The Valve Component

Table of Contents

Introduction

A Valve element represents a component that will be inserted into the request processing pipeline for the associated Catalina container (Engine, Host, or Context). Individual Valves have distinct processing capabilities, and are described individually below.

The description below uses the variable name $CATALINA_BASE to refer the base directory against which most relative paths are resolved. If you have not configured Tomcat for multiple instances by setting a CATALINA_BASE directory, then $CATALINA_BASE will be set to the value of $CATALINA_HOME, the directory into which you have installed Tomcat.

Access Logging

Access logging is performed by valves that implement org.apache.catalina.AccessLog interface.

Access Log Valve

Introduction

The Access Log Valve creates log files in the same format as those created by standard web servers. These logs can later be analyzed by standard log analysis tools to track page hit counts, user session activity, and so on. This Valve uses self-contained logic to write its log files, which can be automatically rolled over at midnight each day. (The essential requirement for access logging is to handle a large continuous stream of data with low overhead. This Valve does not use Apache Commons Logging, thus avoiding additional overhead and potentially complex configuration).

This Valve may be associated with any Catalina container (Context, Host, or Engine), and will record ALL requests processed by that container.

Some requests may be handled by Tomcat before they are passed to a container. These include redirects from /foo to /foo/ and the rejection of invalid requests. Where Tomcat can identify the Context that would have handled the request, the request/response will be logged in the AccessLog(s) associated Context, Host and Engine. Where Tomcat cannot identify the Context that would have handled the request, e.g. in cases where the URL is invalid, Tomcat will look first in the Engine, then the default Host for the Engine and finally the ROOT (or default) Context for the default Host for an AccessLog implementation. Tomcat will use the first AccessLog implementation found to log those requests that are rejected before they are passed to a container.

The output file will be placed in the directory given by the directory attribute. The name of the file is composed by concatenation of the configured prefix, timestamp and suffix. The format of the timestamp in the file name can be set using the fileDateFormat attribute. This timestamp will be omitted if the file rotation is switched off by setting rotatable to false.

Warning: If multiple AccessLogValve instances are used, they should be configured to use different output files.

If sendfile is used, the response bytes will be written asynchronously in a separate thread and the access log valve will not know how many bytes were actually written. In this case, the number of bytes that was passed to the sendfile thread for writing will be recorded in the access log valve.

Attributes

The Access Log Valve supports the following configuration attributes:

Attribute Description
className

Java class name of the implementation to use. This MUST be set to org.apache.catalina.valves.AccessLogValve to use the default access log valve.

directory

Absolute or relative pathname of a directory in which log files created by this valve will be placed. If a relative path is specified, it is interpreted as relative to $CATALINA_BASE. If no directory attribute is specified, the default value is "logs" (relative to $CATALINA_BASE).

prefix

The prefix added to the start of each log file's name. If not specified, the default value is "access_log".

suffix

The suffix added to the end of each log file's name. If not specified, the default value is "" (a zero-length string), meaning that no suffix will be added.

fileDateFormat

Allows a customized timestamp in the access log file name. The file is rotated whenever the formatted timestamp changes. The default value is .yyyy-MM-dd. If you wish to rotate every hour, then set this value to .yyyy-MM-dd.HH. The date format will always be localized using the locale en_US.

rotatable

Flag to determine if log rotation should occur. If set to false, then this file is never rotated and fileDateFormat is ignored. Default value: true

renameOnRotate

By default for a rotatable log the active access log file name will contain the current timestamp in fileDateFormat. During rotation the file is closed and a new file with the next timestamp in the name is created and used. When setting renameOnRotate to true, the timestamp is no longer part of the active log file name. Only during rotation the file is closed and then renamed to include the timestamp. This is similar to the behavior of most log frameworks when doing time based rotation. Default value: false

pattern

A formatting layout identifying the various information fields from the request and response to be logged, or the word common or combined to select a standard format. See below for more information on configuring this attribute.

encoding

Character set used to write the log file. An empty string means to use the system default character set. Default value: use the system default character set.

locale

The locale used to format timestamps in the access log lines. Any timestamps configured using an explicit SimpleDateFormat pattern (%{xxx}t) are formatted in this locale. By default the default locale of the Java process is used. Switching the locale after the AccessLogValve is initialized is not supported. Any timestamps using the common log format (CLF) are always formatted in the locale en_US.

requestAttributesEnabled

Set to true to check for the existence of request attributes (typically set by the RemoteIpValve and similar) that should be used to override the values returned by the request for remote address, remote host, server port and protocol. If the attributes are not set, or this attribute is set to false then the values from the request will be used. If not set, the default value of false will be used.

conditionIf

Turns on conditional logging. If set, requests will be logged only if ServletRequest.getAttribute() is not null. For example, if this value is set to important, then a particular request will only be logged if ServletRequest.getAttribute("important") != null. The use of Filters is an easy way to set/unset the attribute in the ServletRequest on many different requests.

conditionUnless

Turns on conditional logging. If set, requests will be logged only if ServletRequest.getAttribute() is null. For example, if this value is set to junk, then a particular request will only be logged if ServletRequest.getAttribute("junk") == null. The use of Filters is an easy way to set/unset the attribute in the ServletRequest on many different requests.

condition

The same as conditionUnless. This attribute is provided for backwards compatibility.

buffered

Flag to determine if logging will be buffered. If set to false, then access logging will be written after each request. Default value: true

maxLogMessageBufferSize

Log message buffers are usually recycled and re-used. To prevent excessive memory usage, if a buffer grows beyond this size it will be discarded. The default is 256 characters. This should be set to larger than the typical access log message size.

resolveHosts

This attribute is no longer supported. Use the connector attribute enableLookups instead.

If you have enableLookups on the connector set to true and want to ignore it, use %a instead of %h in the value of pattern.

