[jboss-jira] [JBoss JIRA] (JGRP-1480) NAKACK retransmission request for message 0
[
https://issues.jboss.org/browse/JGRP-1480?page=com.atlassian.jira.plugin....
]
Dennis Reed commented on JGRP-1480:
-----------------------------------
The root issue appears to be in NAKACK#rebroadcastMessages, which is called during a
FLUSH_RECONCILE when the digests on different nodes
do not match (due to the multicast packet loss), and the requester node has not received
any messages from the local_addr node.
NAKACK#rebroadcastMessages requests messages [my_high, their_high]. It appears this
should be [my_high + 1, their_high] instead.
NAKACK retransmission request for message 0
-------------------------------------------
Key: JGRP-1480
URL: https://issues.jboss.org/browse/JGRP-1480
Project: JGroups
Issue Type: Bug
Affects Versions: 2.6.20
Reporter: Dennis Reed
Assignee: Dennis Reed
The following warning is logged. Since there is no message with ID 0, it should not be
requested for retransmission.
WARN [org.jgroups.protocols.pbcast.NAKACK] (OOB-19,10.1.1.1:55200)
(requester=10.1.1.2:55200, local_addr=10.1.1.1:55200) message 10.1.1.1:55200::0 not found
in retransmission table of 10.1.1.1:55200:
[0 : 7 (7) (size=7, missing=0, highest stability=0)]
It occurred during a FLUSH when a node is trying to join the cluster, in a network that
is dropping some multicast packets.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira