| Summary: | slapd fails to stop with /etc/init.d/slapd stop | ||
|---|---|---|---|
| Product: | Gentoo Linux | Reporter: | Janne Pikkarainen <jabapi> |
| Component: | [OLD] Server | Assignee: | Ryan Phillips (RETIRED) <rphillips> |
| Status: | RESOLVED FIXED | ||
| Severity: | major | ||
| Priority: | High | ||
| Version: | unspecified | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Package list: | Runtime testing required: | --- | |
|
Description
Janne Pikkarainen
2004-03-01 06:47:54 UTC
I am unable to reproduce this bug with openldap 2.1.26. Starting and stopping slapd works fine for me. I suggest that it is something to do with your particular setup. Hmm, let me try with 2.1.26 -- I must confess that the latest 2.1.x version I tried was 2.1.23. I'll also give 2.0.27-r5 a try, the latest I really have tried is 2.0.27-r4, but since changelog didn't tell me anything related to this bug, I thought it would still be alive. Ok. It works fine with 2.0.27-r5, but not with 2.1.26 (tried with my own slapd.conf and out-of-the-box slapd.conf). With 2.0.27-r5 there were slapd.pid and slapd.args files in /var/run/openldap/, but not with 2.1.26, so that must be why restarting slapd fails. I wonder what's going on - I have these lines in my slapd.conf but still in some versions of OpenLDAP the pid/args file doesn't get created during startup. --- pidfile /var/run/openldap/slapd.pid argsfile /var/run/openldap/slapd.args --- Maybe slapd crashed *before* you tried to shut it down? Does 'ps xa | grep slapd' show something? No, slapd did not crash before /etc/init.d/slapd stop. I know it because that ldap database was in active use all the time. Is this still a problem? Does the directory /var/run/openldap exist with owner ldap:ldap? ; ls -l /var/run | fgrep openldap drwxr-xr-x 2 ldap ldap 136 nov 17 19:35 openldap And does your /etc/init.d/slapd say this: ; cat /etc/init.d/slapd | egrep " --stop" start-stop-daemon --stop --quiet --pidfile /var/run/openldap/slapd.pid Nowadays this works ok. (In reply to comment #7) > Nowadays this works ok. Closing a stale bug. |