Bug 1028795

Summary: There should be no downtime for a scaled app
Product: OpenShift Online Reporter: Oleg Fayans <ofayans>
Component: ContainersAssignee: Jhon Honce <jhonce>
Status: CLOSED NOTABUG QA Contact: libra bugs <libra-bugs>
Severity: high Docs Contact:
Priority: unspecified    
Version: 1.xCC: amarecek, bmeng
Target Milestone: ---   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2013-11-11 03:12:59 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description Oleg Fayans 2013-11-10 19:57:17 UTC 
Description of problem:

When I restart or git push the scaled app during the load, there are several seconds when the app is unavalable. In fact, it happens even without any load

Version-Release number of selected component (if applicable):
reproduced on devenv-stage_562

How reproducible:
always

Steps to Reproduce:
1. rhc app create myphp php-5.3 -s
2. scale up the app. Make sure the app has 2 gears
3. create some local content, git commit and push
4. During git push constantly check the app's url for availability

Actual results:

there are several moments when I get 503 - service unavailable

Expected results:
the app url should be available always

Additional info:
tcms testcase: 
https://tcms.engineering.redhat.com/case/318063/?from_plan=4962
Comment 1 Meng Bo 2013-11-11 03:12:59 UTC 
@ofayans

If we want to restart the app without down time, we should have at least two haproxies.

The case is out of date.

In your step, the app should have a small downtime to restart the haproxy cartridge. But you can see the web gears are rolling restart.