Indirect Checkpoint And Non-Yielding Scheduler Problems

Parikshit Savjani has a post describing an issue you might experience with indirect checkpoint post SQL Server-2012:

One of the scenarios where skewed distribution of dirty pages in the DPList is common is tempdb. Starting SQL Server 2016, indirect checkpoint is turned ON by default with target_recovery_time set to 60 for model database. Since tempdb database is derived from model during startup, it inherits this property from model database and has indirect checkpoint enabled by default. As a result of the skewed DPList distribution in tempdb, depending on the workload, you may experience excessive spinlock contention and exponential backoffs on DPList on tempdb. In scenarios when the DPList has grown very long, the recovery writer may produce a non-yielding scheduler dump as it iterates through the long list (20k-30k) and tries to acquire spinlock and waits with exponential backoff if spinlock is taken by multiple IOC routines for removal of pages.

This is worth taking a close read.

Related Posts

Sending SQL Server Notifications To Slack

Alessandro Alpi shows how to integrate SQL Server notifications with Slack: Now, how can we send notifications from SQL Server in an easier way than using custom code or a Slack incoming webhook? Is there any integration or a Slack app?  Yes. And guess what? I think you’ll like it because you don’t need to write […]

Read More

Running SQL Server On Cluster Shared Volumes

Sreekanth Bandarla continues a series on clustered shared volumes: In the previous part of this series, we have seen what a cluster shared volume is and what are the advantages and other considerations to keep in mind when deploying CSVs for SQL Server workloads. In this article, I will walk though actual installation of a failover cluster Instance […]

Read More

Categories

June 2017
MTWTFSS
« May Jul »
 1234
567891011
12131415161718
19202122232425
2627282930