Quantcast
Channel: Mellanox Interconnect Community: Message List
Viewing all articles
Browse latest Browse all 6211

Re: “Invalid module format” error while loading nv_peer_mem in CentOS 6.6

$
0
0

This is exactly what I'm thinking, with the addition that host2 has been working fine (with that module loaded) for almost 2 months.

OS and kernel are the same (CentOS 6.6, 2.6.32-504.el6.x86_64), also Nvidia drivers never changed so are still the same, as well as Infiniband driver.

These are the only differences in kernel modules (lsmod command), all the rest is identical:

                           HOST1

ib_core                  125047  13 nv_peer_mem,rdma_ucm,ib_ucm,rdma_cm,iw_cm,ib_ipoib,ib_cm,ib_uverbs,ib_umad,mlx5_ib,mlx4_ib,ib_sa,ib_mad

ib_umad                14390  4

ipv6                       334932  342 ip6t_REJECT,nf_conntrack_ipv6,nf_defrag_ipv6,ib_ipoib,ib_core,ib_addr

nvidia                    8594822  2 nvidia_uvm,nv_peer_mem

nv_peer_mem      4006  0      <<--- module loaded !

 

                         HOST2

ib_core                 125047  12 rdma_ucm,ib_ucm,rdma_cm,iw_cm,ib_ipoib,ib_cm,ib_uverbs,ib_umad,mlx5_ib,mlx4_ib,ib_sa,ib_mad

ib_umad               14390  0

ipv6                       334932  426 ip6t_REJECT,nf_conntrack_ipv6,nf_defrag_ipv6,ib_ipoib,ib_core,ib_addr

nvidia                    8594822  1 nvidia_uvm

 

 

What could have caused the change, and how to sinchronize again both hosts ?

Thanks and bye,

   Stefano


Viewing all articles
Browse latest Browse all 6211

Trending Articles



<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>