ops to be correct syntactically, and can cause Traffic Server to not start if invalid. Please use with caution.

[2]

It is not recommended to change the Routing Name of a Delivery Service after deployment because this changes its Delivery FQDN (i.e. <routing-name>.<deliveryservice>.<cdn-domain>), which means that SSL certificates may need to be updated and clients using the Delivery Service will need to be transitioned to the new Delivery URL.

Delivery Service Types

One of the most important settings when creating the delivery service is the selection of the delivery service type. This type determines the routing method and the primary storage for the delivery service.

Name	Description
HTTP	HTTP Content Routing - The Traffic Router DNS auth server returns its own IP address on DNS queries, and the client gets redirected to a specific cache in the nearest cache group using HTTP 302. Use this for long sessions like HLS/HDS/Smooth live streaming, where a longer setup time is not a. problem.
DNS	DNS Content Routing - The Traffic Router DNS auth server returns an edge cache IP address to the client right away. The client will find the cache quickly but the Traffic Router can not route to a cache that already has this content in the cache group. Use this for smaller objects like web page images / objects.
HTTP_NO_CACHE	HTTP Content Routing, but the caches will not actually cache the content, they act as just proxies. The MID tier is bypassed.
HTTP_LIVE	HTTP Content routing, but where for “standard” HTTP content routing the objects are stored on disk, for this delivery service type the objects are stored on the RAM disks. Use this for linear TV. The MID tier is bypassed for this type.
HTTP_LIVE_NATNL	HTTP Content routing, same as HTTP_LIVE, but the MID tier is NOT bypassed.
DNS_LIVE_NATNL	DNS Content routing, but where for “standard” DNS content routing the objects are stored on disk, for this delivery service type the objects are stored on the RAM disks. Use this for linear TV. The MID tier is NOT bypassed for this type.
DNS_LIVE	DNS Content routing, same as DNS_LIVE_NATNL, but the MID tier is bypassed.
ANY_MAP	ANY_MAP is not known to Traffic Router. For this deliveryservice, the “Raw remap text” field in the input form will be used as the remap line on the cache.
STEERING	The Delivery Service will be used to route to other delivery services. The target delivery services and the routing weights for those delivery services will be defined by an admin or steering user. For more information see the steering feature documentation
CLIENT_STEERING	Simliar to STEERING except that a client can send a request to Traffic Router with a query param of trred=false and Traffic Router will return a HTTP 200 reponse with a body that contians a list of Delivery Service URLs that the client connect to. Therefore the client is doing the steering, not the Traffic Router.

Note

Once created, the Traffic Ops user interface does not allow you to change the delivery service type; the drop down is greyed out. There are many things that can go wrong when changing the type, and it is safer to delete the delivery service, and recreate it.

Federations

Federations allow for other (federated) CDNs (at a different ISP, MSO, etc) to add a list of resolvers and a CNAME to a delivery service Traffic Ops. When a request is made from one of federated CDN’s clients, Traffic Router will return the CNAME configured in the federation mapping. This allows the federated CDN to serve the content without the content provider changing the URL, or having to manage multiple URLs.

Before adding a federation in the Traffic Ops UI, a user with the federations role needs to be created. This user will be assigned to the federation and will be able to add resolvers to the federation via the Traffic Ops Federation API.

Header Rewrite Options and DSCP

Most header manipulation and per-delivery service configuration overrides are done using the ATS Header Rewrite Plugin. Traffic Control allows you to enter header rewrite rules to be applied at the edge and at the mid level. The syntax used in Traffic Ops is the same as the one described in the ATS documentation, except for some special strings that will get replaced:

Traffic Ops Entry	Gets Replaced with
__RETURN__	A newline
__CACHE_IPV4__	The cache’s IPv4 address

The deliveryservice screen also allows you to set the DSCP value of traffic sent to the client. This setting also results in a header_rewrite rule to be generated and applied to at the edge.

Note

The DSCP setting in the UI is only for setting traffic towards the client, and gets applied after the initial TCP handshake is complete, and the HTTP request is received (before that the cache can’t determine what deliveryservice this request is for, and what DSCP to apply), so the DSCP feature can not be used for security settings - the TCP SYN-ACK is not going to be DSCP marked.

Token Based Authentication

Token based authentication or signed URLs is implemented using the Traffic Server url_sig plugin. To sign a URL at the signing portal take the full URL, without any query string, and add on a query string with the following parameters:

Client IP address

The client IP address that this signature is valid for.

