Hadoop和Flink漏洞修复_青藤云hadoop yarn web漏洞如何修复

Hadoop漏洞修复

• CVE-2021-33036: Hadoop YARN REST API 未授权访问导致远程代码执行漏洞
• Cluster Overview 未授权访问漏洞
• Hadoop 未授权访问漏洞

修复方法1

• 修复方法1

1、进入hadoop配置文件目录，创建密钥文件
cd /home/flink/hadoop-3.3.2/etc/hadoop
echo "hadoop" > hadoop-http-auth-signature-secret

2、备份core-site.xml文件，在配置文件configuration块中添加如下配置项，hadoop.http.authentication.signature.secret.file配置对应密钥文件路径是否完整可访问到。
<configuration>
	<property>
		<name>hadoop.http.filter.initializers</name>
		<value>org.apache.hadoop.security.AuthenticationFilterInitializer</value>
	</property>
	<property>
		<name>hadoop.http.authentication.type</name>
		<value>simple</value>
	</property>
	<property>
		<name>hadoop.http.authentication.token.validity</name>
		<value>3600</value>
	</property>
	<property>
		<name>hadoop.http.authentication.signature.secret.file</name>
		<value>/home/flink/hadoop-3.3.2/etc/hadoop/hadoop-http-auth-signature-secret</value>
	</property>
	<property>
		<name>hadoop.http.authentication.cookie.domain</name>
		<value></value>
	</property>
	<property>
		<name>hadoop.http.authentication.simple.anonymous.allowed</name>
		<value>true</value>
	</property>
</configuration>
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31

修复方法2

• 下载jersey-server-1.19.jar

CSDN地址下载：https://download.csdn.net/download/zhao138969/89487778   对应jersey-server-1.19.jar
CSDN地址下载：https://download.csdn.net/download/zhao138969/89494178 对应hadoop-auth-3.3.2.jar
1
2

• 修改方法2

1、#备份原有jersey-server-1.19.jar文件
 cd ${HADOOP_HOME}/share/hadoop/common/lib/
 mv jersey-server-1.19.jar jersey-server-1.19.jar_bak
 cd ${HADOOP_HOME}/share/hadoop/hdfs/lib/
mv jersey-server-1.19.jar jersey-server-1.19.jar_bak
cd ${HADOOP_HOME}/share/hadoop/common/lib
mv  hadoop-auth-3.3.2.jar hadoop-auth-3.3.2.jar_bak

2、上传加固的jersey-server-1.19.jar到对应目录下
cd ${HADOOP_HOME}/share/hadoop/common/lib/
rz -be jersey-server-1.19.jar
cd ${HADOOP_HOME}/share/hadoop/hdfs/lib/
rz -be jersey-server-1.19.jar
cd ${HADOOP_HOME}/share/hadoop/common/lib
rz -be hadoop-auth-3.3.2.jar

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16

• 重启Hadoop集群

stop-dfs.sh
start-dfs.sh
1
2

• 验证方法

#在浏览器中进行访问
http://ip:50070  #出现错误
http://ip:50070?user.name=hadoop  #访问正确
1
2
3

Flink漏洞修复

• Apache Flink任意 Jar 包上传导致远程代码执行漏洞
• Apache Flink未授权访问漏洞

修改方法

• 修改方法

1、创建cer目录然后，生产证书。
cd /home/flink/flink-1.15.2/conf/cer
#创建证书，密码并设置为Ccod@2024
keytool -genkeypair -alias ca -keystore ca.keystore -dname "CN=Sample CA" -storepass ca_keystore_password -keypass ca_key_password -keyalg RSA -keysize 4096 -ext "bc=ca:true"
keytool -exportcert -keystore ca.keystore -alias ca -storepass ca_keystore_password -file ca.cer
keytool -importcert -keystore ca.truststore -alias ca -storepass Ccod@2024 -file ca.cer -noprompt
keytool -genkeypair -alias flink.rest -keystore rest.signed.keystore -dname "CN=flink.company.org" -ext "SAN=dns:flink.company.org" -storepass Ccod@2024 -keypass Ccod@2024 -keyalg RSA -keysize 4096
keytool -certreq -alias flink.rest -keystore rest.signed.keystore -storepass Ccod@2024 -keypass Ccod@2024 -file rest.csr
keytool -gencert -alias ca -keystore ca.keystore -storepass ca_keystore_password -keypass ca_key_password -ext "SAN=dns:flink.company.org,10.130.47.192" -infile rest.csr -outfile rest.cer
keytool -importcert -keystore rest.signed.keystore -storepass Ccod@2024 -file ca.cer -alias ca -noprompt
keytool -importcert -keystore rest.signed.keystore -storepass Ccod@2024 -keypass Ccod@2024 -file rest.cer -alias flink.rest -noprompt

2、 修改flink配置文件
cp flink-conf.yaml flink-conf.yaml_bak
vim  flink-conf.yaml #以下配置项

security.ssl.rest.authentication-enabled: false
security.ssl.internal.enabled: false
security.ssl.verify-hostname: false
security.ssl.rest.enabled: true
security.ssl.rest.keystore: /home/flink/flink-1.15.2/conf/cer/rest.signed.keystore
security.ssl.rest.truststore: /home/flink/flink-1.15.2/conf/cer/ca.truststore
security.ssl.rest.key-password: Ccod@2024
security.ssl.rest.truststore-password: Ccod@2024
security.ssl.rest.keystore-password: Ccod@2024
web.submit.enabled: false
web.cancel.enabled: false
jobmanager.web.submit.enable: false  #该配置项配置上后,则无法通过web页面提交jar任务，只有通过命令来启动job任务。

3、重启flink
start-cluster.sh
stop-cluster.sh

4、命令行启动flink job task
flink run -d  -c com.channelsoft.flinkstatics.job.ServiceDetailJob -p 3  ./flink-statics-1.1.1.3-SNAPSHOT-jar-with-dependencies.jar  -zookeeper 10.130.47.192:2181

5、登录flink https:ip/8180端口
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37

最简单粗暴的方法

• 待完成