cloudstack-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Likitha Shetty (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (CLOUDSTACK-5485) Vmware - Whe 10 hourly snapshots are scheduled at the same time , we see only 5 of them being processed actively at the same time.
Date Fri, 27 Dec 2013 12:32:52 GMT

    [ https://issues.apache.org/jira/browse/CLOUDSTACK-5485?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13857465#comment-13857465
] 

Likitha Shetty commented on CLOUDSTACK-5485:
--------------------------------------------

Analysis
----------
For agent communication in Management Sever on port 8250 and in SSVM, CPVM and KVM hosts an
Indirect agent communication thread pool is used.
The name of the thread pool in MS is AgentManager- and in SSVM is Agent-

In MS, size of the above thread pool is _workers_+10 where _workers_ is a global configuration
parameter with 5 as the default value.
But In SSVM, the size of the thread pool is not configurable using this global config. With
5 as default, only 5 snapshots can be backed up to secondary storage simultaneously.

Workaround
--------------- 
To increase the number of simultaneous commands processed in SSVM(increase the count of worker
threads), modify the agent properties file in SSVM to specify the number of _workers_.
To backup say 10 snapshots simultaneously to secondary storage -
1. service cloud stop
2. update /usr/local/cloud/systemvm/conf/agent.properties file in SSVM to add line _workers=10_
3. service cloud start

> Vmware - Whe 10 hourly snapshots are scheduled at the same time , we see only 5 of them
being processed actively at the same time. 
> -----------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: CLOUDSTACK-5485
>                 URL: https://issues.apache.org/jira/browse/CLOUDSTACK-5485
>             Project: CloudStack
>          Issue Type: Bug
>      Security Level: Public(Anyone can view this level - this is the default.) 
>          Components: Management Server
>    Affects Versions: 4.3.0
>         Environment: Build from 4.3
>            Reporter: Sangeetha Hariharan
>            Assignee: Likitha Shetty
>            Priority: Critical
>             Fix For: 4.3.0
>
>         Attachments: vmware.rar
>
>
> Vmware - When 10 hourly snapshots are scheduled at the same time , we see only 5 of them
being processed actively at the same time.
> Set up :
> Advanced Zone with 2 5.1 ESXI hosts.
> Steps to reproduce the problem:
> 1. Deploy 5 Vms in each of the hosts , so we start with 10 Vms.
> 2. Start concurrent snapshots for ROOT volumes of all the Vms.
> Noticed that on the Vsphere client , only 5 of these snapshots gets executed in parallel.Rest
of them get picked up when the current  snapshot jobs complete.
> Why cant we do more than 5 parallel tasks ?



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

Mime
View raw message