jclouds-notifications mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "ASF subversion and git services (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (JCLOUDS-450) Support min/max count and client token fields for AWS EC2 RunInstances
Date Wed, 12 Feb 2014 05:59:20 GMT

    [ https://issues.apache.org/jira/browse/JCLOUDS-450?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13898830#comment-13898830

ASF subversion and git services commented on JCLOUDS-450:

Commit bf4626016b426d1fbc0e3e4d4e7499ea3aca9ac2 in branch refs/heads/cleanup-5976159-1.7.x
from [~abayer]
[ https://git-wip-us.apache.org/repos/asf?p=jclouds.git;h=bf46260 ]

JCLOUDS-450. Adding support for EC2 MaxCount option.

Note - no live test for this, because it's very hard to guarantee a
situation where it would be relevant.

> Support min/max count and client token fields for AWS EC2 RunInstances
> ----------------------------------------------------------------------
>                 Key: JCLOUDS-450
>                 URL: https://issues.apache.org/jira/browse/JCLOUDS-450
>             Project: jclouds
>          Issue Type: Improvement
>          Components: jclouds-compute
>            Reporter: Andrew Bayer
>            Assignee: Andrew Bayer
>             Fix For: 1.8.0, 1.7.2
> AWS's RunInstances API method (http://docs.aws.amazon.com/AWSEC2/latest/APIReference/ApiReference-query-RunInstances.html)
allows for specifying minimum and maximum count - if given, you'll get at least the minimum
number and no more than the maximum, unless the minimum is greater than the number of instances
you can launch in that AZ/region (say, if you've already got enough instances of a size that's
got a per-region limit such that creating min count more would push over the limit), in which
case you don't get any. It'd be nice to support this, both to simplify multiple instance creation
(rather than X separate calls to RunInstances, just one...) and to take advantage of the behavior
if you try to launch more than you can.
> As well, we should have some form of support for RunInstances' idempotency - see http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/Run_Instance_Idempotency.html
- if you pass a client token to the call, each subsequent call with the same arguments (including
the client token) will return the same response, only varying in the HTTP response code. So
you can keep making the same call (say, with a > 1 min count) until you get a terminating

This message was sent by Atlassian JIRA

View raw message