Update mariadb-server suspect_timeout to default value to align
with garbd's suspect_timeout
In openstack-helm-infra, it launch evs.suspect_timeout=PT30S
for mariadb-server in configmap, mariadb-etc. This setting is
for three mariadb-server pod deployment, every mariadb-server
with same setting suspect_timeout=30s. But after change to two
mariadb-server and one garbd arbitrator. Setting in configmap
mariadb-etc evs.suspect_timeout=PT30S, only takes effect for 2
mariadb-server, for garbd arbitrator, it use galera default
setting evs.suspect_timeout=PT5S. If mariadb-server-1 exit
abnormal, after 5s, garbd arbitrator suspects mariadb-server-1
is dead, but as not reach 30s, mariadb-server-0 thinks mariadb-server-1
is not dead. In this state, quorum fail, garbd arbitrator and
mariadb-server-0 both set to none primary component, service
down.
For fix solution, set value.conf.data.config_override to override
wsrep_provider_option in mariadb helm chart, which makes garbd
arbitrator and mariadb-server launch with same setting for
"evs.suspect_timeout=PT5S", default value. By this way, mariadb
server recovery time will also improve. To update setting for
"evs.suspect_timeout", it should both update override for mariadb
and garbd helm chart.
Setting for "gmcast.listen_addr=tcp://0.0.0.0:<port>", takes
effect for both ipv4 and ipv6. So keeps such setting.
Reviewed: https:/ /review. opendev. org/747124 /git.openstack. org/cgit/ starlingx/ config/ commit/ ?id=245023894ac fb163b4ed73ccde d72914550d982c
Committed: https:/
Submitter: Zuul
Branch: r/stx.3.0
commit 245023894acfb16 3b4ed73ccded729 14550d982c
Author: Martin, Chen <email address hidden>
Date: Thu Aug 20 16:26:50 2020 +0800
Update mariadb-server suspect_timeout to default value to align
with garbd's suspect_timeout
In openstack- helm-infra, it launch evs.suspect_ timeout= PT30S timeout= 30s. But after change to two timeout= PT30S, only takes effect for 2 timeout= PT5S. If mariadb-server-1 exit server- 0 both set to none primary component, service data.config_ override to override provider_ option in mariadb helm chart, which makes garbd suspect_ timeout= PT5S", default value. By this way, mariadb suspect_ timeout" , it should both update override for mariadb
for mariadb-server in configmap, mariadb-etc. This setting is
for three mariadb-server pod deployment, every mariadb-server
with same setting suspect_
mariadb-server and one garbd arbitrator. Setting in configmap
mariadb-etc evs.suspect_
mariadb-server, for garbd arbitrator, it use galera default
setting evs.suspect_
abnormal, after 5s, garbd arbitrator suspects mariadb-server-1
is dead, but as not reach 30s, mariadb-server-0 thinks mariadb-server-1
is not dead. In this state, quorum fail, garbd arbitrator and
mariadb-
down.
For fix solution, set value.conf.
wsrep_
arbitrator and mariadb-server launch with same setting for
"evs.
server recovery time will also improve. To update setting for
"evs.
and garbd helm chart.
Setting for "gmcast. listen_ addr=tcp: //0.0.0. 0:<port> ", takes
effect for both ipv4 and ipv6. So keeps such setting.
Reference link for wsrep option and galera cluster quorum /mariadb. com/kb/ en/wsrep_ provider_ options/ /galeracluster. com/library/ documentation/ weighted- quorum. html
https:/
https:/
Closes-Bug: 1888546
Change-Id: I92af77fab929c9 f598b7dc41543db 6ad6238f812
Signed-off-by: Martin, Chen <email address hidden>