Values for the pattern attribute are made up of literal text strings, combined with pattern identifiers prefixed by the "%" character to cause replacement by the corresponding variable value from the current request and response. The following pattern codes are supported:

  • %a - Remote IP address
  • %A - Local IP address
  • %b - Bytes sent, excluding HTTP headers, or '-' if zero
  • %B - Bytes sent, excluding HTTP headers
  • %h - Remote host name (or IP address if enableLookups for the connector is false)
  • %H - Request protocol
  • %l - Remote logical username from identd (always returns '-')
  • %m - Request method (GET, POST, etc.)
  • %p - Local port on which this request was received. See also %{xxx}p below.
  • %q - Query string (prepended with a '?' if it exists)
  • %r - First line of the request (method and request URI)
  • %s - HTTP status code of the response
  • %S - User session ID
  • %t - Date and time, in Common Log Format
  • %u - Remote user that was authenticated (if any), else '-'
  • %U - Requested URL path
  • %v - Local server name
  • %D - Time taken to process the request, in millis
  • %T - Time taken to process the request, in seconds
  • %F - Time taken to commit the response, in millis
  • %I - Current request thread name (can compare later with stacktraces)
  • %X - Connection status when response is completed:
    • X = Connection aborted before the response completed.
    • + = Connection may be kept alive after the response is sent.
    • - = Connection will be closed after the response is sent.

There is also support to write information incoming or outgoing headers, cookies, session or request attributes and special timestamp formats. It is modeled after the Apache HTTP Server log configuration syntax. Each of them can be used multiple times with different xxx keys:

  • %{xxx}i write value of incoming header with name xxx
  • %{xxx}o write value of outgoing header with name xxx
  • %{xxx}c write value of cookie with name xxx
  • %{xxx}r write value of ServletRequest attribute with name xxx
  • %{xxx}s write value of HttpSession attribute with name xxx
  • %{xxx}p write local (server) port (xxx==local) or remote (client) port (xxx=remote)
  • %{xxx}t write timestamp at the end of the request formatted using the enhanced SimpleDateFormat pattern xxx

All formats supported by SimpleDateFormat are allowed in %{xxx}t. In addition the following extensions have been added:

  • sec - number of seconds since the epoch
  • msec - number of milliseconds since the epoch
  • msec_frac - millisecond fraction

These formats cannot be mixed with SimpleDateFormat formats in the same format token.

Furthermore one can define whether to log the timestamp for the request start time or the response finish time:

  • begin or prefix begin: chooses the request start time
  • end or prefix end: chooses the response finish time

By adding multiple %{xxx}t tokens to the pattern, one can also log both timestamps.

The shorthand pattern pattern="common" corresponds to the Common Log Format defined by '%h %l %u %t "%r" %s %b'.

The shorthand pattern pattern="combined" appends the values of the Referer and User-Agent headers, each in double quotes, to the common pattern.

When Tomcat is operating behind a reverse proxy, the client information logged by the Access Log Valve may represent the reverse proxy, the browser or some combination of the two depending on the configuration of Tomcat and the reverse proxy. For Tomcat configuration options see Proxies Support and the Proxy How-To. For reverse proxies that use mod_jk, see the generic proxy documentation. For other reverse proxies, consult their documentation.

Extended Access Log Valve

Introduction

The Extended Access Log Valve extends the Access Log Valve class, and so uses the same self-contained logging logic. This means it implements many of the same file handling attributes. The main difference to the standard AccessLogValve is that ExtendedAccessLogValve creates log files which conform to the Working Draft for the Extended Log File Format defined by the W3C.

Attributes

The Extended Access Log Valve supports all configuration attributes of the standard Access Log Valve. Only the values used for className and pattern differ.

Attribute Description
className

Java class name of the implementation to use. This MUST be set to org.apache.catalina.valves.ExtendedAccessLogValve to use the extended access log valve.

pattern

A formatting layout identifying the various information fields from the request and response to be logged. See below for more information on configuring this attribute.

Values for the pattern attribute are made up of format tokens. Some of the tokens need an additional prefix. Possible prefixes are c for "client", s for "server", cs for "client to server", sc for "server to client" or x for "application specific". Furthermore some tokens are completed by an additional selector. See the W3C specification for more information about the format.

The following format tokens are supported:

  • bytes - Bytes sent, excluding HTTP headers, or '-' if zero
  • c-dns - Remote host name (or IP address if enableLookups for the connector is false)
  • c-ip - Remote IP address
  • cs-method - Request method (GET, POST, etc.)
  • cs-uri - Request URI
  • cs-uri-query - Query string (prepended with a '?' if it exists)
  • cs-uri-stem - Requested URL path
  • date - The date in yyyy-mm-dd format for GMT
  • s-dns - Local host name
  • s-ip - Local IP address
  • sc-status - HTTP status code of the response
  • time - Time the request was served in HH:mm:ss format for GMT
  • time-taken - Time (in seconds as floating point) taken to serve the request
  • x-threadname - Current request thread name (can compare later with stacktraces)

For any of the x-H(XXX) the following method will be called from the HttpServletRequest object:

  • x-H(authType): getAuthType
  • x-H(characterEncoding): getCharacterEncoding
  • x-H(contentLength): getContentLength
  • x-H(locale): getLocale
  • x-H(protocol): getProtocol
  • x-H(remoteUser): getRemoteUser
  • x-H(requestedSessionId): getRequestedSessionId
  • x-H(requestedSessionIdFromCookie): isRequestedSessionIdFromCookie
  • x-H(requestedSessionIdValid): isRequestedSessionIdValid
  • x-H(scheme): getScheme
  • x-H(secure): isSecure

There is also support to write information about headers cookies, context, request or session attributes and request parameters.

  • cs(XXX) for incoming request headers with name XXX
  • sc(XXX) for outgoing response headers with name XXX
  • x-A(XXX) for the servlet context attribute with name XXX
  • x-C(XXX) for the first cookie with name XXX
  • x-O(XXX) for a concatenation of all outgoing response headers with name XXX
  • x-P(XXX) for the URL encoded (using UTF-8) request parameter with name XXX
  • x-R(XXX) for the request attribute with name XXX
  • x-S(XXX) for the session attribute with name XXX

Access Control

Remote Address Valve

Introduction

The Remote Address Valve allows you to compare the IP address of the client that submitted this request against one or more regular expressions, and either allow the request to continue or refuse to process the request from this client. A Remote Address Valve can be associated with any Catalina container (Engine, Host, or Context), and must accept any request presented to this container for processing before it will be passed on.

The syntax for regular expressions is different than that for 'standard' wildcard matching. Tomcat uses the java.util.regex package. Please consult the Java documentation for details of the expressions supported.

Optionally one can append the server connector port separated with a semicolon (";") to allow different expressions for each connector.

The behavior when a request is refused can be changed to not deny but instead set an invalid authentication header. This is useful in combination with the context attribute preemptiveAuthentication="true".

Note: There is a caveat when using this valve with IPv6 addresses. Format of the IP address that this valve is processing depends on the API that was used to obtain it. If the address was obtained from Java socket using Inet6Address class, its format will be x:x:x:x:x:x:x:x. That is, the IP address for localhost will be 0:0:0:0:0:0:0:1 instead of the more widely used ::1. Consult your access logs for the actual value.

See also: Remote Host Valve, Remote IP Valve.

Attributes

The Remote Address Valve supports the following configuration attributes:

Attribute Description
className

Java class name of the implementation to use. This MUST be set to org.apache.catalina.valves.RemoteAddrValve.

allow

A regular expression (using java.util.regex) that the remote client's IP address is compared to. If this attribute is specified, the remote address MUST match for this request to be accepted. If this attribute is not specified, all requests will be accepted UNLESS the remote address matches a deny pattern.

deny

A regular expression (using java.util.regex) that the remote client's IP address is compared to. If this attribute is specified, the remote address MUST NOT match for this request to be accepted. If this attribute is not specified, request acceptance is governed solely by the allow attribute.

denyStatus

HTTP response status code that is used when rejecting denied request. The default value is 403. For example, it can be set to the value 404.

addConnectorPort

Append the server connector port to the client IP address separated with a semicolon (";"). If this is set to true, the expressions configured with allow and deny is compared against ADDRESS;PORT where ADDRESS is the client IP address and PORT is the Tomcat connector port which received the request. The default value is false.

invalidAuthenticationWhenDeny

When a request should be denied, do not deny but instead set an invalid authentication header. This only works if the context has the attribute preemptiveAuthentication="true" set. An already existing authentication header will not be overwritten. In effect this will trigger authentication instead of deny even if the application does not have a security constraint configured.

This can be combined with addConnectorPort to trigger authentication depending on the client and the connector that is used to access an application.

Example 1

To allow access only for the clients connecting from localhost:

<Valve className="org.apache.catalina.valves.RemoteAddrValve"
   allow="127\.\d+\.\d+\.\d+|::1|0:0:0:0:0:0:0:1"/>

Example 2

To allow unrestricted access for the clients connecting from localhost but for all other clients only to port 8443:

<Valve className="org.apache.catalina.valves.RemoteAddrValve"
   addConnectorPort="true"
   allow="127\.\d+\.\d+\.\d+;\d*|::1;\d*|0:0:0:0:0:0:0:1;\d*|.*;8443"/>

Example 3

To allow unrestricted access to port 8009, but trigger basic authentication if the application is accessed on another port:

<Context>
  ...
  <Valve className="org.apache.catalina.valves.RemoteAddrValve"
         addConnectorPort="true"
         invalidAuthenticationWhenDeny="true"
         allow=".*;8009"/>
  <Valve className="org.apache.catalina.authenticator.BasicAuthenticator" />
  ...
</Context>

Remote Host Valve

Introduction

The Remote Host Valve allows you to compare the hostname of the client that submitted this request against one or more regular expressions, and either allow the request to continue or refuse to process the request from this client. A Remote Host Valve can be associated with any Catalina container (Engine, Host, or Context), and must accept any request presented to this container for processing before it will be passed on.

The syntax for regular expressions is different than that for 'standard' wildcard matching. Tomcat uses the java.util.regex package. Please consult the Java documentation for details of the expressions supported.

Optionally one can append the server connector port separated with a semicolon (";") to allow different expressions for each connector.

The behavior when a request is refused can be changed to not deny but instead set an invalid authentication header. This is useful in combination with the context attribute preemptiveAuthentication="true".

Note: This filter processes the value returned by method ServletRequest.getRemoteHost(). To allow the method to return proper host names, you have to enable "DNS lookups" feature on a Connector.

See also: Remote Address Valve, HTTP Connector configuration.

Attributes

The Remote Host Valve supports the following configuration attributes:

Attribute Description
className

Java class name of the implementation to use. This MUST be set to org.apache.catalina.valves.RemoteHostValve.

allow

A regular expression (using java.util.regex) that the remote client's hostname is compared to. If this attribute is specified, the remote hostname MUST match for this request to be accepted. If this attribute is not specified, all requests will be accepted UNLESS the remote hostname matches a deny pattern.

deny

A regular expression (using java.util.regex) that the remote client's hostname is compared to. If this attribute is specified, the remote hostname MUST NOT match for this request to be accepted. If this attribute is not specified, request acceptance is governed solely by the allow attribute.

denyStatus

HTTP response status code that is used when rejecting denied request. The default value is 403. For example, it can be set to the value 404.

addConnectorPort

Append the server connector port to the client hostname separated with a semicolon (";"). If this is set to true, the expressions configured with allow and deny is compared against HOSTNAME;PORT where HOSTNAME is the client hostname and PORT is the Tomcat connector port which received the request. The default value is false.

invalidAuthenticationWhenDeny

When a request should be denied, do not deny but instead set an invalid authentication header. This only works if the context has the attribute preemptiveAuthentication="true" set. An already existing authentication header will not be overwritten. In effect this will trigger authentication instead of deny even if the application does not have a security constraint configured.

This can be combined with addConnectorPort to trigger authentication depending on the client and the connector that is used to access an application.

