workflow_manager 0.7.4 → 0.7.8

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 34acafd6e9a82cd8f09832c5bcc6466c7b03b85cf816e08db78d969c5f597824
-  data.tar.gz: 4ecf1ab2713a0e328eef76f30e2766b7d822b15b177935985985670eec50a9a0
+  metadata.gz: 8c60e97cd0b4f5c15220a65c77e296b90c38bcdf0218adea5700d6e71556c494
+  data.tar.gz: 201c8eb50cf7647bf4fa53abbd64586ec05b118a82e0fb053f4102977b3c6cb8
 SHA512:
-  metadata.gz: 39166f4c016c84a718741764ee4a9461fe8892be2a7a615b92e96c81c2077c0d1d72b1effe776a138d2ed6a381065a7b598489a960c06064b1a9fbdf26c52bd6
-  data.tar.gz: c1a1708f94b3775f466f64df39d3424dbc22736b3433ba44e6e53a4e5422828c7d49d4b8494163af37f0e68f9b79e296e3ff5d7279476d8d37b109e521e802e0
+  metadata.gz: c976d0a6b684536d78657b3d21b9cd682e51cbefdbd11c00ed4f65be802f3be749d2e25a0cdf61c620a7153f112131626511af9fc434f0fe24886f0b688cbd8a
+  data.tar.gz: ca7beaebea2c29ce6da4dcb15185258aac2f22b57f773017f8f876b36c81643b212d7e09edd528051ee0695138b945fbd13e5201da4e54c4eadada605d15a07b

data/bin/wfm_monitoring

@@ -1,7 +1,7 @@
 #!/usr/bin/env ruby
 # encoding: utf-8
 # 20121112 masa workflow manager client
-Version = '20210625-165025'
+Version = '20211104-160323'
 
 require 'drb/drb' 
 require 'workflow_manager/optparse_ex'
@@ -13,10 +13,11 @@ opt = OptionParser.new do |o|
   o.on(:server, 'druby://localhost:12345', '-d server', '--server', 'workflow manager URI (default: druby://localhost:12345)')
   o.on(:log, '-o logdir', '--logdir', 'directory of standard output and standard error file outputted after the job')
   o.on(:cores, '-c cores', '--cores', 'Number of cores to request for g-sub')
-  o.on(:nodes, '-n nodes', '--nodes', 'Comma separated list of nodes to submit to for g-sub')
+#  o.on(:nodes, '-n nodes', '--nodes', 'Comma separated list of nodes to submit to for g-sub')
   o.on(:ram, '-r RAM', '--RAM', 'Amount of RAM to request in Gigs for g-sub')
   o.on(:scratch, '-s scratch', '--scratch', 'Amount of scratch space to request in Gigs for g-sub')
-  o.on(:queue, '-q queue', '--queue', 'Queue name')
+#  o.on(:queue, '-q queue', '--queue', 'Queue name')
+  o.on(:nice, '-i nice', '--nice', 'Nice')
   o.parse!(ARGV)
 end
 unless script_file = ARGV[0] and script_file =~ /\.sh/
@@ -49,7 +50,8 @@ sge_options = []
 sge_options << "-c #{opt.cores}" if opt.cores
 sge_options << "-r #{opt.ram}" if opt.ram
 sge_options << "-s #{opt.scratch}" if opt.scratch
-sge_options << "-n #{opt.nodes}" if opt.nodes
+#sge_options << "-n #{opt.nodes}" if opt.nodes
+sge_options << "-i #{opt.nice}" if opt.nice
 
 script_content = File.read(script_file)
 workflow_manager = DRbObject.new_with_uri(uri)

data/config/environments/redis.conf

@@ -1,9 +1,9 @@
 loglevel debug
 logfile ./logs/redis.log
-databases 4
+databases 5
 save 300 10
 rdbcompression yes
 dir ./dbs
 dbfilename redis.rdb
 maxmemory 10gb
-port 6380
+port 6379

data/lib/job_checker.rb

@@ -1,6 +1,9 @@
 require 'sidekiq'
 require 'redis'
 
