-
Notifications
You must be signed in to change notification settings - Fork 3.6k
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
InfluxDB high memory usage and very slow to start #6759
Comments
@MatMeredith There are number of startup improvements in the currently nightlies. Can you try running a nightly? |
@jwilder Thank you -- I'll try this. Does the memory usage seem right though? When I restarted (with no database) memory use was negligible. Memory use has since been steadily growing. The database size has reached 4GB so far, and memory usage has increased to ~2GB. I assume it will continue increasing until it reaches the 5.5GB it was at previously. Is this expected? Is RAM usage expected to increase with total DB size? I do also have a couple of continuous queries running, aggregating / downsampling all of the data received. I don't know if this is relevant to any of the issues I'm seeing. |
It depends on the shape of your data and what queries are running. The more series and kinds of queries running will incur more RAM usage. |
Seems reasonable. But my number of series and queries haven’t changed since I restarted InfluxDB and yet RAM usage has increased steadily from ~500MB to 2GB and (based on my previous run) will continue to increase to 5.5GB. Would you expect this increase over time? From: Jason Wilder [mailto:notifications@github.com] It depends on the shape of your data and what queries are running. The more series and kinds of queries running will incur more RAM usage. — |
Slow startup should be fixed by #6743 |
I've been running a POC InfluxDB system (InfluxDB v0.12.1 git: 0.12 e094138) for ~a month. Over that time the DB has grown to ~15GB.
I'm logging 1000 measurements with just one tag (the name of the server that the measurement was collected from) and using Grafana to generate dashboards.
Looking at memory usage I noticed that InfluxDB was using ~5.5 GB so tried restarting the process to see if memory usage would be reduced.
"service influxdb restart" completed OK, but InfluxDB never started listening on port 8083 or responding to requests. I left it for ~30 minutes before giving up and deleting the DB. InfluxDB then started instantly.
Before deleting the DB I ran influxdb with the cpuprofile and memprofile options and left it running for a minute or so. Outupt attached if this is helpful.
This is obviously a critical issue that we need to resolve before we can take InfluxDB into production. Any ideas?
profile.zip
The text was updated successfully, but these errors were encountered: