I asked this question here: https://devtalk.nvidia.com/default/topic/1042440/deepstream-sdk-2-0-on-centos-rhel-7/#5287545
, but it is probably more appropriate in this forum (Deepstream for Tesla).
Is there any plan to support CentOS/RHEL 7? All of the other dependencies are available for RHEL 7, i.e.
NVIDIA Display Driver ver. 410.72
NVIDIA Video SDK 8.2
NVIDIA cuDNN 7.3
NVIDIA TensorRT 5.0
NVIDIA CUDA 10.0
The DeepStream binaries built against libraries in Ubuntu 16.04 are my only impediment to deploying with our RHEL 7 based software.
Dear DeepStream SDK developers,
I too have this question. We are required to use CentOS/RHEL 7. We are not allowed to use Ubuntu.
Can you send your company name to me privately?
Hello,
We are also interested in working with CentOS/RHEL for our deployment. Compatible DeepStream binaries would really help us on this way!
Thank you.
Aurélien
Dear DeepStream SDK developers,
This topic appears stalled. What is the current status of making DeepStream binaries for Centos - RHEL?
We didn’t test deepstream in Centos-RHEL. Maybe there are some problems when you run it in Centos-RHEL.
Dear Chris Ding,
Thank you very much for the reply.
Here is a related link:
https://devtalk.nvidia.com/default/topic/1042440/deepstream-sdk-2-0-on-centos-rhel-7/
From this link, the issue raised with using code compiled for Ubuntu on RHEL - Centos 7:
“…are built against Ubuntu’s version of libstdc++, so CXXABI and GLIBCXX do not match up.”
I have not invested the time to make a spec file for building a rpm because I am concerned that I will receive seg faults if the binaries are not compiled for the RHEL - Centos binaries.
Is this a valid concern?
Hi everybody.
Once again the topic is stalled. Any news about DeepStream for CentOS/RHEL 7?
Thanks.
You can run DS container on centos.