aurora-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Maxim Khutornenko (JIRA)" <j...@apache.org>
Subject [jira] [Reopened] (AURORA-208) Add sla_list_safe_domain command into aurora_admin client
Date Fri, 21 Feb 2014 23:17:21 GMT

     [ https://issues.apache.org/jira/browse/AURORA-208?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Maxim Khutornenko reopened AURORA-208:
--------------------------------------


> Add sla_list_safe_domain command into aurora_admin client
> ---------------------------------------------------------
>
>                 Key: AURORA-208
>                 URL: https://issues.apache.org/jira/browse/AURORA-208
>             Project: Aurora
>          Issue Type: Task
>          Components: Client
>            Reporter: Maxim Khutornenko
>            Assignee: Maxim Khutornenko
>
> sla_list_safe_domain
> Usage: sla_list_safe_domain --cluster=cluster --attribute={rack | host} percentage duration
[--override_jobs=filename] [--exclude_attr=filename][--list_jobs]
> Returns a list of racks or hosts where it would be safe to kill tasks without violating
their job SLA: percentage of tasks that stayed up within the last “duration” secs|mins|hrs|days.
The SLA can be specified globally per cluster as pair of percentage and duration values or
per job in a file. 
> --cluster: 
> Aurora cluster name.
> --attribute: 
> Currently supported attributes “host” or “rack”.
> percentage:
> Percentage of tasks required to be up within the duration. Applied to all jobs except
those listed in --override file. 
> duration:
> Time interval (now - value) for the percentage of up tasks. Applied to all jobs except
those listed in --override_jobs file. Format: <value>{secs|mins|hrs|days}.
> --override_jobs: 
> An optional file to load job specific SLAs that will override cluster-wide command line
percentage and duration values. The file can have multiple lines in the following format:
> role/env/job percentage duration
> --exclude_attr:
> An optional text file listing attribute values (one per line) to exclude from the result
set if found.
> --list_jobs:
> Lists all affected job keys with projected new SLAs if their tasks get killed.
> Examples: 
> sla_list_safe_domain --cluster=smf1 --attribute=host 85 10mins
> sla_list_safe_domain --cluster=smf1 --filename=~/rack.txt
> Example (--attribute=rack):
> aurora_admin list_safe_sla_domain --cluster=cluster --attribute=rack 95 2hrs
> rack1 host1
> rack2 host2
> rack2 host3
> rack3 host1
> Example output (--attribute=host):
> aurora_admin list_safe_sla_domain --cluster=cluster --attribute=host 95 2hrs
> host1
> host2
> host3
> host4
> Example output (--list_jobs):
> aurora_admin list_safe_sla_domain --cluster=cluster --attribute=rack 95 2hrs --list_jobs
> rack1 host1 cluster/role/prod/job1 	96.00 2hrs
> rack2 host2 cluster/role/prod/job2 	97.65 2hrs
> rack2 host2 cluster/role/prod/job3 	95.05 2hrs



--
This message was sent by Atlassian JIRA
(v6.1.5#6160)

Mime
View raw message