Proxies Support

Load Balancer Draining Valve

Introduction

When using mod_jk or mod_proxy_ajp, the client's session id is used to determine which back-end server will be used to serve the request. If the target node is being "drained" (in mod_jk, this is the DISABLED state; in mod_proxy_ajp, this is the Drain (N) state), requests for expired sessions can actually cause the draining node to fail to drain.

Unfortunately, AJP-based load-balancers cannot prove whether the client-provided session id is valid or not and therefore will send any requests for a session that appears to be targeted to that node to the disabled (or "draining") node, causing the "draining" process to take longer than necessary.

This Valve detects requests for invalid sessions, strips the session information from the request, and redirects back to the same URL, where the load-balancer should choose a different (active) node to handle the request. This will accelerate the "draining" process for the disabled node(s).

The activation state of the node is sent by the load-balancer in the request, so no state change on the node being disabled is necessary. Simply configure this Valve in your valve pipeline and it will take action when the activation state is set to "disabled".

You should take care to register this Valve earlier in the Valve pipeline than any authentication Valves, because this Valve should be able to redirect a request before any authentication Valve saves a request to a protected resource. If this happens, a new session will be created and the draining process will stall because a new, valid session will be established.

Attributes

The Load Balancer Draining Valve supports the following configuration attributes:

Attribute Description
className

Java class name of the implementation to use. This MUST be set to org.apache.catalina.valves.LoadBalancerDrainingValve.

redirectStatusCode

Allows setting a custom redirect code to be used when the client is redirected to be re-balanced by the load-balancer. The default is 307 TEMPORARY_REDIRECT.

ignoreCookieName

When used with ignoreCookieValue, a client can present this cookie (and accompanying value) that will cause this Valve to do nothing. This will allow you to probe your disabled node before re-enabling it to make sure that it is working as expected.

ignoreCookieValue

When used with ignoreCookieName, a client can present a cookie (and accompanying value) that will cause this Valve to do nothing. This will allow you to probe your disabled node before re-enabling it to make sure that it is working as expected.

Remote IP Valve

Introduction

Tomcat port of mod_remoteip, this valve replaces the apparent client remote IP address and hostname for the request with the IP address list presented by a proxy or a load balancer via a request headers (e.g. "X-Forwarded-For").

Another feature of this valve is to replace the apparent scheme (http/https), server port and request.secure with the scheme presented by a proxy or a load balancer via a request header (e.g. "X-Forwarded-Proto").

This Valve may be used at the Engine, Host or Context level as required. Normally, this Valve would be used at the Engine level.

If used in conjunction with Remote Address/Host valves then this valve should be defined first to ensure that the correct client IP address is presented to the Remote Address/Host valves.

Note: By default this valve has no effect on the values that are written into access log. The original values are restored when request processing leaves the valve and that always happens earlier than access logging. To pass the remote address, remote host, server port and protocol values set by this valve to the access log, they are put into request attributes. Publishing these values here is enabled by default, but AccessLogValve should be explicitly configured to use them. See documentation for requestAttributesEnabled attribute of AccessLogValve.

The names of request attributes that are set by this valve and can be used by access logging are the following:

  • org.apache.catalina.AccessLog.RemoteAddr
  • org.apache.catalina.AccessLog.RemoteHost
  • org.apache.catalina.AccessLog.Protocol
  • org.apache.catalina.AccessLog.ServerPort
  • org.apache.tomcat.remoteAddr

Attributes

The Remote IP Valve supports the following configuration attributes:

Attribute Description
className

Java class name of the implementation to use. This MUST be set to org.apache.catalina.valves.RemoteIpValve.

remoteIpHeader

Name of the HTTP Header read by this valve that holds the list of traversed IP addresses starting from the requesting client. If not specified, the default of x-forwarded-for is used.

internalProxies

Regular expression (using java.util.regex) that a proxy's IP address must match to be considered an internal proxy. Internal proxies that appear in the remoteIpHeader will be trusted and will not appear in the proxiesHeader value. If not specified the default value of 10\.\d{1,3}\.\d{1,3}\.\d{1,3}|192\.168\.\d{1,3}\.\d{1,3}|169\.254\.\d{1,3}\.\d{1,3}|127\.\d{1,3}\.\d{1,3}\.\d{1,3}|172\.1[6-9]{1}\.\d{1,3}\.\d{1,3}|172\.2[0-9]{1}\.\d{1,3}\.\d{1,3}|172\.3[0-1]{1}\.\d{1,3}\.\d{1,3} will be used.

proxiesHeader

Name of the HTTP header created by this valve to hold the list of proxies that have been processed in the incoming remoteIpHeader. If not specified, the default of x-forwarded-by is used.

requestAttributesEnabled

Set to true to set the request attributes used by AccessLog implementations to override the values returned by the request for remote address, remote host, server port and protocol. Request attributes are also used to enable the forwarded remote address to be displayed on the status page of the Manager web application. If not set, the default value of true will be used.

trustedProxies

Regular expression (using java.util.regex) that a proxy's IP address must match to be considered an trusted proxy. Trusted proxies that appear in the remoteIpHeader will be trusted and will appear in the proxiesHeader value. If not specified, no proxies will be trusted.

protocolHeader

Name of the HTTP Header read by this valve that holds the protocol used by the client to connect to the proxy. If not specified, the default of null is used.

portHeader

Name of the HTTP Header read by this valve that holds the port used by the client to connect to the proxy. If not specified, the default of null is used.

protocolHeaderHttpsValue

Value of the protocolHeader to indicate that it is an HTTPS request. If not specified, the default of https is used.

httpServerPort

Value returned by ServletRequest.getServerPort() when the protocolHeader indicates http protocol and no portHeader is present. If not specified, the default of 80 is used.

httpsServerPort

Value returned by ServletRequest.getServerPort() when the protocolHeader indicates https protocol and no portHeader is present. If not specified, the default of 443 is used.

changeLocalPort

If true, the value returned by ServletRequest.getLocalPort() and ServletRequest.getServerPort() is modified by the this valve. If not specified, the default of false is used.

SSL Valve

Introduction

When using mod_proxy_http, the client SSL information is not included in the protocol (unlike mod_jk and mod_proxy_ajp). To make the client SSL information available to Tomcat, some additional configuration is required. In httpd, mod_headers is used to add the SSL information as HTTP headers. In Tomcat, this valve is used to read the information from the HTTP headers and insert it into the request.

Note: Ensure that the headers are always set by httpd for all requests to prevent a client spoofing SSL information by sending fake headers.

To configure httpd to set the necessary headers, add the following:

<IfModule ssl_module>
  RequestHeader set SSL_CLIENT_CERT "%{SSL_CLIENT_CERT}s"
  RequestHeader set SSL_CIPHER "%{SSL_CIPHER}s"
  RequestHeader set SSL_SESSION_ID "%{SSL_SESSION_ID}s"
  RequestHeader set SSL_CIPHER_USEKEYSIZE "%{SSL_CIPHER_USEKEYSIZE}s"
</IfModule>

Attributes

The SSL Valve supports the following configuration attribute:

Attribute Description
className

Java class name of the implementation to use. This MUST be set to org.apache.catalina.valves.SSLValve.

sslClientCertHeader

Allows setting a custom name for the ssl_client_cert header. If not specified, the default of ssl_client_cert is used.

sslCipherHeader

Allows setting a custom name for the ssl_cipher header. If not specified, the default of ssl_cipher is used.

sslSessionIdHeader

Allows setting a custom name for the ssl_session_id header. If not specified, the default of ssl_session_id is used.

sslCipherUserKeySizeHeader

Allows setting a custom name for the ssl_cipher_usekeysize header. If not specified, the default of ssl_cipher_usekeysize is used.

Single Sign On Valve

Introduction

The Single Sign On Valve is utilized when you wish to give users the ability to sign on to any one of the web applications associated with your virtual host, and then have their identity recognized by all other web applications on the same virtual host.

See the Single Sign On special feature on the Host element for more information.

Attributes

The Single Sign On Valve supports the following configuration attributes:

Attribute Description
className

Java class name of the implementation to use. This MUST be set to org.apache.catalina.authenticator.SingleSignOn.

requireReauthentication

Default false. Flag to determine whether each request needs to be reauthenticated to the security Realm. If "true", this Valve uses cached security credentials (username and password) to reauthenticate to the Realm each request associated with an SSO session. If "false", the Valve can itself authenticate requests based on the presence of a valid SSO cookie, without rechecking with the Realm.

cookieDomain

Sets the host domain to be used for sso cookies.

Authentication

The valves in this section implement org.apache.catalina.Authenticator interface.

Basic Authenticator Valve

Introduction

The Basic Authenticator Valve is automatically added to any Context that is configured to use BASIC authentication.

If any non-default settings are required, the valve may be configured within Context element with the required values.

Attributes

The Basic Authenticator Valve supports the following configuration attributes:

Attribute Description
alwaysUseSession

Should a session always be used once a user is authenticated? This may offer some performance benefits since the session can then be used to cache the authenticated Principal, hence removing the need to authenticate the user via the Realm on every request. This may be of help for combinations such as BASIC authentication used with the JNDIRealm or DataSourceRealms. However there will also be the performance cost of creating and GC'ing the session. If not set, the default value of false will be used.

cache

Should we cache authenticated Principals if the request is part of an HTTP session? If not specified, the default value of true will be used.

changeSessionIdOnAuthentication

Controls if the session ID is changed if a session exists at the point where users are authenticated. This is to prevent session fixation attacks. If not set, the default value of true will be used.

charset

Controls if the WWW-Authenticate HTTP header includes a charset authentication parameter as per RFC 7617. The only permitted options are null, the empty string and UTF-8. If UTF-8 is specified then the charset authentication parameter will be sent with that value and the provided user name and optional password will be converted from bytes to characters using UTF-8. Otherwise, no charset authentication parameter will be sent and the provided user name and optional password will be converted from bytes to characters using ISO-8859-1. The default value is null

className

Java class name of the implementation to use. This MUST be set to org.apache.catalina.authenticator.BasicAuthenticator.

disableProxyCaching

Controls the caching of pages that are protected by security constraints. Setting this to false may help work around caching issues in some browsers but will also cause secured pages to be cached by proxies which will almost certainly be a security issue. securePagesWithPragma offers an alternative, secure, workaround for browser caching issues. If not set, the default value of true will be used.

securePagesWithPragma

Controls the caching of pages that are protected by security constraints. Setting this to false may help work around caching issues in some browsers by using Cache-Control: private rather than the default of Pragma: No-cache and Cache-control: No-cache. If not set, the default value of false will be used.

secureRandomAlgorithm

Name of the algorithm to use to create the java.security.SecureRandom instances that generate session IDs. If an invalid algorithm and/or provider is specified, the platform default provider and the default algorithm will be used. If not specified, the default algorithm of SHA1PRNG will be used. If the default algorithm is not supported, the platform default will be used. To specify that the platform default should be used, do not set the secureRandomProvider attribute and set this attribute to the empty string.

secureRandomClass

Name of the Java class that extends java.security.SecureRandom to use to generate SSO session IDs. If not specified, the default value is java.security.SecureRandom.

secureRandomProvider

Name of the provider to use to create the java.security.SecureRandom instances that generate SSO session IDs. If an invalid algorithm and/or provider is specified, the platform default provider and the default algorithm will be used. If not specified, the platform default provider will be used.

jaspicCallbackHandlerClass

Name of the Java class of the javax.security.auth.callback.CallbackHandler implementation which should be used by JASPIC. If none is specified the default org.apache.catalina.authenticator.jaspic.CallbackHandlerImpl will be used.

Digest Authenticator Valve

Introduction

The Digest Authenticator Valve is automatically added to any Context that is configured to use DIGEST authentication.

If any non-default settings are required, the valve may be configured within Context element with the required values.

