These are chat archives for dereneaton/ipyrad

13th
Oct 2016
Edgardo M. Ortiz
@edgardomortiz
Oct 13 2016 00:12
Hello @dereneaton @isaacovercast , this has happened in a couple of jobs:
  ipyrad [v.0.4.4]
  Interactive assembly and analysis of RAD-seq data
 -------------------------------------------------------------
  loading Assembly: eri-r1-88-12x
  from saved path: /scratch/02728/emo347/ddRAD/ericaceae/f_untrimmed_ipyrad/eri-r1-88-12x.json
  host compute node: [24 cores] on nid00313

  Step 4: Joint estimation of error rate and heterozygosity
  [####################] 100%  inferring [H, E]      | 0:39:45

  Step 5: Consensus base calling
  Mean error  [0.00406 sd=0.00284]
  Mean hetero [0.02061 sd=0.00791]
  [####################] 100%  consensus calling     | 0:09:01

  Step 6: Clustering at 0.88 similarity across 95 samples
  [####################] 100%  concat/shuffle input  | 0:00:09
  [####################] 100%  clustering across     | 0:00:00

  Encountered an unexpected error (see ./ipyrad_log.txt)
  Error message is below -------------------------------
local variable 'proc' referenced before assignment
Isaac Overcast
@isaacovercast
Oct 13 2016 18:13
@edgardomortiz I see what might be causing this (os failing to allocate pty for some reason). I added some code to step 6 to at least catch the exception. It doesn't fix the problem, but maybe will help us debug it. can you git pull and try again?
Edgardo M. Ortiz
@edgardomortiz
Oct 13 2016 18:23
Thanks @isaacovercast , it might take a while to test the new code, the TACC cluster has a huge queue today.
I have another question: Because of the particular setup of this cluster I have to start ipcluster before calling ipyrad as @dereneaton told me, however, I am getting 'Hub connection request timed out' errors when submitting simultaneous jobs, is there a way to avoid them? Usually just one of the ipyrad jobs gets to succesfully connect to ipcluster.