C=<client IP address>

Expiration

The Expiration time (seconds since epoch) of this signature.

E=<expiration time in secs since unix epoch>

Algorithm

The Algorithm used to create the signature. Only 1 (HMAC_SHA1) and 2 (HMAC_MD5) are supported at this time

A=<algorithm number>

Key index

Index of the key used. This is the index of the key in the configuration file on the cache. The set of keys is a shared secret between the signing portal and the edge caches. There is one set of keys per reverse proxy domain (fqdn).

K=<key index used>

Parts

Parts to use for the signature, always excluding the scheme (http://). parts0 = fqdn, parts1..x is the directory parts of the path, if there are more parts to the path than letters in the parts param, the last one is repeated for those. Examples:

1: use fqdn and all of URl path 0110: use part1 and part 2 of path only 01: use everything except the fqdn

P=<parts string (0's and 1's)>

Signature

The signature over the parts + the query string up to and including “S=”.

S=<signature>

See also

The url_sig README.

Generate URL Sig Keys

To generate a set of random signed url keys for this delivery service and store them in Traffic Vault, click the Generate URL Sig Keys button at the bottom of the delivery service details screen.

Parent Selection

Parameters in the Edge (child) profile that influence this feature:

Name	Filename	Default	Description
CONFIG proxy.config. http.parent_proxy_routing_enable	records.config	INT 1	enable parent selection. This is a required setting.
CONFIG proxy.config. url_remap.remap_required	records.config	INT 1	required for parent selection.
CONFIG proxy.config. http.no_dns_just_forward_to_parent	records.config	INT 0	See
CONFIG proxy.config. http.uncacheable_requests_bypass_parent	records.config	INT 1
CONFIG proxy.config. http.parent_proxy_routing_enable	records.config	INT 1
CONFIG proxy.config. http.parent_proxy.retry_time	records.config	INT 300
CONFIG proxy.config. http.parent_proxy.fail_threshold	records.config	INT 10
CONFIG proxy.config. http.parent_proxy.total_connect_attempts	records.config	INT 4
CONFIG proxy.config. http.parent_proxy.per_parent_connect_attempts	records.config	INT 2
CONFIG proxy.config. http.parent_proxy.connect_attempts_timeout	records.config	INT 30
CONFIG proxy.config. http.forward.proxy_auth_to_parent	records.config	INT 0
CONFIG proxy.config. http.parent_proxy_routing_enable	records.config	INT 0
CONFIG proxy.config. http.parent_proxy.file	records.config	STRING parent.config
CONFIG proxy.config. http.parent_proxy.connect_attempts_timeout	records.config	INT 3
algorithm	parent.config	urlhash	The algorithm to use.

Parameters in the Mid (parent) profile that influence this feature:

Name	Filename	Default	Description
domain_name	CRConfig.json		Only parents with the same value as the edge are going to be used as parents (to keep separation between CDNs)
weight	parent.config	1.0	The weight of this parent, translates to the number of replicas in the consistent hash ring. This parameter only has effect with algorithm at the client set to “consistent_hash”
port	parent.config	80	The port this parent is listening on as a forward proxy.
use_ip_address	parent.config	0	1 means use IP(v4) address of this parent in the parent.config, 0 means use the host_name.domain_name concatenation.

Qstring Handling

Delivery services have a Query String Handling option that, when set to ignore, will automatically add a regex remap to that delivery service’s config. There may be times this is not preferred, or there may be requirements for one delivery service or server(s) to behave differently. When this is required, the psel.qstring_handling parameter can be set in either the delivery service profile or the server profile, but it is important to note that the server profile will override ALL delivery services assigned to servers with this profile parameter. If the parameter is not set for the server profile but is present for the Delivery Service profile, this will override the setting in the delivery service. A value of “ignore” will not result in the addition of regex remap configuration.

Name	Filename	Default	Description
psel.qstring_handling	parent.config		Sets qstring handling without the use of regex remap for a delivery service when assigned to a delivery service profile, and overrides qstring handling for all delivery services for associated servers when assigned to a server profile. Value must be “consider” or “ignore”.

Multi Site Origin

Note

The configuration of this feature changed significantly between ATS version 5 and >= 6. Some configuration in Traffic Control is different as well. This documentation assumes ATS 6 or higher. See Configure Multi Site Origin for more details.

Normally, the mid servers are not aware of any redundancy at the origin layer. With Multi Site Origin enabled this changes - Traffic Server (and Traffic Ops) are now made aware of the fact there are multiple origins, and can be configured to do more advanced failover and loadbalancing actions. A prerequisite for MSO to work is that the multiple origin sites serve identical content with identical paths, and both are configured to serve the same origin hostname as is configured in the deliveryservice Origin Server Base URL field. See the Apache Traffic Server docs for more information on that cache’s implementation.

With This feature enabled, origin servers (or origin server VIP names for a site) are going to be entered as servers in to the Traiffic Ops UI. Server type is “ORG”.

Parameters in the mid profile that influence this feature:

Name	Filename	Default	Description
CONFIG proxy.config. http.parent_proxy_routing_enable	records.config	INT 1	enable parent selection. This is a required setting.
CONFIG proxy.config. url_remap.remap_required	records.config	INT 1	required for parent selection.

Parameters in the deliveryservice profile that influence this feature:

Name	Filename	Default	Description
mso.parent_retry	parent.config	-	Either simple_retry, dead_server_retry or both.
mso.algorithm	parent.config	consistent_hash	The algorithm to use. consisten_hash, strict, true, false, or latched. consisten_hash - spreads requests across multiple parents simultaneously based on hash of content URL. strict - strict Round Robin spreads requests across multiple parents simultaneously based on order of requests. true - same as strict, but ensures that requests from the same IP always go to the same parent if available. false - uses only a single parent at any given time and switches to a new parent only if the current parent fails. latched - same as false, but now, a failed parent will not be retried.
mso.unavailable_server_retry_response_codes	parent.config	“503”	Quoted, comma separated list of HTTP status codes that count as a unavailable_server_retry_response_code.
mso.max_unavailable_server_retries	parent.config	1	How many times an unavailable server will be retried.
mso.simple_retry_response_codes	parent.config	“404”	Quoted, comma separated list of HTTP status codes that count as a simple retry response code.
mso.max_simple_retries	parent.config	1	How many times a simple retry will be done.

see Configure Multi Site Origin for a quick how to on this feature.

Traffic Router Profile

Name	Config_file	Description
location	dns.zone	Location to store the DNS zone files in the local file system of Traffic Router.
location	http-log4j.properties	Location to find the log4j.properties file for Traffic Router.
location	dns-log4j.properties	Location to find the dns-log4j.properties file for Traffic Router.
location	geolocation.properties	Location to find the log4j.properties file for Traffic Router.
CDN_name	rascal-config.txt	The human readable name of the CDN for this profile.
CoverageZoneJsonURL	CRConfig.xml	The location (URL) to retrieve the coverage zone map file in JSON format from.
geolocation.polling.url	CRConfig.json	The location (URL) to retrieve the geo database file from.
geolocation.polling.interval	CRConfig.json	How often to refresh the coverage geo location database in ms
coveragezone.polling.interval	CRConfig.json	How often to refresh the coverage zone map in ms
coveragezone.polling.url	CRConfig.json	The location (URL) to retrieve the coverage zone map file in JSON format from.
deepcoveragezone.polling.interval	CRConfig.json	How often to refresh the deep coverage zone map in ms
deepcoveragezone.polling.url	CRConfig.json	The location (URL) to retrieve the deep coverage zone map file in JSON format from.
tld.soa.expire	CRConfig.json	The value for the expire field the Traffic Router DNS Server will respond with on Start of Authority (SOA) records.
tld.soa.minimum	CRConfig.json	The value for the minimum field the Traffic Router DNS Server will respond with on SOA records.
tld.soa.admin	CRConfig.json	The DNS Start of Authority admin. Should be a valid support email address for support if DNS is not working correctly.
tld.soa.retry	CRConfig.json	The value for the retry field the Traffic Router DNS Server will respond with on SOA records.
tld.soa.refresh	CRConfig.json	The TTL the Traffic Router DNS Server will respond with on A records.
tld.ttls.NS	CRConfig.json	The TTL the Traffic Router DNS Server will respond with on NS records.
tld.ttls.SOA	CRConfig.json	The TTL the Traffic Router DNS Server will respond with on SOA records.
tld.ttls.AAAA	CRConfig.json	The Time To Live (TTL) the Traffic Router DNS Server will respond with on AAAA records.
tld.ttls.A	CRConfig.json	The TTL the Traffic Router DNS Server will respond with on A records.
tld.ttls.DNSKEY	CRConfig.json	The TTL the Traffic Router DNS Server will respond with on DNSKEY records.
tld.ttls.DS	CRConfig.json	The TTL the Traffic Router DNS Server will respond with on DS records.
api.port	server.xml	The TCP port Traffic Router listens on for API (REST) access.
api.cache-control.max-age	CRConfig.json	The value of the Cache-Control: max-age= header in the API responses of Traffic Router.
api.auth.url	CRConfig.json	The API authentication URL (https://${tmHostname}/api/1.1/user/login); ${tmHostname} is a search and replace token used by Traffic Router to construct the correct URL)
consistent.dns.routing	CRConfig.json	Control whether DNS Delivery Services use consistent hashing on the edge FQDN to select caches for answers. May improve performance if set to true; defaults to false
dnssec.enabled	CRConfig.json	Whether DNSSEC is enabled; this parameter is updated via the DNSSEC administration user interface.
dnssec.allow.expired.keys	CRConfig.json	Allow Traffic Router to use expired DNSSEC keys to sign zones; default is true. This helps prevent DNSSEC related outages due to failed Traffic Control components or connectivity issues.
dynamic.cache.primer.enabled	CRConfig.json	Allow Traffic Router to attempt to prime the dynamic zone cache; defaults to true
dynamic.cache.primer.limit	CRConfig.json	Limit the number of permutations to prime when dynamic zone cache priming is enabled; defaults to 500
keystore.maintenance.interval	CRConfig.json	The interval in seconds which Traffic Router will check the keystore API for new DNSSEC keys
keystore.api.url	CRConfig.json	The keystore API URL (https://${tmHostname}/api/1.1/cdns/name/${cdnName}/dnsseckeys.json; ${tmHostname} and ${cdnName} are search and replace tokens used by Traffic Router to construct the correct URL)
keystore.fetch.timeout	CRConfig.json	The timeout in milliseconds for requests to the keystore API
keystore.fetch.retries	CRConfig.json	The number of times Traffic Router will attempt to load keys before giving up; defaults to 5
keystore.fetch.wait	CRConfig.json	The number of milliseconds Traffic Router will wait before a retry
signaturemanager.expiration.multiplier	CRConfig.json	Multiplier used in conjunction with a zone’s maximum TTL to calculate DNSSEC signature durations; defaults to 5
zonemanager.threadpool.scale	CRConfig.json	Multiplier used to determine the number of cores to use for zone signing operations; defaults to 0.75
zonemanager.cache.maintenance.interval	CRConfig.json	The interval in seconds which Traffic Router will check for zones that need to be resigned or if dynamic zones need to be expired from cache
zonemanager.dynamic.response.expiration	CRConfig.json	A string (e.g.: 300s) that defines how long a dynamic zone
DNSKEY.generation.multiplier	CRConfig.json	Used to deteremine when new keys need to be regenerated. Keys are regenerated if expiration is less than the generation multiplier * the TTL. If the parameter does not exist, the default is 10.
DNSKEY.effective.multiplier	CRConfig.json	Used when creating an effective date for a new key set. New keys are generated with an effective date of old key expiration - (effective multiplier * TTL). Default is 2.

Regex Remap Expression

The regex remap expression allows to to use a regex and resulting match group(s) in order to modify the request URIs that are sent to origin. For example:

^/original/(.*) http://origin.example.com/remapped/$1

Note

If Query String Handling is set to 2 Drop at edge, then you will not be allowed to save a regex remap expression, as dropping query strings actually relies on a regex remap of its own. However, if there is a need to both drop query strings and remap request URIs, this can be accomplished by setting Query String Handling to 1 Do not use in cache key, but pass up to origin, and then using a custom regex remap expression to do the necessary remapping, while simultaneously dropping query strings. The following example will capture the original request URI up to, but not including, the query string and then forward to a remapped URI:

^/([^?]*).* http://origin.example.com/remapped/$1

Delivery Service Regexp

This table defines how requests are matched to the delivery service. There are 3 type of entries possible here:

Name	Description	DS Type	Status
HOST_REGEXP	This is the regular expresion to match the host part of the URL.	DNS and HTTP	Supported
PATH_REGEXP	This is the regular expresion to match the path part of the URL.	HTTP	Beta
HEADER_REGEXP	This is the regular expresion to match on any header in the request.	HTTP	Beta

The Order entry defines the order in which the regular expressions get evaluated. To support CNAMES from domains outside of the Traffic Control top level DNS domain, enter multiple HOST_REGEXP lines.

Example:

Example foo.

Note

In most cases is is sufficient to have just one entry in this table that has a HOST_REGEXP Type, and Order 0. For the movies delivery service in the Kabletown CDN, the entry is simply single HOST_REGEXP set to .*\.movies\..*. This will match every url that has a hostname that ends with movies.cdn1.kabletown.net, since cdn1.kabletown.net is the Kabletown CDN’s DNS domain.

Static DNS Entries

Static DNS entries allow you to create other names under the delivery service domain. You can enter any valid hostname, and create a CNAME, A or AAAA record for it by clicking the Static DNS button at the bottom of the delivery service details screen.

Server Assignments

Click the Server Assignments button at the bottom of the screen to assign servers to this delivery service. Servers can be selected by drilling down in a tree, starting at the profile, then the cache group, and then the individual servers. Traffic Router will only route traffic for this delivery service to servers that are assigned to it.

The Coverage Zone File and ASN Table

The Coverage Zone File (CZF) should contain a cachegroup name to network prefix mapping in the form:

{
  "coverageZones": {
    "cache-group-01": {
      "coordinates": {
        "latitude":  1.1,
        "longitude": 2.2
      },
      "network6": [
        "1234:5678::/64",
        "1234:5679::/64"
      ],
      "network": [
        "192.168.8.0/24",
        "192.168.9.0/24"
      ]
    },
    "cache-group-02": {
      "coordinates": {
        "latitude":  3.3,
        "longitude": 4.4
      },
      "network6": [
        "1234:567a::/64",
        "1234:567b::/64"
      ],
      "network": [
        "192.168.4.0/24",
        "192.168.5.0/24"
      ]
    }
  }
}

The CZF is an input to the Traffic Control CDN, and as such does not get generated by Traffic Ops, but rather, it gets consumed by Traffic Router. Some popular IP management systems output a very similar file to the CZF but in stead of a cachegroup an ASN will be listed. Traffic Ops has the “Networks (ASNs)” view to aid with the conversion of files like that to a Traffic Control CZF file; this table is not used anywhere in Traffic Ops, but can be used to script the conversion using the API.

The script that generates the CZF file is not part of Traffic Control, since it is different for each situation.

Note

The "coordinates" section is optional and may be used by Traffic Router for localization in the case of a CZF “hit” where the zone name does not map to a Cache Group name in Traffic Ops (i.e. Traffic Router will route to the closest Cache Group(s) geographically).

The Deep Coverage Zone File

The Deep Coverage Zone File (DCZF) format is similar to the CZF format but adds a caches list under each deepCoverageZone:

{
  "deepCoverageZones": {
    "location-01": {
      "coordinates": {
        "latitude":  5.5,
        "longitude": 6.6
      },
      "network6": [
        "1234:5678::/64",
        "1234:5679::/64"
      ],
      "network": [
        "192.168.8.0/24",
        "192.168.9.0/24"
      ],
      "caches": [
        "edge-01",
        "edge-02"
      ]
    },
    "location-02": {
      "coordinates": {
        "latitude":  7.7,
        "longitude": 8.8
      },
      "network6": [
        "1234:567a::/64",
        "1234:567b::/64"
      ],
      "network": [
        "192.168.4.0/24",
        "192.168.5.0/24"
      ],
      "caches": [
        "edge-02",
        "edge-03"
      ]
    }
  }
}

Each entry in the caches list is the hostname of an edge cache registered in Traffic Ops which will be used for “deep” caching in that Deep Coverage Zone. Unlike a regular CZF, coverage zones in the DCZF do not map to a Cache Group in Traffic Ops, so currently the deep coverage zone name only needs to be unique.

If the Traffic Router gets a DCZF “hit” for a requested Delivery Service that has Deep Caching enabled, the client will be routed to an available “deep” cache from that zone’s caches list.

Note

The "coordinates" section is optional.

Parameters and Profiles

Parameters are shared between profiles if the set of { name, config_file, value } is the same. To change a value in one profile but not in others, the parameter has to be removed from the profile you want to change it in, and a new parameter entry has to be created (Add Parameter button at the bottom of the Parameters view), and assigned to that profile. It is easy to create new profiles from the Misc > Profiles view - just use the Add/Copy Profile button at the bottom of the profile view to copy an existing profile to a new one. Profiles can be exported from one system and imported to another using the profile view as well. It makes no sense for a parameter to not be assigned to a single profile - in that case it really has no function. To find parameters like that use the Parameters > Orphaned Parameters view. It is easy to create orphaned parameters by removing all profiles, or not assigning a profile directly after creating the parameter.

See also

Profile Parameters in the Configuring Traffic Ops section.

