"no messages seen in" from itself when starting

Bug #1738387 reported by Iwo on 2017-12-15
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Percona XtraDB Cluster moved to https://jira.percona.com/projects/PXC
Status tracked in 5.6
Kenn Takara
Kenn Takara

Bug Description

When node's own IP address is defined in in wsrep_cluter_address


the node is blacklisting itself

     [Warning] WSREP: (cd1938a2, 'tcp://') address 'tcp://' points to own listening address, blacklisting

however later is trying to connect itself:

     [Note] WSREP: (cd1938a2, 'tcp://') connection to peer cd1938a2 with addr tcp:// timed out, no messages seen in PT3S

An output of `ip a l`:

# ip a l
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
    link/ether 08:00:27:25:19:38 brd ff:ff:ff:ff:ff:ff
    inet brd scope global eth0
       valid_lft forever preferred_lft forever
    inet6 fe80::a00:27ff:fe25:1938/64 scope link
       valid_lft forever preferred_lft forever

Tested against PXC 5.7.19-29.22.3 and PXC 5.6.37-26.21-3.

Kenn Takara (kenn-takara) wrote :

The node is reporting that it hasn't heard from that address in 3 seconds (gmcast.peer.timeout), which is valid since that address has been blacklisted.

The issue would be that the liveness check is reporting on blacklisted entries.

Percona now uses JIRA for bug reports so this bug report is migrated to: https://jira.percona.com/browse/PXC-911

To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers