Skip to content
GitLab
  • Menu
Projects Groups Snippets
  • Help
    • Help
    • Support
    • Community forum
    • Submit feedback
    • Contribute to GitLab
  • Sign in / Register
  • Icecast-Server Icecast-Server
  • Project information
    • Project information
    • Activity
    • Labels
    • Members
  • Repository
    • Repository
    • Files
    • Commits
    • Branches
    • Tags
    • Contributors
    • Graph
    • Compare
  • Issues 102
    • Issues 102
    • List
    • Boards
    • Service Desk
    • Milestones
  • Merge requests 5
    • Merge requests 5
  • CI/CD
    • CI/CD
    • Pipelines
    • Jobs
    • Schedules
  • Deployments
    • Deployments
    • Environments
    • Releases
  • Packages & Registries
    • Packages & Registries
    • Package Registry
    • Infrastructure Registry
  • Monitor
    • Monitor
    • Incidents
  • Analytics
    • Analytics
    • Value stream
    • CI/CD
    • Repository
  • External wiki
    • External wiki
  • Activity
  • Graph
  • Create a new issue
  • Jobs
  • Commits
  • Issue Boards
Collapse sidebar
  • Xiph.Org
  • Icecast-ServerIcecast-Server
  • Issues
  • #2293
Closed
Open
Created Oct 21, 2016 by Sander@cloudseer

High CPU usage on macOS Sierra

I am experiencing high CPU (95%) on macOS when connecting to a local Icecast server. The high CPU starts when a source connects, and remains even afer sources are no longer connected, until I restart the server.

Versions tested:

  1. homebrew 2.4.2 with my custom config,
  2. compiled from source 2.4.3 with default config.

When running within a virtualbox Ubuntu server instance, Icecast is behaving nicely.

I've ran a Instruments time + counter trace, data is available from https://dl.dropboxusercontent.com/u/70903/Icecast%20Instruments.zip and attached to this ticket.

It seems to show that 97% of the time fserve_client_waiting is doing a poll, not sure if this is normal I'm not experienced with this subject. 1 every millisecond if I'm reading the counter correctly.

[12:28] <+tbr> I guess this is something we should definitely look into.

Assignee
Assign to
Time tracking