[
https://issues.apache.org/jira/browse/CASSANDRA-7816?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14107681#comment-14107681
]
Michael Penick commented on CASSANDRA-7816:
-------------------------------------------
It seems to be "UP" events that mostly repeat, but I've also seen it happen
with "DOWN" events. I've reproduced the issue several times with the following:
Create a two node cluster, with a control connection listening on "node1".
{code}
ccm create test -n 2:0 -s -i 127.0.0. -b -v 2.0.9
ccm stop node2
ccm start node2
ccm stop node2
ccm start node2
...
{code}
I've ruled out the driver's code using tcpdump. I've attached a tcpdump file
showing the case where an "UP" event has repeated several times. Here's the
command I used:
{code}
sudo tcpdump -s 0 -i lo0 -X src host 127.0.0.1 and dst host 127.0.0.1 and tcp
port 9042 and dst port <the port the driver's connected on>
{code}
This will happen after a couple rounds of starting and stopping the node.
Sorry, this specific to my driver, but it shows that several unique events are
sent as a result of"ccm start node2". For some reason C* sends a status change
node "DOWN" before sending the "UP" event when the node is started. The number
of "DOWN"/"UP" events varies from run to run.
{code}
[INFO]: ControlConnection: Node'127.0.0.2:9042' is down
[INFO]: ControlConnection: Node'127.0.0.2:9042' is up (0x1033004e0)
[INFO]: ControlConnection: Node'127.0.0.2:9042' is up (0x103600000)
[INFO]: ControlConnection: Node'127.0.0.2:9042' is up (0x1003023e0)
[INFO]: ControlConnection: Node'127.0.0.2:9042' is up (0x10020a150)
[INFO]: ControlConnection: Node'127.0.0.2:9042' is up (0x103300230)
[INFO]: ControlConnection: Node'127.0.0.2:9042' is up (0x103300230)
{code}
> Updated the "4.2.6. EVENT" section in the binary protocol specification
> -----------------------------------------------------------------------
>
> Key: CASSANDRA-7816
> URL: https://issues.apache.org/jira/browse/CASSANDRA-7816
> Project: Cassandra
> Issue Type: Improvement
> Components: Documentation & website
> Reporter: Michael Penick
> Priority: Trivial
> Attachments: tcpdump_repeating_status_change.txt, trunk-7816.txt
>
>
> Added "MOVED_NODE" as a possible type of topology change and also specified
> that it is possible to receive the same event multiple times.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