Tools

Generate ISO

Generate ISO is a tool for building custom ISOs for building caches on remote hosts. Currently it only supports Centos 6, but if you’re brave and pure of heart you MIGHT be able to get it to work with other unix-like OS’s.

The interface is mostly self explainatory as it’s got hints.

Field	Explaination
Choose a server from list:	This option gets all the server names currently in the Traffic Ops database and will autofill known values.
OS Version:	There needs to be an _osversions.cfg_ file in the ISO directory that maps the name of a directory to a name that shows up here.
Hostname:	This is the FQDN of the server to be installed. It is required.
Root password:	If you don’t put anything here it will default to the salted MD5 of “Fred”. Whatever put is MD5 hashed and writte to disk.
DHCP:	if yes, other IP settings will be ignored
IP Address:	Required if DHCP=no
Netmask:	Required if DHCP=no
Gateway:	Required if DHCP=no
IPV6 Address:	Optional. /64 is assumed if prefix is omitted
IPV6 Gateway:	Ignored if an IPV4 gateway is specified
Network Device:	Optional. Typical values are bond0, eth4, etc. Note: if you enter bond0, a LACP bonding config will be written
MTU:	If unsure, set to 1500
Specify disk for OS install:	Optional. Typical values are “sda”.

When you click the Download ISO button the folling occurs (all paths relative to the top level of the directory specified in _osversions.cfg_):

1. Reads /etc/resolv.conf to get a list of nameservers. This is a rather ugly hack that is in place until we get a way of configuring it in the interface.
2. Writes a file in the ks_scripts/state.out that contains directory from _osversions.cfg_ and the mkisofs string that we’ll call later.
3. Writes a file in the ks_scripts/network.cfg that is a bunch of key=value pairs that set up networking.
4. Creates an MD5 hash of the password you specify and writes it to ks_scripts/password.cfg. Note that if you do not specify a password “Fred” is used. Also note that we have experienced some issues with webbrowsers autofilling that field.
5. Writes out a disk configuration file to ks_scripts/disk.cfg.
6. mkisofs is called against the directory configured in _osversions.cfg_ and an ISO is generated in memory and delivered to your webbrowser.

You now have a customized ISO that can be used to install Red Hat and derivative Linux installations with some modifications to your ks.cfg file.

Kickstart/Anaconda will mount the ISO at /mnt/stage2 during the install process (at least with 6).

You can directly include the password file anywhere in your ks.cfg file (usually in the top) by doing %include /mnt/stage2/ks_scripts/password.cfg

What we currently do is have 2 scripts, one to do hard drive configuration and one to do network configuration. Both are relatively specific to the environment they were created in, and both are probably wrong for other organizations, however they are currently living in the “misc” directory as examples of how to do things.

We trigger those in a %pre section in ks.cfg and they will write config files to /tmp. We will then include those files in the appropriate places using %pre.

For example this is a section of our ks.cfg file:

%include /mnt/stage2/ks_scripts/packages.txt

%pre
  python /mnt/stage2/ks_scripts/create_network_line.py
  bash /mnt/stage2/ks_scripts/drive_config.sh
%end

These two scripts will then run _before_ anaconda sets up it’s internal structures, then a bit further up in the ks.cfg file (outside of the %pre %end block) we do an

%include /mnt/stage2/ks_scripts/password.cfg
...
%include /tmp/network_line

%include /tmp/drive_config
...

This snarfs up the contents and inlines them.

If you only have one kind of hardware on your CDN it is probably best to just put the drive config right in the ks.cfg.

If you have simple networking needs (we use bonded interfaces in most, but not all locations and we have several types of hardware meaning different ethernet interface names at the OS level etc.) then something like this:

#!/bin/bash
source /mnt/stage2/ks_scripts/network.cfg
echo "network --bootproto=static --activate --ipv6=$IPV6ADDR --ip=$IPADDR --netmask=$NETMASK --gateway=$GATEWAY --ipv6gateway=$GATEWAY --nameserver=$NAMESERVER --mtu=$MTU --hostname=$HOSTNAME" >> /tmp/network.cfg
# Note that this is an example and may not work at all.

You could also put this in the %pre section. Lots of ways to solve it.

We have included the two scripts we use in the “misc” directory of the git repo:

• kickstart_create_network_line.py
• kickstart_drive_config.sh

These scripts were written to support a very narrow set of expectations and environment and are almost certainly not suitable to just drop in, but they might provide a good starting point.

Queue Updates and Snapshot CRConfig

When changing delivery services special care has to be taken so that Traffic Router will not send traffic to caches for delivery services that the cache doesn’t know about yet. In general, when adding delivery services, or adding servers to a delivery service, it is best to update the caches before updating Traffic Router and Traffic Monitor. When deleting delivery services, or deleting server assignments to delivery services, it is best to update Traffic Router and Traffic Monitor first and then the caches. Updating the cache configuration is done through the Queue Updates menu, and updating Traffic Monitor and Traffic Router config is done through the Snapshot CRConfig menu.

Queue Updates

Every 15 minutes the caches should run a syncds to get all changes needed from Traffic Ops. The files that will be updated by the syncds job are:

• records.config
• remap.config
• parent.config
• cache.config
• hosting.config
• url_sig_(.*).config
• hdr_rw_(.*).config
• regex_revalidate.config
• ip_allow.config

A cache will only get updated when the update flag is set for it. To set the update flag, use the Queue Updates menu - here you can schedule updates for a whole CDN or a cache group:

1. Click Tools > Queue Updates.
2. Select the CDN to queueu uodates for, or All.
3. Select the cache group to queue updates for, or All
4. Click the Queue Updates button.
5. When the Queue Updates for this Server? (all) window opens, click OK.

To schedule updates for just one cache, use the “Server Checks” page, and click the in the UPD column. The UPD column of Server Checks page will change show a when updates are pending for that cache.

Snapshot CRConfig

Every 60 seconds Traffic Monitor will check with Traffic Ops to see if a new CRConfig snapshot exists; Traffic Monitor polls Traffic Ops for a new CRConfig, and Traffic Router polls Traffic Monitor for the same file. This is necessary to ensure that Traffic Monitor sees configuration changes first, which helps to ensure that the health and state of caches and delivery services propagates properly to Traffic Router. See Traffic Router Profile for more information on the CRConfig file.

To create a new snapshot, use the Tools > Snapshot CRConfig menu:

1. Click Tools > Snapshot CRConfig.

2. Verify the selection of the correct CDN from the Choose CDN drop down and click Diff CRConfig. On initial selection of this, the CRConfig Diff window says the following:

   There is no existing CRConfig for [cdn] to diff against… Is this the first snapshot??? If you are not sure why you are getting this message, please do not proceed! To proceed writing the snapshot anyway click the ‘Write CRConfig’ button below.

   If there is an older version of the CRConfig, a window will pop up showing the differences between the active CRConfig and the CRConfig about to be written.

3. Click Write CRConfig.

4. When the This will push out a new CRConfig.json. Are you sure? window opens, click OK.

5. The Successfully wrote CRConfig.json! window opens, click OK.

Invalidate Content

Invalidating content on the CDN is sometimes necessary when the origin was mis-configured and something is cached in the CDN that needs to be removed. Given the size of a typical Traffic Control CDN and the amount of content that can be cached in it, removing the content from all the caches may take a long time. To speed up content invalidation, Traffic Ops will not try to remove the content from the caches, but it makes the content inaccessible using the regex_revalidate ATS plugin. This forces a revalidation of the content, rather than a new get.

Note

This method forces a HTTP revalidation of the content, and not a new GET - the origin needs to support revalidation according to the HTTP/1.1 specification, and send a 200 OK or 304 Not Modified as applicable.

To invalidate content:

1. Click Tools > Invalidate Content
2. Fill out the form fields:

• Select the Delivery Service
• Enter the Path Regex - this should be a PCRE compatible regular expression for the path to match for forcing the revalidation. Be careful to only match on the content you need to remove - revalidation is an expensive operation for many origins, and a simple /.* can cause an overload condition of the origin.
• Enter the Time To Live - this is how long the revalidation rule will be active for. It usually makes sense to make this the same as the Cache-Control header from the origin which sets the object time to live in cache (by max-age or Expires). Entering a longer TTL here will make the caches do unnecessary work.
• Enter the Start Time - this is the start time when the revalidation rule will be made active. It is pre-populated with the current time, leave as is to schedule ASAP.

3. Click the Submit button.

Manage DNSSEC Keys

In order to support DNSSEC in Traffic Router, Traffic Ops provides some actions for managing DNSSEC keys for a CDN and associated Delivery Services. DNSSEC Keys consist of a Key Signing Keys (KSK) which are used to sign other DNSKEY records as well as Zone Signing Keys (ZSK) which are used to sign other records. DNSSEC Keys are stored in Traffic Vault and should only be accessible to Traffic Ops. Other applications needing access to this data, such as Traffic Router, must use the Traffic Ops DNSSEC APIs to retrieve this information.