+require 'uri'
+require 'net/http'
+
 WORKER_INTERVAL = 10 # [s]
 REDIS_CONF = File.expand_path("../../config/environments/redis.conf", __FILE__)
 PORT = if File.exist?(REDIS_CONF)
@@ -40,15 +43,15 @@ class JobChecker
     end
     new_job_script
   end
-  def update_time_status(status, script_basename, user, project_number)
+  def update_time_status(status, script_basename, user, project_number, next_dataset_id, rails_host)
     unless @start_time
       @start_time = Time.now.strftime("%Y-%m-%d %H:%M:%S")
     end
     time = Time.now.strftime("%Y-%m-%d %H:%M:%S")
-    [status, script_basename, [@start_time, time].join("/"), user, project_number].join(',')
+    [status, script_basename, [@start_time, time].join("/"), user, project_number, next_dataset_id, rails_host].join(',')
   end
 
-  def perform(job_id, script_basename, log_file, user, project_id)
+  def perform(job_id, script_basename, log_file, user, project_id, next_dataset_id=nil, rails_host=nil)
     puts "JobID (in JobChecker): #{job_id}"
     db0 = Redis.new(port: PORT, db: 0) # state + alpha DB
     db1 = Redis.new(port: PORT, db: 1) # log DB
@@ -63,10 +66,10 @@ class JobChecker
       #print ret
       state = ret.split(/\n/).last.strip
       #puts "state: #{state}"
-      db0[job_id] = update_time_status(state, script_basename, user, project_id)
+      db0[job_id] = update_time_status(state, script_basename, user, project_id, next_dataset_id, rails_host)
 
       unless state == pre_state
-        db0[job_id] = update_time_status(state, script_basename, user, project_id)
+        db0[job_id] = update_time_status(state, script_basename, user, project_id, next_dataset_id, rails_host)
         project_jobs = eval((db2[project_id]||[]).to_s)
         project_jobs = Hash[*project_jobs]
         project_jobs[job_id] = state
@@ -76,6 +79,11 @@ class JobChecker
       pre_state = state
       sleep WORKER_INTERVAL
     end while state =~ /RUNNING/  or state =~ /PENDING/ or state =~ /---/
+    if next_dataset_id and rails_host
+      uri = URI("#{rails_host}/data_set/#{next_dataset_id}/update_completed_samples")
+      #p uri
+      res = Net::HTTP.get_response(uri)
+    end
   end
 end
 

data/lib/workflow_manager/cluster.rb

@@ -450,11 +450,15 @@ module WorkflowManager
       partition = if i = options.index("-p")
                     options[i+1]
                   end
+      nice = if i = options.index("-i")
+               options[i+1]
+             end
       new_options = []
       new_options << "--mem=#{ram}G" if ram
       new_options << "-n #{cores}" if cores
       new_options << "--tmp=#{scratch}G" if scratch
       new_options << "-p #{partition}" if partition
+      new_options << "--nice=#{nice}" if nice
       new_options.join(" ")
     end
     def submit_job(script_file, script_content, option='')

data/lib/workflow_manager/server.rb

@@ -163,6 +163,7 @@ module WorkflowManager
                     RedisDB.new(1, @redis_conf)
                 end
       @jobs = RedisDB.new(2, @redis_conf)
+      @trees = RedisDB.new(4, @redis_conf)
 
