Jean-Baptiste Denis

Hello, it looks like the same is happening with dsync on my side with release 0.11.1: ``` [2022-05-24T21:56:55] Copied 7.226 TiB (46%) in 1825.398 secs (4.053 GiB/s) 2169 secs left...

After some time, all data has been copied. dsync on MPI nodes are then not doing anything, looping at 100% cpu usage: ``` [pid 715200] epoll_wait(26, [], 16, 0) =...

Hello, Just updating the issue to say that I've been hit by this behavior. It would be nice to have it corrected of course ;) In the meantime as a...

Hi @mhx, no worries at all for the late response. Thank you for having taking the time to answer. Your explanation are quite clear and the suggestion of using an...

Sorry for the inconvenience. You can try msrsync3 at the root of the project folder. On 1/18/24 18:08, falves1 wrote: > I need a version that does not require Python...

Thank you for the kind words ! This program would also need more love from my side. The initial scope was limited to python 2.6 for RHEL 6.x type distribution...

Could you try without “*” ?

Hello, fwiw, I had the problem in 1.4.3 because the scontrol was not in the PATH and the auto-detection didn't work in nhcmain_find_rm. Setting NHC_RM in /etc/sysconfig/nhc worked for me.

I've ran the test again without using checksums in the aggregate form fields, and I've got the same behaviour : ``` upload_set_form_field $upload_field_name '{"filename":"$upload_file_name", "content_type": "$upload_content_type", "path": "$upload_tmp_path"}'; ``` Do...

Please find attached the output of cache-info with and without the nosmt option, when cpuinfo is built with debug log level. [cache-info.nosmt.txt](https://github.com/pytorch/cpuinfo/files/15025313/cache-info.nosmt.txt) [cache-info.smt.txt](https://github.com/pytorch/cpuinfo/files/15025314/cache-info.smt.txt)