Error: no buffer space available

Elastic Stack Beats
1

Am using Heartbeat to ICMP check 2500+ endpoints, and over double the number of TCP ports.

On a RHEL 6 bare metal server, there are no issues. But on an Ubuntu 18 VM, I'm starting to see a large number of these errors:

write ip4 0.0.0.0->x.y.z.w: sendto: no buffer space available

I tried this first:

sysctl -w net.ipv4.neigh.default.gc_thresh1=6144
sysctl -w net.ipv4.neigh.default.gc_thresh2=8192
sysctl -w net.ipv4.neigh.default.gc_thresh3=12288

No difference!

I then tried this:

heartbeat.scheduler.limit: 1048576 (Note: /proc/sys/fs/file-max = 1219493)

But still the same error!

Please can someone help here?

2

Interesting, I'm glad to take a look and see what we can do to fix this. Mind opening an issue @ https://github.com/elastic/beats ? It's not strictly necessary, but would be much appreciated.

3

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.