Galera | BestHub

Qunar Tech Salon

Oct 28, 2021 · Databases

Analysis of a Percona XtraDB Cluster (PXC) Flow‑Control Incident Triggered by Disk I/O and Large TEXT Fields

This article examines a real‑world Percona XtraDB Cluster (PXC) fault where simultaneous write‑node connection spikes were caused by flow‑control triggered by 100% disk I/O on a shared read node, ultimately traced to slow sorting queries involving a mediumtext column, and offers diagnostic steps and mitigation strategies.

Database performanceGaleraMySQL

0 likes · 11 min read

Analysis of a Percona XtraDB Cluster (PXC) Flow‑Control Incident Triggered by Disk I/O and Large TEXT Fields

Qunar Tech Salon

Nov 24, 2020 · Databases

Analysis of Galera GTID vs MySQL GTID Issues During Percona XtraDB Cluster Split at Qunar

This article examines how differences between Galera GTID and MySQL GTID caused replication interruptions during a Percona XtraDB Cluster (PXC) split at Qunar, details the migration procedure, reproduces the problem with step‑by‑step commands, explains GTID concepts, and proposes operational improvements to avoid data inconsistency.

Database OperationsGTIDGalera

0 likes · 17 min read

Analysis of Galera GTID vs MySQL GTID Issues During Percona XtraDB Cluster Split at Qunar

Didi Tech

Jul 27, 2018 · Databases

Root Cause and Fix for Galera MySQL 5.7.17 SST Failure Caused by innodb_undo_table_spaces with RSYNC

The Galera MySQL 5.7.17 SST fails when a pre‑initialized node is added because the rsync‑based SST filter omits the independent undo‑log tablespace files (undo*), causing InnoDB to roll back uncommitted transactions; adding an undo* rule to the filter copies the missing files and resolves the error, a fix now upstream.

DatabaseGaleraMySQL

0 likes · 4 min read

Root Cause and Fix for Galera MySQL 5.7.17 SST Failure Caused by innodb_undo_table_spaces with RSYNC

Architect

Sep 14, 2015 · Databases

MySQL High Availability Solutions: Selection Guide

This article reviews various MySQL high‑availability architectures—including master‑slave replication with keepalived or MHA, Galera‑based clusters, and other approaches—detailing their deployment methods, advantages, limitations, and practical considerations for selecting an appropriate HA solution.

DatabaseGaleraMHA

0 likes · 9 min read

MySQL High Availability Solutions: Selection Guide