def run_htseq_count_process(cluster_name, log, function=None):
'''
Run a htseq-count process.
'''
# initialize the control variable
OK = True
# get the htseq-count option dictionary
htseq_count_option_dict = xlib.get_option_dict(
get_htseq_count_config_file())
# get the experiment identification
experiment_id = htseq_count_option_dict['identification']['experiment_id']
# warn that the log window does not have to be closed
if not isinstance(log, xlib.DevStdOut):
log.write(
'This process might take several minutes. Do not close this window, please wait!\n'
)
# check the htseq-count config file
log.write(f'{xlib.get_separator()}\n')
log.write(f'Checking the {xlib.get_htseq_count_name()} config file ...\n')
(OK, error_list) = check_htseq_count_config_file(strict=True)
if OK:
log.write('The file is OK.\n')
else:
log.write('*** ERROR: The config file is not valid.\n')
log.write('Please correct this file or recreate the config files.\n')
# create the SSH client connection
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write('Connecting the SSH client ...\n')
(OK, error_list,
ssh_client) = xssh.create_ssh_client_connection(cluster_name)
if OK:
log.write('The SSH client is connected.\n')
else:
for error in error_list:
log.write(f'{error}\n')
# create the SSH transport connection
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write('Connecting the SSH transport ...\n')
(OK, error_list,
ssh_transport) = xssh.create_ssh_transport_connection(cluster_name)
if OK:
log.write('The SSH transport is connected.\n')
else:
for error in error_list:
log.write(f'{error}\n')
# create the SFTP client
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write('Connecting the SFTP client ...\n')
sftp_client = xssh.create_sftp_client(ssh_transport)
log.write('The SFTP client is connected.\n')
# warn that the requirements are being verified
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write('Checking process requirements ...\n')
# check the master is running
if OK:
(master_state_code,
master_state_name) = xec2.get_node_state(cluster_name)
if master_state_code != 16:
log.write(
f'*** ERROR: The cluster {cluster_name} is not running. Its state is {master_state_code} ({master_state_name}).\n'
)
OK = False
# check HTSeq is installed
if OK:
(OK, error_list,
is_installed) = xbioinfoapp.is_installed_anaconda_package(
xlib.get_htseq_anaconda_code(), cluster_name, True, ssh_client)
if OK:
if not is_installed:
log.write(
f'*** ERROR: {xlib.get_htseq_name()} is not installed.\n')
OK = False
else:
log.write(
f'*** ERROR: The verification of {xlib.get_htseq_name()} installation could not be performed.\n'
)
# warn that the requirements are OK
if OK:
log.write('Process requirements are OK.\n')
# determine the run directory in the cluster
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write('Determining the run directory in the cluster ...\n')
current_run_dir = xlib.get_cluster_current_run_dir(
experiment_id, xlib.get_htseq_count_code())
command = f'mkdir --parents {current_run_dir}'
(OK, _, _) = xssh.execute_cluster_command(ssh_client, command)
if OK:
log.write(f'The directory path is {current_run_dir}.\n')
else:
log.write(f'*** ERROR: Wrong command ---> {command}\n')
# build the htseq-count process script
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write(
f'Building the process script {get_htseq_count_process_script()} ...\n'
)
(OK, error_list) = build_htseq_count_process_script(
cluster_name, current_run_dir)
if OK:
log.write('The file is built.\n')
if not OK:
log.write('*** ERROR: The file could not be built.\n')
# upload the htseq-count process script to the cluster
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write(
f'Uploading the process script {get_htseq_count_process_script()} to the directory {current_run_dir} ...\n'
)
cluster_path = f'{current_run_dir}/{os.path.basename(get_htseq_count_process_script())}'
(OK, error_list) = xssh.put_file(sftp_client,
get_htseq_count_process_script(),
cluster_path)
if OK:
log.write('The file is uploaded.\n')
else:
for error in error_list:
log.write(f'{error}\n')
# set run permision to the htseq-count process script in the cluster
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write(
f'Setting on the run permision of {current_run_dir}/{os.path.basename(get_htseq_count_process_script())} ...\n'
)
command = f'chmod u+x {current_run_dir}/{os.path.basename(get_htseq_count_process_script())}'
(OK, _, _) = xssh.execute_cluster_command(ssh_client, command)
if OK:
log.write('The run permision is set.\n')
else:
log.write(f'*** ERROR: Wrong command ---> {command}\n')
# build the htseq-count process starter
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write(
f'Building the process starter {get_htseq_count_process_starter()} ...\n'
)
(OK, error_list) = build_htseq_count_process_starter(current_run_dir)
if OK:
log.write('The file is built.\n')
if not OK:
log.write('***ERROR: The file could not be built.\n')
# upload the htseq-count process starter to the cluster
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write(
f'Uploading the process starter {get_htseq_count_process_starter()} to the directory {current_run_dir} ...\n'
)
cluster_path = f'{current_run_dir}/{os.path.basename(get_htseq_count_process_starter())}'
(OK, error_list) = xssh.put_file(sftp_client,
get_htseq_count_process_starter(),
cluster_path)
if OK:
log.write('The file is uploaded.\n')
else:
for error in error_list:
log.write(f'{error}\n')
# set run permision to the htseq-count process starter in the cluster
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write(
f'Setting on the run permision of {current_run_dir}/{os.path.basename(get_htseq_count_process_starter())} ...\n'
)
command = f'chmod u+x {current_run_dir}/{os.path.basename(get_htseq_count_process_starter())}'
(OK, _, _) = xssh.execute_cluster_command(ssh_client, command)
if OK:
log.write('The run permision is set.\n')
else:
log.write(f'*** ERROR: Wrong command ---> {command}\n')
# submit the htseq-count process
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write(
f'Submitting the process script {current_run_dir}/{os.path.basename(get_htseq_count_process_starter())} ...\n'
)
OK = xssh.submit_script(
cluster_name, ssh_client, current_run_dir,
os.path.basename(get_htseq_count_process_starter()), log)
# close the SSH transport connection
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write('Closing the SSH transport connection ...\n')
xssh.close_ssh_transport_connection(ssh_transport)
log.write('The connection is closed.\n')
# close the SSH client connection
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write('Closing the SSH client connection ...\n')
xssh.close_ssh_client_connection(ssh_client)
log.write('The connection is closed.\n')
# warn that the log window can be closed
if not isinstance(log, xlib.DevStdOut):
log.write(f'{xlib.get_separator()}\n')
log.write('You can close this window now.\n')
# execute final function
if function is not None:
function()
# return the control variable
return OK
def run_gzip_process(cluster_name, dataset_type, log, function=None):
'''
Run a gzip process.
'''
# initialize the control variable
OK = True
# get the gzip code and name
gzip_code = xlib.get_gzip_code()
gzip_name = xlib.get_gzip_name()
# get the gzip option dictionary
gzip_option_dict = xlib.get_option_dict(get_gzip_config_file(dataset_type))
# get the experiment identification
experiment_id = gzip_option_dict['identification']['experiment_id']
# get the gzip process script path in the local computer
gzip_process_script = get_gzip_process_script(dataset_type)
# get the gzip process starter path in the local computer
gzip_process_starter = get_gzip_process_starter(dataset_type)
# warn that the log window does not have to be closed
if not isinstance(log, xlib.DevStdOut):
log.write('This process might take several minutes. Do not close this window, please wait!\n')
# check the gzip config file
log.write(f'{xlib.get_separator()}\n')
log.write('Checking the {0} config file ...\n'.format(gzip_name))
(OK, error_list) = check_gzip_config_file(dataset_type, strict=True)
if OK:
log.write('The file is OK.\n')
else:
log.write('*** ERROR: The config file is not valid.\n')
log.write('Please correct this file or recreate the config files.\n')
# create the SSH client connection
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write('Connecting the SSH client ...\n')
(OK, error_list, ssh_client) = xssh.create_ssh_client_connection(cluster_name)
if OK:
log.write('The SSH client is connected.\n')
else:
for error in error_list:
log.write(f'{error}\n')
# create the SSH transport connection
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write('Connecting the SSH transport ...\n')
(OK, error_list, ssh_transport) = xssh.create_ssh_transport_connection(cluster_name)
if OK:
log.write('The SSH transport is connected.\n')
else:
for error in error_list:
log.write(f'{error}\n')
# create the SFTP client
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write('Connecting the SFTP client ...\n')
sftp_client = xssh.create_sftp_client(ssh_transport)
log.write('The SFTP client is connected.\n')
# warn that the requirements are being verified
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write('Checking process requirements ...\n')
# check the master is running
if OK:
(master_state_code, master_state_name) = xec2.get_node_state(cluster_name)
if master_state_code != 16:
log.write(f'*** ERROR: The cluster {cluster_name} is not running. Its state is {master_state_code} ({master_state_name}).\n')
OK = False
# warn that the requirements are OK
if OK:
log.write('Process requirements are OK.\n')
# determine the run directory in the cluster
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write('Determining the run directory in the cluster ...\n')
if dataset_type == 'reference':
current_run_dir = xlib.get_cluster_current_run_dir('reference', gzip_code)
elif dataset_type == 'database':
current_run_dir = xlib.get_cluster_current_run_dir('database', gzip_code)
else:
current_run_dir = xlib.get_cluster_current_run_dir(experiment_id, gzip_code)
command = f'mkdir --parents {current_run_dir}'
(OK, stdout, stderr) = xssh.execute_cluster_command(ssh_client, command)
if OK:
log.write(f'The directory path is {current_run_dir}.\n')
else:
log.write(f'*** ERROR: Wrong command ---> {command}\n')
# build the gzip process script
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write('Building the process script {0} ...\n'.format(gzip_process_script))
(OK, error_list) = build_gzip_process_script(cluster_name, dataset_type, current_run_dir)
if OK:
log.write('The file is built.\n')
else:
log.write('*** ERROR: The file could not be built.\n')
# upload the gzip process script to the cluster
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write('Uploading the process script {0} to the directory {1} ...\n'.format(gzip_process_script, current_run_dir))
cluster_path = '{0}/{1}'.format(current_run_dir, os.path.basename(gzip_process_script))
(OK, error_list) = xssh.put_file(sftp_client, gzip_process_script, cluster_path)
if OK:
log.write('The file is uploaded.\n')
else:
for error in error_list:
log.write(f'{error}\n')
# set run permision to the gzip process script in the cluster
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write('Setting on the run permision of {0}/{1} ...\n'.format(current_run_dir, os.path.basename(gzip_process_script)))
command = 'chmod u+x {0}/{1}'.format(current_run_dir, os.path.basename(gzip_process_script))
(OK, stdout, stderr) = xssh.execute_cluster_command(ssh_client, command)
if OK:
log.write('The run permision is set.\n')
else:
log.write(f'*** ERROR: Wrong command ---> {command}\n')
# build the gzip process starter
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write('Building the process starter {0} ...\n'.format(gzip_process_starter))
(OK, error_list) = build_gzip_process_starter(dataset_type, current_run_dir)
if OK:
log.write('The file is built.\n')
else:
log.write('***ERROR: The file could not be built.\n')
# upload the gzip process starter to the cluster
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write('Uploading the process starter {0} to the directory {1} ...\n'.format(gzip_process_starter, current_run_dir))
cluster_path = '{0}/{1}'.format(current_run_dir, os.path.basename(gzip_process_starter))
(OK, error_list) = xssh.put_file(sftp_client, gzip_process_starter, cluster_path)
if OK:
log.write('The file is uploaded.\n')
else:
for error in error_list:
log.write(f'{error}\n')
# set run permision to the gzip process starter in the cluster
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write('Setting on the run permision of {0}/{1} ...\n'.format(current_run_dir, os.path.basename(gzip_process_starter)))
command = 'chmod u+x {0}/{1}'.format(current_run_dir, os.path.basename(gzip_process_starter))
(OK, stdout, stderr) = xssh.execute_cluster_command(ssh_client, command)
if OK:
log.write('The run permision is set.\n')
else:
log.write(f'*** ERROR: Wrong command ---> {command}\n')
# submit the gzip process
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write('Submitting the process script {0}/{1} ...\n'.format(current_run_dir, os.path.basename(gzip_process_starter)))
OK = xssh.submit_script(cluster_name, ssh_client, current_run_dir, os.path.basename(gzip_process_starter), log)
# close the SSH transport connection
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write('Closing the SSH transport connection ...\n')
xssh.close_ssh_transport_connection(ssh_transport)
log.write('The connection is closed.\n')
# close the SSH client connection
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write('Closing the SSH client connection ...\n')
xssh.close_ssh_client_connection(ssh_client)
log.write('The connection is closed.\n')
# warn that the log window can be closed
if not isinstance(log, xlib.DevStdOut):
log.write(f'{xlib.get_separator()}\n')
log.write('You can close this window now.\n')
# execute final function
if function is not None:
function()
# return the control variable
return OK
def upload_read_dataset(cluster_name, log, function=None):
'''
Upload the read dataset to the cluster.
