def start_job_monitor(self):
try:
while True:
newjobids = jobmodel_helper.check_new_jobs()
if (newjobids[0] != "nojobs"):
print "new jobs\n \n \n"
for jobid in newjobids:
self.start_job(jobid)
time.sleep(10)
except Exception, e:
print 'Unable to process in worker thread: ' + str(e)
def start_job_monitor(self):
try:
while True:
newjobids = jobmodel_helper.check_new_jobs()
if (newjobids[0] != "nojobs"):
print("new jobs\n \n \n")
for jobid in newjobids:
self.start_job(jobid)
time.sleep(10)
except Exception as e:
print('Unable to process in worker thread: ' + str(e))
def start_job(self, jobid):
print "running job with id %s"%jobid
jobid = str(jobid)
working_directory = os.path.join(DARE_HOME, "darefiles/", str(jobid))
conffile = os.path.join(working_directory , str(jobid) + "-job.cfg")
os.system("mkdir -p "+ working_directory)
self.prepare_conf(jobid, conffile)
jobmodel_helper.update_job_status(jobid, 2)
# start the dare here:
job_type = jobmodel_helper.get_jobtype(jobid)
if(job_type == "tophatfusion"):
exec_file = "tophatfusion_dare.py"
elif(job_type == "chipseq"):
mapping_tool = jobmodel_helper.get_mapping_tool(jobid)
print "mapping toool \n\n\n", mapping_tool
if (mapping_tool == "BOWTIE"):
exec_file = "scalable_chipseq_2_dare.py"
else:
exec_file = "scalable_chipseq_dare.py"
else:
exec_file = "bfast_dare.py"
p1 = Popen(["python", os.path.join(DARE_HOME,"examples",exec_file), "-c", conffile], stdout=PIPE)
print "p1.pid",p1.pid
jobmodel_helper.update_job_pid(jobid, p1.pid)
#sts = os.waitpid(p1.pid, 0)[1]
#update_job_status(jobid, 4)
print '\n completed starting job%s' %jobid
def start_job(self, jobid):
print("running job with id %s"%jobid)
jobid = str(jobid)
working_directory = os.path.join(DARE_HOME, "darefiles/", str(jobid))
conffile = os.path.join(working_directory , str(jobid) + "-job.cfg")
os.system("mkdir -p "+ working_directory)
self.prepare_conf(jobid, conffile)
jobmodel_helper.update_job_status(jobid, 2)
# start the dare here:
job_type = jobmodel_helper.get_jobtype(jobid)
if(job_type == "tophatfusion"):
exec_file = "tophatfusion_dare.py"
elif(job_type == "chipseq"):
mapping_tool = jobmodel_helper.get_mapping_tool(jobid)
print("mapping toool \n\n\n", mapping_tool)
if (mapping_tool == "BOWTIE"):
exec_file = "scalable_chipseq_2_dare.py"
else:
exec_file = "scalable_chipseq_dare.py"
else:
exec_file = "bfast_dare.py"
p1 = Popen(["python", os.path.join(DARE_HOME,"examples",exec_file), "-c", conffile], stdout=PIPE)
print("p1.pid",p1.pid)
jobmodel_helper.update_job_pid(jobid, p1.pid)
#sts = os.waitpid(p1.pid, 0)[1]
#update_job_status(jobid, 4)
print('\n completed starting job%s' %jobid)
def prepare_conf(self, jobid, conffile):
config = ConfigParser.ConfigParser()
cfgfile = open(conffile,'w')
newjobinfo = jobmodel_helper.get_jobinfo(jobid)
section_name = 'JOB'
config.add_section(section_name)
config.set(section_name ,"jobid" , jobid)
config.set(section_name ,"dare_home", DARE_HOME)
for jobinfo in newjobinfo:
config.set(section_name ,jobinfo.key , jobinfo.value)
config.write(cfgfile)
cfgfile.close()
def prepare_conf(self, jobid, conffile):
config = configparser.ConfigParser()
cfgfile = open(conffile,'w')
newjobinfo = jobmodel_helper.get_jobinfo(jobid)
section_name = 'JOB'
config.add_section(section_name)
config.set(section_name ,"jobid" , jobid)
config.set(section_name ,"dare_home", DARE_HOME)
for jobinfo in newjobinfo:
config.set(section_name ,jobinfo.key , jobinfo.value)
config.write(cfgfile)
cfgfile.close()
def run(self):
# create a logfile
LOG_FILENAME = self.job_conf["log_filename"]
print LOG_FILENAME
self.logger = logging.getLogger("dare_multijob")
hdlr = logging.FileHandler(LOG_FILENAME)
formatter = logging.Formatter("%(asctime)s %(levelname)s %(message)s")
hdlr.setFormatter(formatter)
self.logger.addHandler(hdlr)
self.logger.setLevel(logging.INFO)
# first info in log file
self.logger.info("Job id is " + self.job_conf["jobid"])
