cloudstack-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Joe Brockmeier" <...@zonker.net>
Subject Re: Review Request: CLOUDSTACK-605: Host physical CPU is incorrectly calculated for Vmware hosts
Date Wed, 09 Jan 2013 20:05:14 GMT

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/8511/#review15209
-----------------------------------------------------------

Ship it!


This looks sane to me. Thanks much for the patch! 

- Joe Brockmeier


On Dec. 11, 2012, 1:11 p.m., Koushik Das wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/8511/
> -----------------------------------------------------------
> 
> (Updated Dec. 11, 2012, 1:11 p.m.)
> 
> 
> Review request for cloudstack and Kelven Yang.
> 
> 
> Description
> -------
> 
> There is a mismatch in host CPU count displayed in CS UI (displayed under the 'Statistics'
tab when a host is selected) compared to that in vCenter. The mismatch happens for ESX hosts
having > 1 cpu package. This will affect VM allocation logic.
>  
> Currently host CPU is computed based on below logic (HostHardwareSummary is defined in
Vmware library)
>  
> HostHardwareSummary hardwareSummary = getHostHardwareSummary(); 
> summary.setCpuCount(hardwareSummary.getNumCpuCores()*hardwareSummary.getNumCpuPkgs());
>  
> The correct logic is to use hardwareSummary.getNumCpuCores() only. 
> 
> 
> This addresses bug CLOUDSTACK-605.
> 
> 
> Diffs
> -----
> 
>   vmware-base/src/com/cloud/hypervisor/vmware/mo/HostMO.java 3557048 
> 
> Diff: https://reviews.apache.org/r/8511/diff/
> 
> 
> Testing
> -------
> 
> Verified on esx hosts with 1 and 2 CPU packages.
> 
> 
> Thanks,
> 
> Koushik Das
> 
>


Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message