记一次升级Tomcat

总述

    JDK都要出12了，而我们项目使用的jdk却仍然还停留在JDK1.6。为了追寻技术的发展的脚步，我这边准备将项目升级到JDK1.8。而作为一个web项目，我们的容器使用的是Tomcat。看了下Tomcat版本与JDK版本之间的兼容关系http://tomcat.apache.org/whichversion.html以及网上所传的各种JDK1.8和Tomcat7不兼容的问题， 我决定将Tomcat升级到8。我这里本地验证采用的tomcat版本是8.5.38https://tomcat.apache.org/download-80.cgi。

问题一：请求js文件报404错误

    其实这个问题严格来讲不是升级到Tomcat8出现的问题，而是升级到Tomcat9出现的问题。正好我开始尝试的是Tomcat9，无法解决这个问题才降到Tomcat8。所以这里一并记录下来。

    这个问题在从Tomcat6升级到Tomcat7之后也会存在，原因如下，在项目代码中对js的请求路径中包含了{、}等特殊符号：

<script type="text/javascript" src="${ctx}/js/common/include_css.js?{'ctx':'${ctx}','easyui':'easyui'}"></script>

    前台会发现加载js的时候报了404的错误，后台报错信息如下：

Invalid character found in the request target.The valid characters are defined in RFC 7230 and RFC3986

    出现这个问题的原因是因为Tomcat升级之后对安全进行了升级，其中就有对请求中的特殊字符进行校验，具体校验规则参照下面的代码：

(InternalInputBuffer、InternalAprInputBuffer、InternalNioInputBuffer)

/**
 * Read the request line. This function is meant to be used during the
 * HTTP request header parsing. Do NOT attempt to read the request body
 * using it.
 *
 * @throws IOException If an exception occurs during the underlying socket
 * read operations, or if the given buffer is not big enough to accommodate
 * the whole line.
 */
@Override
public boolean parseRequestLine(boolean useAvailableDataOnly)
	throws IOException {
	int start = 0;
	//
	// Skipping blank lines
	//
	byte chr = 0;
	do {
		// Read new bytes if needed
		if (pos >= lastValid) {
			if (!fill())
				throw new EOFException(sm.getString("iib.eof.error"));
		}
		// Set the start time once we start reading data (even if it is
		// just skipping blank lines)
		if (request.getStartTime() < 0) {
			request.setStartTime(System.currentTimeMillis());
		}
		chr = buf[pos++];
	} while ((chr == Constants.CR) || (chr == Constants.LF));
	pos--;
	// Mark the current buffer position
	start = pos;
	//
	// Reading the method name
	// Method name is a token
	//
	boolean space = false;
	while (!space) {
		// Read new bytes if needed
		if (pos >= lastValid) {
			if (!fill())
				throw new EOFException(sm.getString("iib.eof.error"));
		}
		// Spec says method name is a token followed by a single SP but
		// also be tolerant of multiple SP and/or HT.
		if (buf[pos] == Constants.SP || buf[pos] == Constants.HT) {
			space = true;
			request.method().setBytes(buf, start, pos - start);
		} else if (!HttpParser.isToken(buf[pos])) {
			throw new IllegalArgumentException(sm.getString("iib.invalidmethod"));
		}
		pos++;
	}
	// Spec says single SP but also be tolerant of multiple SP and/or HT
	while (space) {
		// Read new bytes if needed
		if (pos >= lastValid) {
			if (!fill())
				throw new EOFException(sm.getString("iib.eof.error"));
		}
		if (buf[pos] == Constants.SP || buf[pos] == Constants.HT) {
			pos++;
		} else {
			space = false;
		}
	}
	// Mark the current buffer position
	start = pos;
	int end = 0;
	int questionPos = -1;
	//
	// Reading the URI
	//
	boolean eol = false;
	while (!space) {
		// Read new bytes if needed
		if (pos >= lastValid) {
			if (!fill())
				throw new EOFException(sm.getString("iib.eof.error"));
		}
		// Spec says single SP but it also says be tolerant of HT
		if (buf[pos] == Constants.SP || buf[pos] == Constants.HT) {
			space = true;
			end = pos;
		} else if ((buf[pos] == Constants.CR)
				   || (buf[pos] == Constants.LF)) {
			// HTTP/0.9 style request
			eol = true;
			space = true;
			end = pos;
		} else if ((buf[pos] == Constants.QUESTION) && (questionPos == -1)) {
			questionPos = pos;
		} else if (HttpParser.isNotRequestTarget(buf[pos])) {
			throw new IllegalArgumentException(sm.getString("iib.invalidRequestTarget"));
		}
		pos++;
	}
	request.unparsedURI().setBytes(buf, start, end - start);
	if (questionPos >= 0) {
		request.queryString().setBytes(buf, questionPos + 1,
									   end - questionPos - 1);
		request.requestURI().setBytes(buf, start, questionPos - start);
	} else {
		request.requestURI().setBytes(buf, start, end - start);
	}
	// Spec says single SP but also says be tolerant of multiple SP and/or HT
	while (space) {
		// Read new bytes if needed
		if (pos >= lastValid) {
			if (!fill())
				throw new EOFException(sm.getString("iib.eof.error"));
		}
		if (buf[pos] == Constants.SP || buf[pos] == Constants.HT) {
			pos++;
		} else {
			space = false;
		}
	}
	// Mark the current buffer position
	start = pos;
	end = 0;
	//
	// Reading the protocol
	// Protocol is always "HTTP/" DIGIT "." DIGIT
	//
	while (!eol) {
		// Read new bytes if needed
		if (pos >= lastValid) {
			if (!fill())
				throw new EOFException(sm.getString("iib.eof.error"));
		}
		if (buf[pos] == Constants.CR) {
			end = pos;
		} else if (buf[pos] == Constants.LF) {
			if (end == 0)
				end = pos;
			eol = true;
		} else if (!HttpParser.isHttpProtocol(buf[pos])) {
			// 关键点在这一句，如果校验不通过，则会报参数异常
			throw new IllegalArgumentException(sm.getString("iib.invalidHttpProtocol"));
		}
		pos++;
	}
	if ((end - start) > 0) {
		request.protocol().setBytes(buf, start, end - start);
	} else {
		request.protocol().setString("");
	}
	return true;
}

我们进一步跟进HttpParser中的方法：

public static boolean isNotRequestTarget(int c) {
	// Fast for valid request target characters, slower for some incorrect
	// ones
	try {
		// 关键在于这个数组
		return IS_NOT_REQUEST_TARGET[c];
	} catch (ArrayIndexOutOfBoundsException ex) {
		return true;
	}
}
// Combination of multiple rules from RFC7230 and RFC 3986. Must be
// ASCII, no controls plus a few additional characters excluded
if (IS_CONTROL[i] || i > 127 ||
		i == ' ' || i == '\"' || i == '#' || i == '<' || i == '>' || i == '\\' ||
		i == '^' || i == '`'  || i == '{' || i == '|' || i == '}') {
	// 可以看到只有在REQUEST_TARGET_ALLOW数组中的值才不会设置成true，所以我们需要追踪REQUEST_TARGET_ALLOW数组的赋值
	if (!REQUEST_TARGET_ALLOW[i]) {
		IS_NOT_REQUEST_TARGET[i] = true;
	}
}
String prop = System.getProperty("tomcat.util.http.parser.HttpParser.requestTargetAllow");
if (prop != null) {
	for (int i = 0; i < prop.length(); i++) {
		char c = prop.charAt(i);
		// 可以看到在配置文件中配置了tomcat.util.http.parser.HttpParser.requestTargetAllow并且包含{、}、|的时候，REQUEST_TARGET_ALLOW数组中的值才会为true
		if (c == '{' || c == '}' || c == '|') {
			REQUEST_TARGET_ALLOW[c] = true;
		} else {
			log.warn(sm.getString("httpparser.invalidRequestTargetCharacter",
					Character.valueOf(c)));
		}
	}
}

    解决办法: 其实通过源码分析不难得到解决办法

在Tomcat的catalina.properties文件中添加以下语句：

tomcat.util.http.parser.HttpParser.requestTargetAllow={}|

当然需要注意的是，这个后门在Tomcat8.5以后就无法使用的，Tomcat9之后的解决办法暂时未找到，可能只有对URL进行编码了。

问题二：Cookie设置报错

     这个问题就是在升级到Tomcat8.5以上的时候会出现的，具体原因是Tomcat8.5采用的Cookie处理类是：

Rfc6265CookieProcessor,而在之前使用的处理类是LegacyCookieProcessor。该处理类对domai进行了校验：

private void validateDomain(String domain) {
	int i = 0;
	int prev = -1;
	int cur = -1;
	char[] chars = domain.toCharArray();
	while (i < chars.length) {
		prev = cur;
		cur = chars[i];
		if (!domainValid.get(cur)) {
			throw new IllegalArgumentException(sm.getString(
					"rfc6265CookieProcessor.invalidDomain", domain));
		}
		// labels must start with a letter or number
		if ((prev == '.' || prev == -1) && (cur == '.' || cur == '-')) {
			throw new IllegalArgumentException(sm.getString(
					"rfc6265CookieProcessor.invalidDomain", domain));
		}
		// labels must end with a letter or number
		if (prev == '-' && cur == '.') {
			throw new IllegalArgumentException(sm.getString(
					"rfc6265CookieProcessor.invalidDomain", domain));
		}
		i++;
	}
	// domain must end with a label
	if (cur == '.' || cur == '-') {
		throw new IllegalArgumentException(sm.getString(
				"rfc6265CookieProcessor.invalidDomain", domain));
	}
}

新的Cookie规范对domain有以下要求

1、必须是1-9、a-z、A-Z、. 、- （注意是-不是_）这几个字符组成

2、必须是数字或字母开头 （所以以前的cookie的设置为.XX.com 的机制要改为 XX.com 即可）

3、必须是数字或字母结尾

原来的代码设置domain时如下：

cookie.setDomain(".aaa.com");

这就导致设置domain的时候不符合新的规范，直接报错如下：

java.lang.IllegalArgumentException: An invalid domain [.aaa.com] was specified for this cookie
        at org.apache.tomcat.util.http.Rfc6265CookieProcessor.validateDomain(Rfc6265CookieProcessor.java:181)
        at org.apache.tomcat.util.http.Rfc6265CookieProcessor.generateHeader(Rfc6265CookieProcessor.java:123)
        at org.apache.catalina.connector.Response.generateCookieString(Response.java:989)
        at org.apache.catalina.connector.Response.addCookie(Response.java:937)
        at org.apache.catalina.connector.ResponseFacade.addCookie(ResponseFacade.java:386)

    解决办法(以下3中任意一种皆可)

1. 修改原来代码为：

   cookie.setDomain("aaa.com");
   

2. 如果是Spring-boot环境，直接替换默认的Cookie处理类：

   @Configuration
   @ConditionalOnExpression("${tomcat.useLegacyCookieProcessor:false}")
   public class LegacyCookieProcessorConfiguration {
       @Bean
       EmbeddedServletContainerCustomizer embeddedServletContainerCustomizerLegacyCookieProcessor() {
           return new EmbeddedServletContainerCustomizer() {
               @Override
               public void customize(ConfigurableEmbeddedServletContainer factory) {
                   if (factory instanceof TomcatEmbeddedServletContainerFactory) {
                       TomcatEmbeddedServletContainerFactory tomcatFactory =
                               (TomcatEmbeddedServletContainerFactory) factory;
                       tomcatFactory.addContextCustomizers(new TomcatContextCustomizer() {
                           @Override
                           public void customize(Context context) {
                               context.setCookieProcessor(new LegacyCookieProcessor());
                           }
                       });
                   }
               }
           };
       }
   }
   

3. 在Tomcat的context.xml中增加如下配置，指定Cookie的处理类：

   <CookieProcessor className="org.apache.tomcat.util.http.LegacyCookieProcessor" />
   

参考链接

https://blog.csdn.net/fy_sun123/article/details/73115381

http://ju.outofmemory.cn/entry/367186

https://www.cnblogs.com/lr393993507/p/7755867.html

http://tomcat.apache.org/tomcat-8.5-doc/config/cookie-processor.html