Detailed bug description:
Syslog
Steps to reproduce:
1. Create environment with: 3 controller, 20 ceph-osd, 1000 computes
2. Start deployment
Expected results:
Logs from all nodes collected on Fuel node
Actual result:
Logs collected not from all nodes. Syslog use near of 100% of CPU
Reproducibility:
Each time
Workaround:
Increase NOFILE limit^
1. On fuel node create file /etc/systemd/system/rsyslog.service.d/limits.conf with content:
[Service]
LimitNOFILE=16384
2. execute systemctl daemon-reload
Impact:
Fuel can't get logs from all nodes
Description of the environment:
Operation system: Ubuntu-14
Versions of components: Mitaka
Reference architecture: 3 controller, 20 ceph-osd, 1000 computes
Network model: VXLAN
Related projects installed: none
Detailed bug description: system/ rsyslog. service. d/limits. conf with content:
Syslog
Steps to reproduce:
1. Create environment with: 3 controller, 20 ceph-osd, 1000 computes
2. Start deployment
Expected results:
Logs from all nodes collected on Fuel node
Actual result:
Logs collected not from all nodes. Syslog use near of 100% of CPU
Reproducibility:
Each time
Workaround:
Increase NOFILE limit^
1. On fuel node create file /etc/systemd/
[Service]
LimitNOFILE=16384
2. execute systemctl daemon-reload
Impact:
Fuel can't get logs from all nodes
Description of the environment:
Operation system: Ubuntu-14
Versions of components: Mitaka
Reference architecture: 3 controller, 20 ceph-osd, 1000 computes
Network model: VXLAN
Related projects installed: none