[pgpool-hackers: 3928] Re: [pgpool-general: 7593] Re: Re: Promote specific backend to primary in streaming replication

Tatsuo Ishii ishii at sraoss.co.jp
Sun Jun 13 16:16:32 JST 2021


>   Just my ¢2: you could use
>   -t --takeover
>   this properly conveys the notion that the newly promoted node takes
> over the existing primary node.

Hum. According to https://stackoverflow.com/questions/59124567/whats-the-difference-between-failover-and-takeover-in-ha

     Takeover is normally a dynamic action, associated with an Active
     / Active configuration, which can have minimal or no outage
     impact for one Active Node to takeover pending or in-progress
     work from the other (failed) Active node.

     Failover is normally associated with an Active / Passive
     configuration, implying some level of automated or manual action
     being required for the Passive node to begin processing work, if
     the Active node fails.

Although "takeover" and "failover" are compared here, I think
"switchover" is in the same group of "failover". So if streaming
replication's primary/standby configuration can be regarded as
Active/Active configuration, probably "takeover" is more
appropriate. On the other hand, if it can be regarded as
Active/Passive configuration, "switchover" is more appropriate.

I feel like this: "some level of automated or manual action being
required for the Passive node to begin processing work, if the Active
node fails" applies to the promoting operation of standby in streaming
replication configuration. So my impression is "switchover" is more
appropriate.

Best regards,
--
Tatsuo Ishii
SRA OSS, Inc. Japan
English: http://www.sraoss.co.jp/index_en.php
Japanese:http://www.sraoss.co.jp


More information about the pgpool-hackers mailing list