[JENKINS-13265] Ldap connection failed - jenkins loosing FQDN of ldap server

Jenkins is configured to use LDAP login which works perfectly. After some time - some times directly after restarting tomcat - the login fails. When checking catalina.out the attached exception is thrown. The reason for the failed login is clear - Jenkins is configured to connect to ldap.my.domain - but when this happens Jenkins forgets "ldap" and tries to connect to my.domain only - for sure this does not work as this is no valid host name.

To fix this, we need to restart Tomcat until it failes again.

---
Originally reported by jomu78, imported from: Ldap connection failed - jenkins loosing FQDN of ldap server

• status: Open
  
• priority: Major
  
• component(s): ldap-plugin
  
• label(s): ConnectExcetion, ldap, plugin
  
• resolution: Unresolved
  
• votes: 4
  
• watchers: 6
  
• imported: 2025-12-09
  

Raw content of original issue

Jenkins is configured to use LDAP login which works perfectly. After some time - some times directly after restarting tomcat - the login fails. When checking catalina.out the attached exception is thrown. The reason for the failed login is clear - Jenkins is configured to connect to ldap.my.domain - but when this happens Jenkins forgets "ldap" and tries to connect to my.domain only - for sure this does not work as this is no valid host name. 


To fix this, we need to restart Tomcat until it failes again. 

• environment: Jenkins in all 1.4xx versions, Linux X86, SLES 10, Apache Tomcat 7.0.14

1 attachment

- [exception.txt](https://issues.jenkins.io/secure/attachment/21682/exception.txt)

Start with the ldap-plugin code paths that create LDAP connections and read the configured server name, then inspect the attached exception.txt and the catalina.out failure described in the report. Reproduce the intermittent login failure if possible and verify that connections consistently use the configured ldap.my.domain hostname rather than my.domain.

Written by the indexing model from the issue text.