'''
# initialize the control variable
OK = True
# get the read transfer config file
read_transfer_config_file = get_read_transfer_config_file()
# warn that the log window must not be closed
if not isinstance(log, xlib.DevStdOut):
log.write(
'This process might take several minutes. Do not close this window, please wait!\n'
)
# get and validate the read transfer config file
log.write('{0}\n'.format(xlib.get_separator()))
log.write('The read transfer config file is been validating ...\n')
if validate_read_transfer_config_file(strict=True):
log.write('The config file is OK.\n')
else:
log.write('*** ERROR: The read transfer config file is not valid.\n')
log.write('Please correct this file or recreate the config files.\n')
OK = False
# create the SSH client connection
if OK:
(OK, error_list, ssh_client) = xssh.create_ssh_client_connection(
cluster_name, 'master')
for error in error_list:
log.write('{0}\n'.format(error))
# create the SSH transport connection
if OK:
(OK, error_list, ssh_transport) = xssh.create_ssh_transport_connection(
cluster_name, 'master')
for error in error_list:
log.write('{0}\n'.format(error))
# create the SFTP client
if OK:
sftp_client = xssh.create_sftp_client(ssh_transport)
# get the options dictionary
if OK:
read_transfer_options_dict = xlib.get_option_dict(
read_transfer_config_file)
# get the experiment identification and create the experiment reads directory
if OK:
# get the experiment identification
experiment_id = read_transfer_options_dict['identification'][
'experiment_id']
# Get the directory of read and results datasets of the experiment
cluster_experiment_reads_dir = xlib.get_cluster_experiment_read_dataset_dir(
experiment_id, xlib.get_uploaded_read_dataset_name())
cluster_experiment_result_dir = xlib.get_cluster_experiment_result_dir(
experiment_id)
# create the experiment reads directory
log.write('{0}\n'.format(xlib.get_separator()))
log.write(
'The reads directory {0} in the cluster is being created ...\n'.
format(cluster_experiment_reads_dir))
command = 'mkdir --parents {0}'.format(cluster_experiment_reads_dir)
(OK, stdout,
stderr) = xssh.execute_cluster_command(ssh_client, command)
if OK:
log.write('The directory is created.\n')
else:
log.write('*** ERROR: Wrong command ---> {0}\n'.format(command))
# create the experiment run result directory
log.write('{0}\n'.format(xlib.get_separator()))
log.write(
'The run result directory {0} in the cluster is being created ...\n'
.format(cluster_experiment_result_dir))
command = 'mkdir --parents {0}'.format(cluster_experiment_result_dir)
(OK, stdout,
stderr) = xssh.execute_cluster_command(ssh_client, command)
if OK:
log.write('The directory is created.\n')
else:
log.write('*** ERROR: Wrong command ---> {0}\n'.format(command))
# upload the read dataset
if OK:
# get the sections list
sections_list = []
for section in read_transfer_options_dict.keys():
sections_list.append(section)
sections_list.sort()
# for each section "file-n"
for section in sections_list:
# verify than the section identification is like file-n
if re.match('^file-[0-9]+$', section):
# get local path and cluster directory
local_path = read_transfer_options_dict[section]['local_path']
# upload the reference file in the cluster
log.write('{0}\n'.format(xlib.get_separator()))
log.write('The file {0} is being uploaded to {1} ...\n'.format(
local_path, cluster_experiment_reads_dir))
cluster_path = '{0}/{1}'.format(cluster_experiment_reads_dir,
os.path.basename(local_path))
(OK, error_list) = xssh.put_file(sftp_client, local_path,
cluster_path)
if OK:
log.write('The file has been uploaded.\n')
else:
for error in error_list:
log.write('{0}\n'.format(error))
break
# close the SSH transport connection
if OK:
xssh.close_ssh_transport_connection(ssh_transport)
# close the SSH client connection
if OK:
xssh.close_ssh_client_connection(ssh_client)
# warn that the log window can be closed
if not isinstance(log, xlib.DevStdOut):
log.write('{0}\n'.format(xlib.get_separator()))
log.write('You can close this window now.\n')
# execute final function
if function is not None:
function()
# return the control variable
return OK
def install_node_infrastructure_software(cluster_name, node_name, log):
'''
Install infraestructure software in a node.
