[Date Prev][Date Next]
[Chronological]
[Thread]
[Top]
(ITS#4240) back-meta loses "connection" with remote backend
Full_Name: Perry Nguyen
Version: openldap-2.3.12
OS: Linux FC4
URL: ftp://ftp.openldap.org/incoming/
Submission from: (NULL) (66.245.252.239)
After a long period of time (days or more) slapd needs to be restarted to be
able to retrieve results from a meta backend. This does not seem to be related
to idletimeout as I've tested against another openldap server with the
idletimeout set to a low value (10 seconds) and back-meta was able to return
results after being timed out multiple times.
I don't know what's causing this right now, this is just an issue report. My
only current workaround is to restart the ldap server once a day.
slapd.conf bit:
database meta
readonly on
nretries forever
suffix "ou=sso,ou=ecmbi,o=ibm"
uri "ldaps://bluepages.ibm.com/c=us,ou=sso,ou=ecmbi,o=ibm"
suffixmassage "c=us,ou=sso,ou=ecmbi,o=ibm" "c=us,ou=bluepages,o=ibm.com"
uri "ldaps://bluepages.ibm.com/c=cn,ou=sso,ou=ecmbi,o=ibm"
suffixmassage "c=cn,ou=sso,ou=ecmbi,o=ibm" "c=cn,ou=bluepages,o=ibm.com"
uri "ldap:///ou=sso,ou=ecmbi,o=ibm"
suffixmassage "ou=sso,ou=ecmbi,o=ibm" "ou=SSO Stub,ou=ecmbi,o=ibm"
command output:
[pfnguyen@wassup ~]$ ldapsearch -x notesshortname=pfnguyen 1.1
# extended LDIF
#
# LDAPv3
# base <> with scope sub
# filter: notesshortname=pfnguyen
# requesting: 1.1
#
# search result
search: 2
result: 0 Success
[pfnguyen@wassup ~]$ sudo /sbin/service ldap restart
Stopping slapd: [ OK ]
Stopping slurpd: [ OK ]
Starting slapd: [ OK ]
Starting slurpd: [ OK ]
[pfnguyen@wassup ~]$ !ldaps
ldapsearch -x notesshortname=pfnguyen 1.1
# extended LDIF
#
# LDAPv3
# base <> with scope sub
# filter: notesshortname=pfnguyen
# requesting: 1.1
#
# 828197897, us, sso, ecmbi, ibm
dn: uid=828197897,c=us,ou=sso,ou=ecmbi,o=ibm
# search result
search: 2
result: 0 Success
# numResponses: 2
# numEntries: 1