frankjkelly opened a new issue #9241:
URL: https://github.com/apache/pulsar/issues/9241
**Describe the bug**
Under heavy load Pulsar Proxy exhibited
`io.netty.util.internal.OutOfDirectMemoryError`
**To Reproduce**
Have not yet been able to reproduce (we increased the number of proxies from
2 to 3 to work around this issue)
The Proxy is configured with -Xmx1G and the pod itself has 2G memory (limits
== requests)
**Expected behavior**
A clear and concise description of what you expected to happen.
**Screenshots**
Saw this in the log file
```
{"timestamp":1610637633834,"log":"15:20:33.834 [pulsar-proxy-io-2-1] WARN
org.apache.pulsar.proxy.server.ProxyConnection -
[/127.0.0.1:36988] Got exception OutOfDirectMemoryError : failed to allocate
16777216 byte(s) of direct memory (used: 1073741824, max:
1073741824) io.netty.util.internal.OutOfDirectMemoryError: failed to
allocate 16777216 byte(s) of direct memory (used: 1073741824, max:
1073741824)","stream":"stdout","time":"2021-01-14T15:20:33.834842879Z"}
```
Noticed this in Grafana Dashboard
**Additional context**
- OS: Linux
- Deployment: Kubernetes
- Pulsar: 2.6.1
CC: @addisonj
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
