[pgpool-general: 7559] Re: VIP with one node

[Tatsuo Ishii]

https://www.pgpool.net/docs/latest/en/html/planning.html#PLANNING-POSTGRESQL

"Thus we recommend to deploy Pgpool-II in more than 3 odd numbers."

(This should have been read as "Thus we recommend to deploy Pgpool-II
in more than or equal to 3 odd numbers." though.)

> Wow, first I heard of that --> "We do not recommend 2-node pgpool
> configuration. "
> 
> 
> 
> Tatsuo Ishii wrote on 5/5/2021 5:17 PM:
>>> Dear Tatsuo,
>>>
>>> thank you for your answer.
>>>
>>> I do recognize the dangers of a split-brain scenario.
>>>
>>> With a 2-node setup, however, it is possible to configure PgPool to
>>> have the VIP on one node while the other node is down (unreachable),
>>> even though it may very well be a split-brain situation:
>>>
>>> Node B loses connection to  Node A. Neither of them has the quorum (in
>>> fact there can never be a quorum in a 2-node cluster), still Node B
>>> will have the VIP as well as Node A.
>>>
>>> With that risk in mind, why can we not have such an option for
>>> clusters with more than 2 nodes?
>> We do not recommend 2-node pgpool configuration. The option for 2-node
>> which your are referring to is enable_consensus_with_half_votes (the
>> default is off). It was created just for a compromise because there
>> have been already too many 2-node users.
>>
>>> Regards,
>>>
>>> tamas
>>>
>>>
>>> 2021. 05. 05. 14:04 keltezéssel, Tatsuo Ishii írta:
>>>>> Dear all,
>>>>>
>>>>> I have fairly common 3-node cluster, with each node running a PgPool
>>>>> and a PostreSQL instance.
>>>>>
>>>>> I have set up priorities so that:
>>>>>     - when all 3 nodes are up, the 1st node is gonna have the VIP,
>>>>>     - when the 1st node is down, the 2nd node is gonna have the VIP, and
>>>>>     - when both the 1st and the 2nd nodes are down, then the 3rd node
>>>>> should get the VIP.
>>>>>
>>>>> My problem is that when only 1 node is up, the VIP is not brought up,
>>>>> because there is no quorum.
>>>>> How can I get PgPool to bring up the VIP to the only remaining node,
>>>> There's no way to do that.
>>>>
>>>>> which still could and should serve requests?
>>>> No, because it's dangerous. Think about a split brain scenario: node A
>>>> loses connection to node B and c. In this case node A loses the
>>>> quorum, while node B and C have the quorum and will keep the VIP. If
>>>> we allow the VIP to node A, there are 2 VIPs are up. We surely want to
>>>> avoid it.
>>>>
>>>> Best regards,
>>>> --
>>>> Tatsuo Ishii
>>>> SRA OSS, Inc. Japan
>>>> English: http://www.sraoss.co.jp/index_en.php
>>>> Japanese:http://www.sraoss.co.jp
>>>> _______________________________________________
>>>> pgpool-general mailing list
>>>> pgpool-general at pgpool.net
>>>> http://www.pgpool.net/mailman/listinfo/pgpool-general
>> _______________________________________________
>> pgpool-general mailing list
>> pgpool-general at pgpool.net
>> http://www.pgpool.net/mailman/listinfo/pgpool-general
>