Our MySQL support engineers encountered a rather unusual cause of replication lag. We covered the most common causes of replication lag in our previous article about MySQL parallel replication. In this particular case, the root cause was more strange. The servers were very high spec with NVMe disks and 10Gbit Ethernet. Analysis showed that the […]
How our MariaDB / MySQL Support engineers resolved a mysterious problem of MariaDB freezing every few hours for no apparent reason.
New Shattered Silicon Monitoring (SSM) 126.96.36.199.5.3 release is now available with additional features and fixes
MySQL parallel replication is a useful feature, but there are some pitfalls to watch out for. Getting it wrong can cripple performance.
SSM is a fork of PMM 1.x with additional features, updates and fixes. Notable new features in the initial release include QAN query harvesting via slow log in RDS, manual node cleanup functionality, and ARM64 builds.
Troubleshooting performance degradation caused by pileup of queries stuck in “Waiting for table flush” state in MySQL and MariaDB.
Troubleshooting performance degradation caused by pileup of queries stuck in “Waiting for query cache lock” state in MySQL and MariaDB.
A summary of basics of MariaDB and MySQL performance tuning, along with a few more advanced hints and tips. Beyond these, further tuning becomes very workload specific.
When you DROP or TRUNCATE a table or partition in MySQL and you have a large buffer pool, you may noticed a complete server lock-up or stall of multiple seconds. Mitigation and Workarounds.