[Openstack] #239037: Simultaneous Post requests of large objects could fail with "503 Service Unavailable"

balu chandra balu9463 at gmail.com
Tue Dec 3 15:07:13 UTC 2013


Adding the server details,
I am using HP proliant DL380 G7 series with 10GbE NIC.

On 12/2/13, balu chandra <balu9463 at gmail.com> wrote:
> Hello,
> Mailing to a larger audience the question posted at
> https://answers.launchpad.net/swift/+question/239037
>
> I am testing for performance numbers on a 2 node default
> swift(grizzly) configuration. The results were not encouraging though.
>
> With 10 concurrent users when uploading 15 objects per user with a
> mixed object size varying from 1KB to 5GB, I see a failure rate of 10%
> with "503 internal server error" or "503 service unavailable error".
> The failures was not restricted to bigger sized objects.
> I am using 3 local disks on each server and having default
> configuration on the Auditor.
> Increasing the timeout to 60 seconds and more only marginally reduced
> the failure rate.
> I would be interested to know if these numbers are expected or there
> are possible configuration tweaks to improve the success rate for
> multiple parallel requests.
> Any performance results for swift that I could be pointed to would be
> of great help.
>




More information about the Openstack mailing list