-
-
Notifications
You must be signed in to change notification settings - Fork 53
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Snuba consumes high cpu #5825
Comments
Thanks for writing in. Could you answer these questions before proceeding?
|
Hello @hubertdeng123 for question 1, I hope the attached screenshot answers your question. And for question 2, I am using t3a.xlarge as EC2 from aws, which has 4 vCPU and 16 GB of memory. Thanks, |
As somebody who also deploys Sentry on AWS, I would strongly advice against using t3a instances. Their baseline performance is too low, and the burst performance does not scale well with how Sentry operates. Look into using m7i instances, at least, or m7i-flex if minimising costs is a concern. |
Having the same issue on digitalocean droplet after upgrading to 24.3.0. CPU is constantly at 100% |
Can you see if the solution suggested here helps in this case as well? |
This issue has gone three weeks without activity. In another week, I will close it. But! If you comment or otherwise update it, I will reset the clock, and if you remove the label "A weed is but an unloved flower." ― Ella Wheeler Wilcox 🥀 |
Can you try the advice posted here? |
hmm actually runs worse without the rust, can see an extra 20% on top for the CPU utilisation from both sentry and arrouns 5% for snuba. have run the sentry-self-hosted-post-process-forwarder-errors-1 traces
sentry-self-hosted-metrics-consumer-1 traces
sentry-self-hosted-generic-metrics-consumer-1 traces
|
Moved this over to the snuba repo for more domain-specific triage. |
Related: getsentry/self-hosted#2931 (comment) |
Self-Hosted Version
24.3.0
CPU Architecture
x86_x64
Docker Version
25.0.4
Docker Compose Version
2.24.7
Steps to Reproduce
I have installed sentry self hosted on an aws EC2. The only different that I have made is the following.
In the process of creating the EC2, I attached a new HDD disk to the server and mounted it. I used HDD disk to mount 2 volumes sentry-kafka and sentry-postgres. I edited the following file install/create-docker-volumes.sh (attached to the issue).
create-docker-volumes.txt
So, now I can mount all the docker volumes data and storage of both sentry-kafka and sentry-postgres to the HDD disk.
The reason why I did this is to save some money and use huge disk of HDD instead of SSD because these volumes consume huge amount of disk space.
Expected Result
normal behavior and events receiving and issue monitoring.
Actual Result
no events no issues high cpu utilization even after uninstalling sentry completely and installing it again.
It worked for a while around day or day and half, then sentry stops to report. No new events or issues or transactions.
I had to reset it and re-install it from the beginning but now even before adding any project cpu is very high above 90%.
I tried to add some logs from the docker compose logs command output.
logs.txt
So any idea regrading what to do next?
Thanks,
Event ID
No response
The text was updated successfully, but these errors were encountered: