locked
Compute node stuck in provisioning state RRS feed

  • Question

  • Dear all,
    I have configured <g class="gr_ gr_37 gr-alert gr_gramm gr_inline_cards gr_run_anim Grammar multiReplace" data-gr-id="37" id="37">a HPC</g> head node <g class="gr_ gr_34 gr-alert gr_gramm gr_inline_cards gr_run_anim Grammar multiReplace" data-gr-id="34" id="34">in</g>  a single server and I am are trying to add compute node to the cluster, the compute node is visible in the cluster and when <g class="gr_ gr_40 gr-alert gr_tiny gr_spell gr_inline_cards gr_run_anim ContextualSpelling multiReplace" data-gr-id="40" id="40">i</g> am adding the compute node to the cluster it is stuck in provisioning state and when checked the provisioning log I  am getting the following error

    Time Message
    10/5/2017 11:39:12 PM Two reasons you got this exception:
    1, There is no cached context, you should call GetOrAdd first.
    2, The cancellation token associated with the HpcContext is cancelled, so no further access to the context and please cleanup as soon as possible.

    Time Message
    10/5/2017 11:39:12 PM Nested operation with ID dd0e9388-13e8-4ff5-998d-8a211cc9e4d9 failed

    Tried various possibilities like disabling the secure boot


    Friday, October 6, 2017 6:40 AM

All replies

  • Hi, 

      We will take a check.


    Qiufang Shi

    Sunday, October 8, 2017 3:18 AM
  • Could you check whether your compute node is with Windows deployment service enabled? There is a known issue in HPC Pack 2016 that the compute node must not have WDS enabled. Please uninstall the WDS service and try again.
    Monday, October 9, 2017 1:44 AM