       @system_log = File.join(@log_dir, "system.log")
       @mutex = Mutex.new
@@ -186,12 +187,12 @@ module WorkflowManager
         statuses.each do |job_id, status|
           # puts [job_id, status].join(",")
           # 120249,RUNNING,QC_ventricles_100k.sh,2021-07-30 09:47:04/2021-07-30 09:47:04,masaomi,1535
-          stat, script_basename, time, user, project_number = status.split(",")
+          stat, script_basename, time, user, project_number, next_dataset_id, rails_host = status.split(",")
           if stat == "RUNNING" or stat == "PENDING"
             log_file = logs[job_id]
             log_puts("JobID (in recovery check): #{job_id}")
             puts "JobID (in recovery check): #{job_id}"
-            JobChecker.perform_async(job_id, script_basename, log_file, user, project_number)
+            JobChecker.perform_async(job_id, script_basename, log_file, user, project_number, next_dataset_id, rails_host)
           end
         end
       end
@@ -296,7 +297,7 @@ module WorkflowManager
         Thread.current.kill
       end
     end
-    def start_monitoring3(script_path, script_content, user='sushi_lover', project_number=0, sge_options='', log_dir='')
+    def start_monitoring3(script_path, script_content, user='sushi_lover', project_number=0, sge_options='', log_dir='', next_dataset_id='', rails_host=nil)
       script_basename = File.basename(script_path)
       job_id, log_file, command = @cluster.submit_job(script_path, script_content, sge_options)
       #p command
@@ -304,7 +305,7 @@ module WorkflowManager
       #p job_id
       puts "JobID (in WorkflowManager): #{job_id}"
       sleep 1
-      JobChecker.perform_async(job_id, script_basename, log_file, user, project_number)
+      JobChecker.perform_async(job_id, script_basename, log_file, user, project_number, next_dataset_id, rails_host)
       job_id
     end
     def start_monitoring2(script_path, script_content, user='sushi_lover', project_number=0, sge_options='', log_dir='')
@@ -473,17 +474,23 @@ module WorkflowManager
       job_idsh = if job_ids
                    Hash[*(job_ids.split(',')).map{|job_id| [job_id, true]}.flatten]
                  end
-      s_ = {}
-      unless job_ids
+      if project_number
+        s_ = {}
         @jobs.transaction do |jobs|
           if project_jobs = jobs[project_number]
             s_ = Hash[*eval(project_jobs)]
           end
         end
-      end
-      @statuses.transaction do |statuses|
-        s_.each do |job_id, stat|
-          s << [job_id, statuses[job_id]]
+        @statuses.transaction do |statuses|
+          s_.each do |job_id, stat|
+            s << [job_id, statuses[job_id]]
+          end
+        end
+      else
+        @statuses.transaction do |statuses|
+          statuses.each do |key, value|
+            s << [key, value]
+          end
         end
       end
       if job_ids
@@ -549,6 +556,19 @@ module WorkflowManager
     def cluster_node_list
       @cluster.node_list
     end
+    def save_dataset_tree(project_number, json)
+      @trees.transaction do |trees|
+        trees[project_number] = json
+      end
+      json
+    end
+    def load_dataset_tree(project_number)
+      json = nil
+      @trees.transaction do |trees|
+        json = trees[project_number]
+      end
+      json
+    end
   end
 end
 

data/lib/workflow_manager/version.rb

@@ -1,3 +1,3 @@
 module WorkflowManager
-  VERSION = "0.7.4"
+  VERSION = "0.7.8"
 end

data/start_workflow_manager.sh

@@ -0,0 +1,11 @@
+#!/usr/bin/bash
+source /usr/local/ngseq/etc/lmod_profile
+module load Dev/Ruby/2.6.7
+module load Tools/Redis/6.0.1
+conda activate gtools_env
+which python
+which g-sub
+which g-req
+mkdir -p logs
+mkdir -p dbs
+bundle exec workflow_manager -d druby://fgcz-h-032:40002

data/test/job_list.rb

@@ -1,13 +1,13 @@
 #!/usr/bin/env ruby
 # encoding: utf-8
-# Version = '20210723-134812'
+# Version = '20211001-104513'
 
 PORT = (ARGV[0]||6380).to_i
 require 'redis'
 db0 = Redis.new(port: PORT, db: 0)
 db1 = Redis.new(port: PORT, db: 1)
 db2 = Redis.new(port: PORT, db: 2)
-#db3 = Redis.new(port: 6380, db: 3)
+db4 = Redis.new(port: PORT, db: 4)
 
 class Redis
   def show_all
@@ -18,8 +18,8 @@ class Redis
   end
 end
 
-dbs = [db0, db1, db2]
-db_notes = ["state DB", "log DB", "project job DB"]
+dbs = [db0, db1, db2, db4]
+db_notes = ["state DB", "log DB", "project job DB", "JS tree DB"]
 
 dbs.each.with_index do |db, i|
   note = db_notes[i]
@@ -48,3 +48,10 @@ db2.keys.sort.each do |key|
   value = db2.get(key)
   puts [key, value].join("\t")
 end
+
+puts
+puts "db3, status DB3, project specific"
+db3.keys.sort.each do |key|
+  value = db3.get(key)
+  puts [key, value].join("\t")
+end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: workflow_manager
 version: !ruby/object:Gem::Version
-  version: 0.7.4
+  version: 0.7.8
 platform: ruby
 authors:
 - Functional Genomics Center Zurich
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2021-07-31 00:00:00.000000000 Z
+date: 2021-11-04 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -87,6 +87,7 @@ files:
 - spec/cluster_spec.rb
 - spec/server_spec.rb
 - spec/spec_helper.rb
+- start_workflow_manager.sh
 - test/call_worker4.rb
 - test/call_worker_method.rb
 - test/job_list.rb