[389-users] 389 Ldap Cleanallruv Replica Crash

[Juan Quintanilla]

Hi,

I recently installed 389-ds-base-libs-2.2.6-2.el8.x86_64 and
389-ds-base-2.2.6-2.el8.x86_64 on an ALma Linux 8 Server, but I'm encountering an
issue with removing offline replicas from our existing 389 Ldap.

When the command below is executed on one of the suppliers:

dsconf INSTANCE_NAME repl-tasks cleanallruv --suffix "ou=sample,dc=test,dc=dom"
--replica-id 20 --force-cleaning

The entry is removed from the ldap supplier, and when the change is sent to the secondary
supplier it is also removed with no problem.  The issue is when the change is sent to the
consumer, the slapd process will instantly crash.  When the consumer instance is brought
back up the entry that needed to be removed is gone.

Has anyone encountered a similar issue with the consumers crashing during a cleanallruv
request or cleanruv?

I also tried running a cleanruv task on each server, suppliers have no issue. When the
command is run on the readonly consumers the slapd process crashes.

ldapmodify -x -D "cn=manager" -W <<EOF
dn: cn=replica,cn=ou\3Dsample\2Cdc\3Dtest\2Cdc\3Ddom,cn=mapping tree,cn=config
changetype: modify
replace: nsds5task
nsds5task: CLEANRUV20
EOF

There is no recorded error in the logs to indicate the reason for the crash.

Thanks!


Juan