Craig's PlanetPostgreSQL Archives - Page 2 of 5 - 2ndQuadrant

BDR History and future

October 18, 2016/0 Comments/in Craig's PlanetPostgreSQL /by craig.ringer

BDR is both a patch to PostgreSQL core and an extension on top of PostgreSQL core. How did that come about, and what’s it’s future? Development of BDR was initiated around the time PostgreSQL 9.2 was in development. Arguably earlier if you count things like the extension mechanism. The goal of BDR is, and has […]

BDR is coming to PostgreSQL 9.6

September 21, 2016/39 Comments/in Craig's PlanetPostgreSQL /by craig.ringer

I’m pleased to say that Postgres-BDR is on its way to PostgreSQL 9.6, and even better, it works without a patched PostgreSQL. BDR has always been an extension, but on 9.4 it required a heavily patched PostgreSQL, one that isn’t fully on-disk-format compatible with stock community PostgreSQL 9.4. The goal all along has been to […]

BDR 1.0

August 12, 2016/7 Comments/in Craig's PlanetPostgreSQL /by craig.ringer

I’m pleased to say that we’ve just released Postgres-BDR 1.0, based on PostgreSQL 9.4.9. This release contains significant improvements to DDL replication locking, global sequences, documentation, performance, and more. It also removes the deprecated UDR component in favour of

How to check the lock level taken by operations in PostgreSQL

July 15, 2016/0 Comments/in Craig's PlanetPostgreSQL /by craig.ringer

PostgreSQL’s manual is generally pretty clear about the locks taken by various operations – but nothing’s perfect. If you need to, you can also ask PostgreSQL directly. You can check lock levels trivially with psql or PgAdmin.

PostgreSQL-based application performance: latency and hidden delays

May 23, 2016/2 Comments/in Craig's PlanetPostgreSQL, PostgreSQL /by craig.ringer

If you’re trying to optimise the performance of your PostgreSQL-based application you’re probably focusing on the usual tools: EXPLAIN (BUFFERS, ANALYZE), pg_stat_statements, auto_explain, log_statement_min_duration, etc. Maybe you’re looking into lock contention with log_lock_waits, monitoring your checkpoint performance, etc too. But did you think about network latency? Gamers know about network latency, but did you think […]

Visual PostgreSQL debugging

May 3, 2016/2 Comments/in Craig's PlanetPostgreSQL /by craig.ringer

Having recently been doing some debugging work where many watchpoints, conditional breakpoints etc were necessary I’d like to shout out to a really useful tool: The standalone CDT debugger. It’s part of the Eclipse project, but before you run screaming – it doesn’t require project setup or anything and it serves as a good GUI […]

Don’t set fsync=off if you want to keep your data

April 28, 2016/3 Comments/in Craig's PlanetPostgreSQL /by craig.ringer

There are a lot of amazing features coming in PostgreSQL 9.6, but I’m personally very happy about a really small, simple one that helps close a long-standing user foot-gun. commit a31212b429cd3397fb3147b1a584ae33224454a6 Author: Robert Haas Date: Wed Apr 27 13:46:26 2016 -0400 Change postgresql.conf.sample to say that fsync=off will corrupt data. Discussion: [email protected] Per a suggestion […]

What is SKIP LOCKED for in PostgreSQL 9.5?

April 13, 2016/25 Comments/in Craig's PlanetPostgreSQL /by craig.ringer

PostgreSQL 9.5 introduces a new SKIP LOCKED option to SELECT … FOR [KEY] UPDATE|SHARE. It’s used in the same place as NOWAIT and, like NOWAIT, affects behaviour when the tuple is locked by another transaction. The main utility of SKIP LOCKED is for building simple, reliable and efficient concurrent work queues.

LVM: “Unable to create a snapshot smaller than 2 chunks”

April 12, 2016/0 Comments/in Craig's PlanetPostgreSQL /by craig.ringer

If you try to create an LVM snapshot with lvm lvcreate -s -n snapname -L 100G myvg and it fails with Unable to create a snapshot smaller than 2 chunks. … it’s probably actually trying to tell you “I don’t have 100GB free on this VG, allocate a smaller snapshot”. It’s choosing to phrase it […]

Failover slots for PostgreSQL

February 15, 2016/15 Comments/in Craig's PlanetPostgreSQL /by craig.ringer

Logical decoding and logical replication is getting more attention in the PostgreSQL world. This means we need it working well alongside production HA systems – and it turns out there’s a problem there. Replication slots are not themselves synced to physical replicas so you can’t continue to use a slot after a master failure results […]