Inconsistent DB data in Streaming Replication

Поиск
Список
Период
Сортировка
От Samrat Revagade
Тема Inconsistent DB data in Streaming Replication
Дата
Msg-id CAF8Q-Gxg3PQTf71NVECe-6OzRaew5pWhk7yQtbJgWrFu513s+Q@mail.gmail.com
обсуждение исходный текст
Ответы Re: Inconsistent DB data in Streaming Replication  (Shaun Thomas <sthomas@optionshouse.com>)
Re: Inconsistent DB data in Streaming Replication  (Tom Lane <tgl@sss.pgh.pa.us>)
Re: Inconsistent DB data in Streaming Replication  (Fujii Masao <masao.fujii@gmail.com>)
Re: Inconsistent DB data in Streaming Replication  (Hannu Krosing <hannu@2ndQuadrant.com>)
Список pgsql-hackers
<div dir="ltr"><p class="">Hello,<p class="">We have been trying to figure out possible solutions to the following
problemin streaming replication Consider following scenario:<p class="">If master receives commit command, it writes
andflushes commit WAL records to the disk, It also writes and flushes data page related to this transaction.<p
class="">Themaster then sends WAL records to standby up to the commit WAL record. But before sending these records if
failoverhappens then,  old master is ahead of  standby which is now the new master in terms of DB data leading to
inconsistentdata . <p class=""> <p class="">One solution to avoid this situation is have the master send WAL records to
standbyand wait for ACK from standby committing WAL files to disk and only after that commit data page related to this
transactionon master.<p class="">The main drawback would be increased wait time for the client due to extra round trip
tostandby before master sends ACK to client. Are there any other issues with this approach?<p class=""><br /><p
class="">Thankyou,<p class=""><p class="">Samrat</div> 

В списке pgsql-hackers по дате отправления:

Предыдущее
От: Greg Jaskiewicz
Дата:
Сообщение: Re: unused code in float8_to_char , formatting.c ?
Следующее
От: Shaun Thomas
Дата:
Сообщение: Re: Inconsistent DB data in Streaming Replication