Attributes

The Digest Authenticator Valve supports the following configuration attributes:

Attribute Description
alwaysUseSession

Should a session always be used once a user is authenticated? This may offer some performance benefits since the session can then be used to cache the authenticated Principal, hence removing the need to authenticate the user via the Realm on every request. This may be of help for combinations such as BASIC authentication used with the JNDIRealm or DataSourceRealms. However there will also be the performance cost of creating and GC'ing the session. If not set, the default value of false will be used.

cache

Should we cache authenticated Principals if the request is part of an HTTP session? If not specified, the default value of false will be used.

changeSessionIdOnAuthentication

Controls if the session ID is changed if a session exists at the point where users are authenticated. This is to prevent session fixation attacks. If not set, the default value of true will be used.

className

Java class name of the implementation to use. This MUST be set to org.apache.catalina.authenticator.DigestAuthenticator.

disableProxyCaching

Controls the caching of pages that are protected by security constraints. Setting this to false may help work around caching issues in some browsers but will also cause secured pages to be cached by proxies which will almost certainly be a security issue. securePagesWithPragma offers an alternative, secure, workaround for browser caching issues. If not set, the default value of true will be used.

key

The secret key used by digest authentication. If not set, a secure random value is generated. This should normally only be set when it is necessary to keep key values constant either across server restarts and/or across a cluster.

nonceCacheSize

To protect against replay attacks, the DIGEST authenticator tracks server nonce and nonce count values. This attribute controls the size of that cache. If not specified, the default value of 1000 is used.

nonceCountWindowSize

Client requests may be processed out of order which in turn means that the nonce count values may be processed out of order. To prevent authentication failures when nonce counts are presented out of order the authenticator tracks a window of nonce count values. This attribute controls how big that window is. If not specified, the default value of 100 is used.

nonceValidity

The time, in milliseconds, that a server generated nonce will be considered valid for use in authentication. If not specified, the default value of 300000 (5 minutes) will be used.

opaque

The opaque server string used by digest authentication. If not set, a random value is generated. This should normally only be set when it is necessary to keep opaque values constant either across server restarts and/or across a cluster.

securePagesWithPragma

Controls the caching of pages that are protected by security constraints. Setting this to false may help work around caching issues in some browsers by using Cache-Control: private rather than the default of Pragma: No-cache and Cache-control: No-cache. If not set, the default value of false will be used.

secureRandomAlgorithm

Name of the algorithm to use to create the java.security.SecureRandom instances that generate session IDs. If an invalid algorithm and/or provider is specified, the platform default provider and the default algorithm will be used. If not specified, the default algorithm of SHA1PRNG will be used. If the default algorithm is not supported, the platform default will be used. To specify that the platform default should be used, do not set the secureRandomProvider attribute and set this attribute to the empty string.

secureRandomClass

Name of the Java class that extends java.security.SecureRandom to use to generate SSO session IDs. If not specified, the default value is java.security.SecureRandom.

secureRandomProvider

Name of the provider to use to create the java.security.SecureRandom instances that generate SSO session IDs. If an invalid algorithm and/or provider is specified, the platform default provider and the default algorithm will be used. If not specified, the platform default provider will be used.

validateUri

Should the URI be validated as required by RFC2617? If not specified, the default value of true will be used. This should normally only be set when Tomcat is located behind a reverse proxy and the proxy is modifying the URI passed to Tomcat such that DIGEST authentication always fails.

jaspicCallbackHandlerClass

Name of the Java class of the javax.security.auth.callback.CallbackHandler implementation which should be used by JASPIC. If none is specified the default org.apache.catalina.authenticator.jaspic.CallbackHandlerImpl will be used.

Form Authenticator Valve

Introduction

The Form Authenticator Valve is automatically added to any Context that is configured to use FORM authentication.

If any non-default settings are required, the valve may be configured within Context element with the required values.

Attributes

The Form Authenticator Valve supports the following configuration attributes:

Attribute Description
changeSessionIdOnAuthentication

Controls if the session ID is changed if a session exists at the point where users are authenticated. This is to prevent session fixation attacks. If not set, the default value of true will be used.

characterEncoding

Character encoding to use to read the username and password parameters from the request. If not set, the encoding of the request body will be used.

className

Java class name of the implementation to use. This MUST be set to org.apache.catalina.authenticator.FormAuthenticator.

disableProxyCaching

Controls the caching of pages that are protected by security constraints. Setting this to false may help work around caching issues in some browsers but will also cause secured pages to be cached by proxies which will almost certainly be a security issue. securePagesWithPragma offers an alternative, secure, workaround for browser caching issues. If not set, the default value of true will be used.

landingPage

Controls the behavior of the FORM authentication process if the process is misused, for example by directly requesting the login page or delaying logging in for so long that the session expires. If this attribute is set, rather than returning an error response code, Tomcat will redirect the user to the specified landing page if the login form is submitted with valid credentials. For the login to be processed, the landing page must be a protected resource (i.e. one that requires authentication). If the landing page does not require authentication then the user will not be logged in and will be prompted for their credentials again when they access a protected page.

securePagesWithPragma

Controls the caching of pages that are protected by security constraints. Setting this to false may help work around caching issues in some browsers by using Cache-Control: private rather than the default of Pragma: No-cache and Cache-control: No-cache. If not set, the default value of false will be used.

secureRandomAlgorithm

Name of the algorithm to use to create the java.security.SecureRandom instances that generate session IDs. If an invalid algorithm and/or provider is specified, the platform default provider and the default algorithm will be used. If not specified, the default algorithm of SHA1PRNG will be used. If the default algorithm is not supported, the platform default will be used. To specify that the platform default should be used, do not set the secureRandomProvider attribute and set this attribute to the empty string.

secureRandomClass

Name of the Java class that extends java.security.SecureRandom to use to generate SSO session IDs. If not specified, the default value is java.security.SecureRandom.

secureRandomProvider

Name of the provider to use to create the java.security.SecureRandom instances that generate SSO session IDs. If an invalid algorithm and/or provider is specified, the platform default provider and the default algorithm will be used. If not specified, the platform default provider will be used.

jaspicCallbackHandlerClass

Name of the Java class of the javax.security.auth.callback.CallbackHandler implementation which should be used by JASPIC. If none is specified the default org.apache.catalina.authenticator.jaspic.CallbackHandlerImpl will be used.

SSL Authenticator Valve

Introduction

The SSL Authenticator Valve is automatically added to any Context that is configured to use SSL authentication.

If any non-default settings are required, the valve may be configured within Context element with the required values.

Attributes

The SSL Authenticator Valve supports the following configuration attributes:

Attribute Description
cache

Should we cache authenticated Principals if the request is part of an HTTP session? If not specified, the default value of true will be used.

className

Java class name of the implementation to use. This MUST be set to org.apache.catalina.authenticator.SSLAuthenticator.

changeSessionIdOnAuthentication

Controls if the session ID is changed if a session exists at the point where users are authenticated. This is to prevent session fixation attacks. If not set, the default value of true will be used.

disableProxyCaching

Controls the caching of pages that are protected by security constraints. Setting this to false may help work around caching issues in some browsers but will also cause secured pages to be cached by proxies which will almost certainly be a security issue. securePagesWithPragma offers an alternative, secure, workaround for browser caching issues. If not set, the default value of true will be used.

securePagesWithPragma

Controls the caching of pages that are protected by security constraints. Setting this to false may help work around caching issues in some browsers by using Cache-Control: private rather than the default of Pragma: No-cache and Cache-control: No-cache. If not set, the default value of false will be used.

secureRandomAlgorithm

Name of the algorithm to use to create the java.security.SecureRandom instances that generate session IDs. If an invalid algorithm and/or provider is specified, the platform default provider and the default algorithm will be used. If not specified, the default algorithm of SHA1PRNG will be used. If the default algorithm is not supported, the platform default will be used. To specify that the platform default should be used, do not set the secureRandomProvider attribute and set this attribute to the empty string.

secureRandomClass

Name of the Java class that extends java.security.SecureRandom to use to generate SSO session IDs. If not specified, the default value is java.security.SecureRandom.

secureRandomProvider

Name of the provider to use to create the java.security.SecureRandom instances that generate SSO session IDs. If an invalid algorithm and/or provider is specified, the platform default provider and the default algorithm will be used. If not specified, the platform default provider will be used.

jaspicCallbackHandlerClass

Name of the Java class of the javax.security.auth.callback.CallbackHandler implementation which should be used by JASPIC. If none is specified the default org.apache.catalina.authenticator.jaspic.CallbackHandlerImpl will be used.

SPNEGO Valve

Introduction

The SPNEGO Authenticator Valve is automatically added to any Context that is configured to use SPNEGO authentication.

If any non-default settings are required, the valve may be configured within Context element with the required values.

Attributes

The SPNEGO Authenticator Valve supports the following configuration attributes:

Attribute Description
applyJava8u40Fix

A fix introduced in Java 8 update 40 ( JDK-8048194) onwards broke SPNEGO authentication for IE with Tomcat running on Windows 2008 R2 servers. This option enables a work-around that allows SPNEGO authentication to continue working. The work-around should not impact other configurations so it is enabled by default. If necessary, the workaround can be disabled by setting this attribute to false.

alwaysUseSession

Should a session always be used once a user is authenticated? This may offer some performance benefits since the session can then be used to cache the authenticated Principal, hence removing the need to authenticate the user on every request. This will also help with clients that assume that the server will cache the authenticated user. However there will also be the performance cost of creating and GC'ing the session. For an alternative solution see noKeepAliveUserAgents. If not set, the default value of false will be used.

cache

Should we cache authenticated Principals if the request is part of an HTTP session? If not specified, the default value of true will be used.

className

Java class name of the implementation to use. This MUST be set to org.apache.catalina.authenticator.SpnegoAuthenticator.

changeSessionIdOnAuthentication

Controls if the session ID is changed if a session exists at the point where users are authenticated. This is to prevent session fixation attacks. If not set, the default value of true will be used.

disableProxyCaching

Controls the caching of pages that are protected by security constraints. Setting this to false may help work around caching issues in some browsers but will also cause secured pages to be cached by proxies which will almost certainly be a security issue. securePagesWithPragma offers an alternative, secure, workaround for browser caching issues. If not set, the default value of true will be used.

loginConfigName

The name of the JAAS login configuration to be used to login as the service. If not specified, the default of com.sun.security.jgss.krb5.accept is used.

noKeepAliveUserAgents

Some clients (not most browsers) expect the server to cache the authenticated user information for a connection and do not resend the credentials with every request. Tomcat will not do this unless an HTTP session is available. A session will be available if either the application creates one or if alwaysUseSession is enabled for this Authenticator.

As an alternative to creating a session, this attribute may be used to define the user agents for which HTTP keep-alive is disabled. This means that a connection will only used for a single request and hence there is no ability to cache authenticated user information per connection. There will be a performance cost in disabling HTTP keep-alive.

The attribute should be a regular expression that matches the entire user-agent string, e.g. .*Chrome.*. If not specified, no regular expression will be defined and no user agents will have HTTP keep-alive disabled.

securePagesWithPragma

Controls the caching of pages that are protected by security constraints. Setting this to false may help work around caching issues in some browsers by using Cache-Control: private rather than the default of Pragma: No-cache and Cache-control: No-cache. If not set, the default value of false will be used.

secureRandomAlgorithm

Name of the algorithm to use to create the java.security.SecureRandom instances that generate session IDs. If an invalid algorithm and/or provider is specified, the platform default provider and the default algorithm will be used. If not specified, the default algorithm of SHA1PRNG will be used. If the default algorithm is not supported, the platform default will be used. To specify that the platform default should be used, do not set the secureRandomProvider attribute and set this attribute to the empty string.

secureRandomClass

Name of the Java class that extends java.security.SecureRandom to use to generate SSO session IDs. If not specified, the default value is java.security.SecureRandom.

secureRandomProvider

Name of the provider to use to create the java.security.SecureRandom instances that generate SSO session IDs. If an invalid algorithm and/or provider is specified, the platform default provider and the default algorithm will be used. If not specified, the platform default provider will be used.

storeDelegatedCredential

Controls if the user' delegated credential will be stored in the user Principal. If available, the delegated credential will be available to applications (e.g. for onward authentication to external services) via the org.apache.catalina.realm.GSS_CREDENTIAL request attribute. If not set, the default value of true will be used.

jaspicCallbackHandlerClass

Name of the Java class of the javax.security.auth.callback.CallbackHandler implementation which should be used by JASPIC. If none is specified the default org.apache.catalina.authenticator.jaspic.CallbackHandlerImpl will be used.

Error Report Valve

Introduction

The Error Report Valve is a simple error handler for HTTP status codes that will generate and return HTML error pages. It can also be configured to return pre-defined static HTML pages for specific status codes and/or exception types.

NOTE: Disabling both showServerInfo and showReport will only return the HTTP status code and remove all CSS from the default response.

Attributes

The Error Report Valve supports the following configuration attributes:

Attribute Description
className

Java class name of the implementation to use. This MUST be set to org.apache.catalina.valves.ErrorReportValve to use the default error report valve.

errorCode.nnn

The location of the UTF-8 encoded HTML file to return for the HTTP error code represented by nnn. For example, errorCode.404 specifies the file to return for an HTTP 404 error. The location may be relative or absolule. If relative, it must be relative to $CATALINA_BASE. The special value of errorCode.0 may be used to define a default error page to be used if no error page is defined for a status code. If no matching error page is found, the default Error Report Valve response will be returned.

exceptionType.fullyQualifiedClassName

The location of the UTF-8 encoded HTML file to return if an error has occurred and the javax.servlet.error.exception request attribute has been set to an instance of fullyQualifiedClassName or a sub-class of it. For example, errorCode.java.io.IOException specifies the file to return for an IOException. The location may be relative or absolule. If relative, it must be relative to $CATALINA_BASE. If no matching error page is found, the default Error Report Valve response will be returned.

showReport

Flag to determine if the error report is presented when an error occurs. If set to false, then the error report is not in the HTML response. Default value: true

showServerInfo

Flag to determine if server information is presented when an error occurs. If set to false, then the server version is not returned in the HTML response. Default value: true

Crawler Session Manager Valve

Introduction

Web crawlers can trigger the creation of many thousands of sessions as they crawl a site which may result in significant memory consumption. This Valve ensures that crawlers are associated with a single session - just like normal users - regardless of whether or not they provide a session token with their requests.

This Valve may be used at the Engine, Host or Context level as required. Normally, this Valve would be used at the Engine level.

If used in conjunction with Remote IP valve then the Remote IP valve should be defined before this valve to ensure that the correct client IP address is presented to this valve.

Attributes

The Crawler Session Manager Valve supports the following configuration attributes:

Attribute Description
className

Java class name of the implementation to use. This MUST be set to org.apache.catalina.valves.CrawlerSessionManagerValve.

crawlerIps

Regular expression (using java.util.regex) that client IP is matched against to determine if a request is from a web crawler. By default such regular expression is not set.

crawlerUserAgents

Regular expression (using java.util.regex) that the user agent HTTP request header is matched against to determine if a request is from a web crawler. If not set, the default of .*[bB]ot.*|.*Yahoo! Slurp.*|.*Feedfetcher-Google.* is used.

sessionInactiveInterval

The minimum time in seconds that the Crawler Session Manager Valve should keep the mapping of client IP to session ID in memory without any activity from the client. The client IP / session cache will be periodically purged of mappings that have been inactive for longer than this interval. If not specified the default value of 60 will be used.

Stuck Thread Detection Valve

Introduction

This valve allows to detect requests that take a long time to process, which might indicate that the thread that is processing it is stuck. Additionally it can optionally interrupt such threads to try and unblock them.

When such a request is detected, the current stack trace of its thread is written to Tomcat log with a WARN level.

The IDs and names of the stuck threads are available through JMX in the stuckThreadIds and stuckThreadNames attributes. The IDs can be used with the standard Threading JVM MBean (java.lang:type=Threading) to retrieve other information about each stuck thread.

Attributes

The Stuck Thread Detection Valve supports the following configuration attributes:

Attribute Description
className

Java class name of the implementation to use. This MUST be set to org.apache.catalina.valves.StuckThreadDetectionValve.

threshold

Minimum duration in seconds after which a thread is considered stuck. Default is 600 seconds. If set to 0, the detection is disabled.

Note: since the detection (and optional interruption) is done in the background thread of the Container (Engine, Host or Context) declaring this Valve, the threshold should be higher than the backgroundProcessorDelay of this Container.

interruptThreadThreshold

Minimum duration in seconds after which a stuck thread should be interrupted to attempt to "free" it.

Note that there's no guarantee that the thread will get unstuck. This usually works well for threads stuck on I/O or locks, but is probably useless in case of infinite loops.

Default is -1 which disables the feature. To enable it, the value must be greater or equal to threshold.

Semaphore Valve

Introduction

The Semaphore Valve is able to limit the number of concurrent request processing threads.

org.apache.catalina.valves.SemaphoreValve provides methods which may be overridden by a subclass to customize behavior:

  • controlConcurrency may be overridden to add conditions;
  • permitDenied may be overridden to add error handling when a permit isn't granted.

Attributes

The Semaphore Valve supports the following configuration attributes:

Attribute Description
block

Flag to determine if a thread is blocked until a permit is available. The default value is true.

className

Java class name of the implementation to use. This MUST be set to org.apache.catalina.valves.SemaphoreValve.

concurrency

Concurrency level of the semaphore. The default value is 10.

fairness

Fairness of the semaphore. The default value is false.

interruptible

Flag to determine if a thread may be interrupted until a permit is available. The default value is false.

Comments

Notice: This comments section collects your suggestions on improving documentation for Apache Tomcat.

If you have trouble and need help, read Find Help page and ask your question on the tomcat-users mailing list. Do not ask such questions here. This is not a Q&A section.

The Apache Comments System is explained here. Comments may be removed by our moderators if they are either implemented or considered invalid/off-topic.