Re: replication isn't replicating
El mar, 16 ene 2024 a las 22:47, Brad White () escribió: > Errors from the Primary server > > 2024-01-15 00:01:06.166 CST [1428] ERROR: requested WAL segment > 0001000200A2 has already been removed > 2024-01-15 00:01:06.166 CST [1428] STATEMENT: START_REPLICATION > 2/A200 TIMELINE 1 > 2024-01-15 00:01:11.158 CST [3472] ERROR: requested WAL segment > 0001000200A2 has already been removed > 2024-01-15 00:01:11.158 CST [3472] STATEMENT: START_REPLICATION > 2/A200 TIMELINE 1 > 2024-01-15 00:01:11.158 CST [3472] LOG: could not receive data from > client: An existing connection was forcibly closed by the remote host. > > These log entries mean that some node is requesting a WAL segment that was already removed from the server. > > 2024-01-15 00:01:16.166 CST [664] ERROR: requested WAL segment > 0001000200A2 has already been removed > 2024-01-15 00:01:16.166 CST [664] STATEMENT: START_REPLICATION 2/A200 > TIMELINE 1 > 2024-01-15 00:01:21.161 CST [2016] ERROR: requested WAL segment > 0001000200A2 has already been removed > 2024-01-15 00:01:21.161 CST [2016] STATEMENT: START_REPLICATION > 2/A200 TIMELINE 1 > 2024-01-15 00:01:21.161 CST [2016] LOG: could not receive data from > client: An existing connection was forcibly closed by the remote host. > > [repeat for 55 lines] > > Errors from the backup server > > 2024-01-15 01:13:57.893 CST [2988] LOG: started streaming WAL from > primary at 2/A200 on timeline 1 > 2024-01-15 01:13:57.893 CST [2988] FATAL: could not receive data from WAL > stream: ERROR: requested WAL segment 0001000200A2 has already > been removed > 2024-01-15 01:13:57.893 CST [1792] LOG: waiting for WAL to become > available at 2/A2002000 > 2024-01-15 01:14:02.884 CST [2552] LOG: started streaming WAL from > primary at 2/A200 on timeline 1 > 2024-01-15 01:14:02.884 CST [2552] FATAL: could not receive data from WAL > stream: ERROR: requested WAL segment 0001000200A2 has already > been removed > 2024-01-15 01:14:02.884 CST [1792] LOG: waiting for WAL to become > available at 2/A2002000 > > These are related to the backup not finding that segment, so it means you'll need to resync your backup stream. I assume that you're using barman and using https://docs.pgbarman.org/release/3.9.0/#streaming-backup . Hope it helped. -- -- Emanuel Calvo OnGres Database Engineer | ViaDB Founder
Re: Babelfish for PostgreSQL
> > > There is a method of simulating SQL Server by using PostgreSQL. The > extension is called "Babelfish": https://babelfishpg.org/ > > However, this extension requires modifications to the PostgreSQL source > and building the PostgreSQL with Babelfish modifications from source. While > it is a rather seamless process and goes on without much trouble, it is > still much more complicated than installing the RPM or DEB packages, > depending on your Linux distro of choice. Is there any chance that the > Babelfish modifications will be adopted in the PostgreSQL source proper? > From my DBA point of view, I didn't find any differences in the > functionality of the PostgreSQL database itself. Amazon RDS incorporates > that modification and allows creating Aurora PostgreSQL database with the > Babelfish extension from the menu. > Hi there, I'm currently part of the Babelfish Open Source team. Regarding the installation process it works, but is far from being smooth or ready to get OS packages (there is, although, an ongoing work to get Docker images support in the upcoming months). But most importantly, we won't see this into core anytime soon -- or maybe, we just won't see it as it is. There is a patch for supporting protocol extensibility, but that requires considerable rework on the Babelfish tds and tsql extensions if moving forward. There is a long thread discussing about this in the pgsql-hackers list: https://www.postgresql.org/message-id/flat/CAGBW59d5SjLyJLt-jwNv%2BoP6esbD8SCB%3D%3D%3D11WVe5%3DdOHLQ5wQ%40mail.gmail.com Even though there is intensive work from the AWS team to support and build distributions for Babelfish, it is still a great solution for migrating from SQL Server while keeping legacy applications working. Regards, -- -- Emanuel Calvo -- 3manuek.com OnGres Data Engineer / ViaDB Founder CanalDBA BDFL -- Sumate al Slack canaldba.com! curl -s https://keybase.io/3manuek/key.asc
