Tomcat : Invalid character found in the request target

Invalid character found in the request target. The valid characters are defined in RFC 7230 and RFC 3986

Tomcat的coyote-connector解析到请求中包含了不合法的字符,比如  { }  ( )  ^ ` \  | #  \\

这是典型的400错误，不合法的请求，根据RFC （Request Format Comment）文档的规定，以上字符不能被用在Request Header，也就是request的URI中。

我遇到这个问题是两个tomcat应用之间通讯时发送了一个请求，这个请求包含片段.*\.(css|js|jpg|jpeg|png|gif|ico|woff)$，

很显然，这个请求中包含 \ ( | 这三个不合法的字符；转义是不能解决问题的；因为tomcat会对请求中每个字符都进行检查；

我的解决办法就是 将 .*\.(css|js|jpg|jpeg|png|gif|ico|woff)$ 换成 .*tpl$,.*css$,.*js$,.*jpg$,.*jpeg$,.*bmp$,.*png$,.*gif$,.*ico$,.*woff$

查看tomcat源码

apache-tomcat-7.0.73-src\java\org\apache\coyote\http11\InternalInputBuffer.java

public class InternalInputBuffer extends AbstractInputBuffer<Socket> {
 　　....省略
    /**
     * 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 {
    ...省略
        //
        // 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 (HttpParser.isNotRequestTarget(buf[pos])) {　　
　　　　　　　　//如果请求参数中的字符不是被允许的字符，则抛异常  HttpParser这个类 看下面代码片
                throw new IllegalArgumentException(sm.getString("iib.invalidRequestTarget"));
            }
            pos++;
        }
    ... 省略
        return true;
    }
}

apache-tomcat-7.0.73-src\java\org\apache\tomcat\util\http\parser\HttpParser.java

private static final int ARRAY_SIZE = 128;
private static final boolean[] IS_NOT_REQUEST_TARGET = new boolean[ARRAY_SIZE];
static {
        // Digest field types.for (int i = 0; i < ARRAY_SIZE; i++) {
            // Control> 0-31, 127
            if (i < 32 || i == 127) {
                IS_CONTROL[i] = true;
            }// Not valid for request target.
            // Combination of multiple rules from RFC7230 and RFC 3986. Must be
            // ASCII, no controls plus a few additional characters excluded
　　　　　　　// 不合法的字符在这里都会导致请求不合法而抛异常 请求失败
            if (IS_CONTROL[i] || i > 127 ||
                    i == ' ' || i == '\"' || i == '#' || i == '<' || i == '>' || i == '\\' ||
                    i == '^' || i == '`'  || i == '{' || i == '|' || i == '}') {
                IS_NOT_REQUEST_TARGET[i] = true;
            }
        }
    }

HTTP协议只是一个OSI应用层通讯的标准，在tomcat源码中对HTTP进行了实现，可能在一些Tomcat版本中没有实现对请求字符的限制，可以预见，在Tomcat7.0.64之后的版本以及 Tomcat8、9都会对请求头的字符进行限制。