Hybris 6.1 overall CPU crosses 100% utilization

[former_member1354369]

Hello Hybris Experts,

Greetings!

Hybris overall CPU consumption reaches 100% and then results the eCommerce sites are getting down.

Is any way to track and analyze?

Is any solution to get into normal level ?

Appreciate your help!

Thanks in advance

BR R Venkata karthik

[former_member624549]

Check what cronjobs are running. Other than monitoring the logfile and guessing whats going on you need a java thread dump and check whats running. i Assume your site is not under load.

[Former Member]

Hi

This is common.

Few suggestions from my side based on experience :

1. Check the cache settings as suggested by above. Increase the entity cache max limit to double than what hybris default has.

2. Install softwares such as dynatrace, or get a SAAS offering from Datadog(or any APM tool) to analyse the application performance. Its very difficult and time consuming to check logs and thread dumps. Such tools can help you identify issues in minutes. Then why waste hours reading logs? And that too at a very low price if you take monitoring for a month for one server.

3. This is the best of all : code to perform DTO caching. This will improve the site performance immensely as it directly correlates to the CPU utilisation. If you cache the product DTOs, then the CPU isn't used to unnecessarily convert the models to the Data objects.

4. This is key to verify if your changes works rather than doing hit and trial : Create a jmeter script to simulate the order place scenarios so that you can then verify if your fix actually works. Else we will never known if we have really solved the problem.

Hope this helps.

[former_member633554]

storefront.granule.enabled=false

[arvind-kumar_avinash]

Hi - the simplest thing you can do is to restart the server.

Some of the things I can suggest looking into are:

1. Please check the cache under hAC > Monitoring > Cache ( https://localhost:9002/monitoring/cache )

2. As has suggested, please check the running CronJobs and also the thread dump

3. Check if you find something in the performance tips given below:

• https://help.hybris.com/1811/hcd/8c72d89f866910148f3cdde41bec5ae9.html

• https://help.hybris.com/1811/hcd/8c2780e686691014a134cc8fd1091aa3.html

[former_member1354369]

Hi Arvind/Olivier,

we are restarting the server whenever the CPU crosses 100%, but, we are finding an alternative way, finding the solution what could be the cause is it due to indexing/more requests from end users or usage of Hybris perfory data base operations. Thank you for your notes.

BR R venkata karthik