Re: (ITS#5133) Synchronous replication on slave doesn't notice lost network connection

[stelios.xx.grigoriadis@ericsson.com]

We have created a patch that works for us. An updated version is 
available for 2.4.9

The patch is using the parameter "mastercheckint" in slapd.conf, if 
present. A separate
thread then continously askes the master for the root DSE.

An excerpt from the slapd.conf file:

# from master ldapt1.roadrunner
syncrepl rid=102
   provider=ldap://ldapt1.roadrunner
   type=refreshAndPersist
   retry="10 10 60 +"
   searchbase="dc=example,dc=com"
   filter="(objectClass=*)"
   scope=sub
   schemachecking=off
   sizelimit=500000
   timelimit=360000
   bindmethod=simple
   mastercheckint=5 ### THIS is the new parameter.
   binddn="cn=admin,dc=example,dc=com"
   credentials=secret
updateref  ldap://ldapt1.roadrunner
overlay syncprov


If the master unexpectedly terminates (eg. due to a power failure) the 
replicas will get
a new connection once it's up again.

The patch is available at:
http://www.freewebs.com/stigri/Stelios-Grigoriadis-its5133-080514.patch

It must be run while in openlda-2.4.9 directory.

/Stelios Grigoriadis