Welcome!

Machine Learning Authors: Pat Romanski, Yeshim Deniz, Liz McMillan, Elizabeth White, Zakia Bouachraoui

Related Topics: Java IoT, Microservices Expo, Linux Containers, Machine Learning , Agile Computing, @DXWorldExpo

Java IoT: Article

Sync Your Timeouts: When Load Balancers Cause Database Deadlocks

Have you seen this error message before “java.sql.Exception: ORA-00060: deadlock detected while waiting for resource”?

Have you seen this error message before "java.sql.Exception: ORA-00060: deadlock detected while waiting for resource"?

This is caused when parallel updates require locks on either rows or tables in your database. I recently ran into this exception on an instance of an IBM eCommerce Server. The first thought was that there are simply too many people hitting the same functionality that updates Sales Tax Summary information - which was showing up in the call stack of the exception:

Exception stack trace showing that createOrderTaxes ran into the deadlock issue on the database

The logical conclusion would be to blame this on too many folks accessing this functionality or outdated table statistics causing update statements to run too long causing others to run into that lock. It turned out to be caused by something that wasn't that obvious and wouldn't have shown up in any Exception stack traces or log files. A misconfigured timeout setting on the load balancer caused a re-execute of the original incoming web request. While the first app server was still updating the table and holding the lock - as it had a longer timeout specified as the load balancer - the second app server tried to do the same thing causing that exception.

In this article I'll show you the steps necessary to analyze the symptoms (timeouts and client errors) and to identify and fix the root cause of the problem.

Step #1: Identifying Who and What Is Impacted
Identifying failing results is easy - start by looking at HTTP Response Codes (Web Server Access Log), Severe Log Messages (App Server Logs) or problematic Exception objects (either in log files or other monitoring tools). In our case I identified the SQL Lock Exception, a corresponding severe log message and the resulting HTTP 500 and also traced it back to the individual users and their actions that caused these issues.

Linking the errors to the User Action reveals that the problem happens when adding items to the shopping cart

This impacts our business.

Now we know that this problem impacts a critical feature in our app: Users can't add items to their cart.

Step #2: Understanding the Transaction Flow
Before drilling deeper I typically get an overview of the flow of the transaction from the browser all the way back to the database. This high-level view lets me understand which application components are involved and how they are interconnected. The transaction flow in this case highlights some interesting issues with the "Add Item to Cart" click. It appears to execute more than 33k SQL Statements for this single user interaction, causing 45% time executed just in Oracle:

Transaction Flow highlights several hotspots such as 33k SQL Executions in Total and Load Balancer (IHS) splitting up a request

This Is an Architectural Problem
Getting the full end-to-end execution path for a single user interaction (Add Item to Cart) and seeing how it "branches" out makes it really obvious how these individual problems (too many SQL, High Execution Time, ...) end up impacting end users. Just spotting the individual hotspots without having this connection would make it harder to understand the real root cause.

For steps 3 & 4, and for a list of key takeaways, click here for the full article

More Stories By Andreas Grabner

Andreas Grabner has been helping companies improve their application performance for 15+ years. He is a regular contributor within Web Performance and DevOps communities and a prolific speaker at user groups and conferences around the world. Reach him at @grabnerandi

Comments (0)

Share your thoughts on this story.

Add your comment
You must be signed in to add a comment. Sign-in | Register

In accordance with our Comment Policy, we encourage comments that are on topic, relevant and to-the-point. We will remove comments that include profanity, personal attacks, racial slurs, threats of violence, or other inappropriate material that violates our Terms and Conditions, and will block users who make repeated violations. We ask all readers to expect diversity of opinion and to treat one another with dignity and respect.


CloudEXPO Stories
With more than 30 Kubernetes solutions in the marketplace, it's tempting to think Kubernetes and the vendor ecosystem has solved the problem of operationalizing containers at scale or of automatically managing the elasticity of the underlying infrastructure that these solutions need to be truly scalable. Far from it. There are at least six major pain points that companies experience when they try to deploy and run Kubernetes in their complex environments. In this presentation, the speaker will detail these pain points and explain how cloud can address them.
The deluge of IoT sensor data collected from connected devices and the powerful AI required to make that data actionable are giving rise to a hybrid ecosystem in which cloud, on-prem and edge processes become interweaved. Attendees will learn how emerging composable infrastructure solutions deliver the adaptive architecture needed to manage this new data reality. Machine learning algorithms can better anticipate data storms and automate resources to support surges, including fully scalable GPU-centric compute for the most data-intensive applications. Hyperconverged systems already in place can be revitalized with vendor-agnostic, PCIe-deployed, disaggregated approach to composable, maximizing the value of previous investments.
When building large, cloud-based applications that operate at a high scale, it's important to maintain a high availability and resilience to failures. In order to do that, you must be tolerant of failures, even in light of failures in other areas of your application. "Fly two mistakes high" is an old adage in the radio control airplane hobby. It means, fly high enough so that if you make a mistake, you can continue flying with room to still make mistakes. In his session at 18th Cloud Expo, Lee Atchison, Principal Cloud Architect and Advocate at New Relic, discussed how this same philosophy can be applied to highly scaled applications, and can dramatically increase your resilience to failure.
Machine learning has taken residence at our cities' cores and now we can finally have "smart cities." Cities are a collection of buildings made to provide the structure and safety necessary for people to function, create and survive. Buildings are a pool of ever-changing performance data from large automated systems such as heating and cooling to the people that live and work within them. Through machine learning, buildings can optimize performance, reduce costs, and improve occupant comfort by sharing information within the building and with outside city infrastructure via real time shared cloud capabilities.
As Cybric's Chief Technology Officer, Mike D. Kail is responsible for the strategic vision and technical direction of the platform. Prior to founding Cybric, Mike was Yahoo's CIO and SVP of Infrastructure, where he led the IT and Data Center functions for the company. He has more than 24 years of IT Operations experience with a focus on highly-scalable architectures.