Error message when starting deployment on large environment
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Fuel for OpenStack |
Confirmed
|
Medium
|
Fuel Sustaining | ||
8.0.x |
Won't Fix
|
Medium
|
MOS Maintenance | ||
Mitaka |
Confirmed
|
Medium
|
Fuel Sustaining |
Bug Description
When starting environment deployment on 200 nodes, message "A server error has occurred" appears on UI (screen-shot attached).
However, this error does not affect deployment process.
The reason is that Keep-Alive connection is timed out and nailgun receives 504 error from nginx:
[root@fuel ~]# zgrep error /var/log/
2016/02/17 18:32:54 [error] 310#0: *23422 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 172.18.194.211, server: localhost, request: "PUT /api/clusters/
2016/02/17 18:33:28 [error] 310#0: *29820 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 172.18.194.211, server: localhost, request: "GET /api/nodes/
2016/02/17 18:33:28 [error] 310#0: *29130 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 172.18.194.211, server: localhost, request: "GET /api/notificati
2016/02/17 18:33:36 [error] 310#0: *29821 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 172.18.194.211, server: localhost, request: "GET /api/nodes?
Changed in fuel: | |
status: | New → Confirmed |
importance: | Undecided → Critical |
importance: | Critical → Medium |
importance: | Medium → High |
assignee: | nobody → Fuel Python Team (fuel-python) |
milestone: | none → 8.0-updates |
tags: | added: area-python module-nailgun-agent |
no longer affects: | fuel/newton |
Has been reproduced with 9.0-188