Closed Bug 713051 Opened 14 years ago Closed 14 years ago

[tracking-bug]log aggregation for release engineering infrastructure

Categories

(Release Engineering :: General, defect, P5)

Product:
Release Engineering
Component:
General
Platform:
x86
macOS
Type:
defect

Tracking

(Not tracked)

Status:
RESOLVED WONTFIX

People

(Reporter: jhford, Assigned: jhford)

Details

We have lots of systems that generate lots of logs on lots of machines in lots of colos. When something goes wrong, its a real pain to have to figure out which machine has the error, how to get to it and where the log is on the server. In the case of catastrophic failures, we might not be able to recover logs from the machines. Why don't we start streaming logs from our machines to a centralized facility? I am thinking of things like the system logs on masters, the twistd.log from masters, puppet logs from puppet masters. Sending slave logs would be great, but is probably quite a large scaling problem. This should make it easier to check logs on machines. Additionally, this would allow us to start monitoring all of our logs for metrics and possibly early problem detection. Because this is fairly open ended, I am filing this as a tracking bug. Actual work should be done in dependent bugs.
Status: NEW → RESOLVED
Closed: 14 years ago
Resolution: --- → WONTFIX
Product: mozilla.org → Release Engineering
You need to log in before you can comment on or make changes to this bug.