self.logger.info("RESOURCES used are " + self.job_conf["num_resources"])
try:
# get resource info
# start the big job agents
resource_list = []
self.mjs = []
for i in range(0, int(self.job_conf["num_resources"])):
resource_list.append([])
resource_list[i].append(dict_section(self.config, "resource_" + str(i)))
# create multiple manyjobs
print "Create manyjob service "
self.mjs.append(many_job.many_job_service(resource_list[i], None))
total_number_of_jobs = 0
### run the step
wus_count = 0
for STEP in range(0, int(self.job_conf["num_steps"])):
starttime = time.time()
# job started update status
if self.load_update_env == "true":
jobmodel_helper.update_job_detail_status(self.job_conf["jobid"], "In step " + str(STEP + 1))
step_wus = self.job_conf["step_" + str(STEP)].split(",")
if ("step_" + str(STEP)) not in self.job_conf["ft_steps"].split(","):
### submit the each step wus to bigjob
for wu_count in range(0, len(step_wus)):
wu = dict_section(self.config, step_wus[wu_count].strip())
wus_count = wus_count + 1
self.submit_wu(wu)
self.wait_for_wus(wus_count)
else:
# time.sleep(10)
for wu_count in range(0, len(step_wus)):
fs = dict_section(self.config, step_wus[wu_count].strip())
self.submit_fs(fs)
runtime = time.time() - starttime
self.logger.info("STEP" + str(STEP) + " Runtime: " + str(runtime))
# all jobs done update status
if self.load_update_env == "true":
jobmodel_helper.update_job_detail_status(self.job_conf["jobid"], "")
jobmodel_helper.update_job_status(self.job_conf["jobid"], 4)
for i in range(0, int(self.job_conf["num_resources"])):
self.mjs[i].cancel()
except:
traceback.print_exc(file=sys.stdout)
try:
for i in range(0, int(self.job_conf["num_resources"])):
self.mjs[i].cancel()
except:
sys.exit()
def run(self):
#create a logfile
LOG_FILENAME = self.job_conf["log_filename"]
print(LOG_FILENAME)
self.logger = logging.getLogger('dare_multijob')
hdlr = logging.FileHandler(LOG_FILENAME)
formatter = logging.Formatter('%(asctime)s %(levelname)s %(message)s')
hdlr.setFormatter(formatter)
self.logger.addHandler(hdlr)
self.logger.setLevel(logging.INFO)
#first info in log file
self.logger.info("Job id is " + self.job_conf["jobid"])
self.logger.info("RESOURCES used are " +
self.job_conf["num_resources"])
try:
#get resource info
#start the big job agents
resource_list = []
self.mjs = []
for i in range(0, int(self.job_conf["num_resources"])):
resource_list.append([])
resource_list[i].append(
dict_section(self.config, "resource_" + str(i)))
#create multiple manyjobs
print("Create manyjob service ")
self.mjs.append(
many_job.many_job_service(resource_list[i], None))
total_number_of_jobs = 0
### run the step
wus_count = 0
for STEP in range(0, int(self.job_conf["num_steps"])):
starttime = time.time()
#job started update status
if (self.load_update_env == "true"):
jobmodel_helper.update_job_detail_status(
self.job_conf["jobid"], "In step " + str(STEP + 1))
step_wus = self.job_conf["step_" + str(STEP)].split(',')
if ("step_" +
str(STEP)) not in self.job_conf["ft_steps"].split(','):
### submit the each step wus to bigjob
for wu_count in range(0, len(step_wus)):
wu = dict_section(self.config,
step_wus[wu_count].strip())
wus_count = wus_count + 1
self.submit_wu(wu)
self.wait_for_wus(wus_count)
else:
#time.sleep(10)
for wu_count in range(0, len(step_wus)):
fs = dict_section(self.config,
step_wus[wu_count].strip())
self.submit_fs(fs)
runtime = time.time() - starttime
self.logger.info("STEP" + str(STEP) + " Runtime: " +
str(runtime))
#all jobs done update status
if (self.load_update_env == "true"):
jobmodel_helper.update_job_detail_status(
self.job_conf["jobid"], "")
jobmodel_helper.update_job_status(self.job_conf["jobid"], 4)
for i in range(0, int(self.job_conf["num_resources"])):
self.mjs[i].cancel()
except:
traceback.print_exc(file=sys.stdout)
try:
for i in range(0, int(self.job_conf["num_resources"])):
self.mjs[i].cancel()
except:
sys.exit()