To Manage DNSSEC Keys:

1. Click Tools -> Manage DNSSEC Keys
2. Choose a CDN and click Manage DNSSEC Keys

• If keys have not yet been generated for a CDN, this screen will be mostly blank with just the CDN and DNSSEC Active? fields being populated.
• If keys have been generated for the CDN, the Manage DNSSEC Keys screen will show the TTL and Top Level Domain (TLD) KSK Expiration for the CDN as well as DS Record information which will need to be added to the parent zone of the TLD in order for DNSSEC to work.

The Manage DNSSEC Keys screen also allows a user to perform the following actions:

Activate/Deactivate DNSSEC for a CDN

Fairly straight forward, this button set the dnssec.enabled param to either true or false on the Traffic Router profile for the CDN. The Activate/Deactivate option is only available if DNSSEC keys exist for CDN. In order to active DNSSEC for a CDN a user must first generate keys and then click the Active DNSSEC button.

Generate Keys

Generate Keys will generate DNSSEC keys for the CDN TLD as well as for each Delivery Service in the CDN. It is important to note that this button will create a new KSK for the TLD and, therefore, a new DS Record. Any time a new DS Record is created, it will need to be added to the parent zone of the TLD in order for DNSSEC to work properly. When a user clicks the Generate Keys button, they will be presented with a screen with the following fields:

• CDN: This is not editable and displays the CDN for which keys will be generated
• ZSK Expiration (Days): Sets how long (in days) the Zone Signing Key will be valid for the CDN and associated Delivery Services. The default is 30 days.
• KSK Expiration (Days): Sets how long (in days) the Key Signing Key will be valid for the CDN and associated Delivery Services. The default is 365 days.
• Effective Date (GMT): The time from which the new keys will be active. Traffic Router will use this value to determine when to start signing with the new keys and stop signing with the old keys.

Once these fields have been correctly entered, a user can click Generate Keys. The user will be presented with a confirmation screen to help them understand the impact of generating the keys. If a user confirms, the keys will be generated and stored in Traffic Vault.

Regenerate KSK

Regenerate KSK will create a new Key Signing Key for the CDN TLD. A new DS Record will also be generated and need to be put into the parent zone in order for DNSSEC to work correctly. The Regenerate KSK button is only available if keys have already been generated for a CDN. The intent of the button is to provide a mechanism for generating a new KSK when a previous one expires or if necessary for other reasons such as a security breach. When a user goes to generate a new KSK they are presented with a screen with the following options:

• CDN: This is not editable and displays the CDN for which keys will be generated
• KSK Expiration (Days): Sets how long (in days) the Key Signing Key will be valid for the CDN and associated Delivery Services. The default is 365 days.
• Effective Date (GMT): The time from which the new KSK and DS Record will be active. Since generating a new KSK will generate a new DS Record that needs to be added to the parent zone, it is very important to make sure that an effective date is chosen that allows for time to get the DS Record into the parent zone. Failure to get the new DS Record into the parent zone in time could result in DNSSEC errors when Traffic Router tries to sign responses.

Once these fields have been correctly entered, a user can click Generate KSK. The user will be presented with a confirmation screen to help them understand the impact of generating the KSK. If a user confirms, the KSK will be generated and stored in Traffic Vault.

Additionally, Traffic Ops also performs some systematic management of DNSSEC keys. This management is necessary to help keep keys in sync for Delivery Services in a CDN as well as to make sure keys do not expire without human intervention.

Generation of keys for new Delivery Services

If a new Delivery Service is created and added to a CDN that has DNSSEC enabled, Traffic Ops will create DNSSEC keys for the Delivery Service and store them in Traffic Vault.

Regeneration of expiring keys for a Delivery Service

Traffic Ops has a process, controlled by cron, to check for expired or expiring keys and re-generate them. The process runs at 5 minute intervals to check and see if keys are expired or close to expiring (withing 10 minutes by default). If keys are expired for a Delivery Service, traffic ops will regenerate new keys and store them in Traffic Vault. This process is the same for the CDN TLD ZSK, however Traffic Ops will not re-generate the CDN TLD KSK systematically. The reason is that when a KSK is regenerated for the CDN TLD then a new DS Record will also be created. The new DS Record needs to be added to the parent zone before Traffic Router attempts to sign with the new KSK in order for DNSSEC to work correctly. Therefore, management of the KSK needs to be a manual process.