'''
# initialize the control variable
OK = True
# get the infrastructure software installation script path in local compute
local_script_path = get_infrastructure_software_installation_script()
# set the infrastructure software installation script path in node
node_script_path = './{0}'.format(os.path.basename(local_script_path))
# set the infrastructure software installation log path in node
node_log_path = node_script_path[:node_script_path.find('.sh')] + '.log'
# build the infrastructure software installation script
if OK:
log.write('{0}\n'.format(xlib.get_separator()))
log.write(
'Building the infrastructure software installation script {0} ...\n'
.format(local_script_path))
(OK, error_list
) = build_infrastructure_software_installation_script(cluster_name)
if OK:
log.write('The file is built.\n')
else:
for error in error_list:
log.write('{0}\n'.format(error))
# create the SSH client connection
if OK:
log.write('{0}\n'.format(xlib.get_separator()))
log.write(
'Connecting the SSH client to node {0} ...\n'.format(node_name))
(OK, error_list, ssh_client) = xssh.create_ssh_client_connection(
cluster_name, node_name)
if OK:
log.write('The SSH client is connected.\n')
else:
for error in error_list:
log.write('{0}\n'.format(error))
# create the SSH transport connection
if OK:
log.write('{0}\n'.format(xlib.get_separator()))
log.write(
'Connecting the SSH transport to node {0} ...\n'.format(node_name))
(OK, error_list, ssh_transport) = xssh.create_ssh_transport_connection(
cluster_name, node_name)
if OK:
log.write('The SSH transport is connected.\n')
else:
for error in error_list:
log.write('{0}\n'.format(error))
# create the SFTP client
if OK:
log.write('{0}\n'.format(xlib.get_separator()))
log.write(
'Connecting the SFTP client to node {0} ...\n'.format(node_name))
sftp_client = xssh.create_sftp_client(ssh_transport)
log.write('The SFTP client is connected.\n')
# upload the infraestructe software installation script to the node
if OK:
log.write('{0}\n'.format(xlib.get_separator()))
log.write(
'Uploading the infraestructe software installation script to the node {0} ...\n'
.format(node_name))
(OK, error_list) = xssh.put_file(sftp_client, local_script_path,
node_script_path)
if OK:
log.write('The file is uploaded.\n')
else:
for error in error_list:
log.write('{0}\n'.format(error))
# set run permision to the infraestructe software installation script in the cluster
if OK:
log.write('{0}\n'.format(xlib.get_separator()))
log.write('Setting on the run permision ...\n')
command = 'chmod u+x {0}'.format(node_script_path)
(OK, stdout,
stderr) = xssh.execute_cluster_command(ssh_client, command)
if OK:
log.write('The run permision is set.\n')
else:
log.write('*** ERROR: Wrong command ---> {0}\n'.format(command))
# submit the infraestructe software installation script
if OK:
log.write('{0}\n'.format(xlib.get_separator()))
log.write(
'Submitting the infraestructe software installation script in the node {0} ...\n'
.format(node_name))
command = '{0} &>{1} &'.format(node_script_path, node_log_path)
(OK, stdout,
stderr) = xssh.execute_cluster_command(ssh_client, command)
if OK:
log.write('The script is submitted.\n')
else:
log.write('*** ERROR: Wrong command ---> {0}\n'.format(command))
# close the SSH transport connection
if OK:
log.write('{0}\n'.format(xlib.get_separator()))
log.write('Closing the SSH transport connection ...\n')
xssh.close_ssh_transport_connection(ssh_transport)
log.write('The connection is closed.\n')
# close the SSH client connection
if OK:
log.write('{0}\n'.format(xlib.get_separator()))
log.write('Closing the SSH client connection ...\n')
xssh.close_ssh_client_connection(ssh_client)
log.write('The connection is closed.\n')
# return the control variable
return OK
def upload_database_dataset(cluster_name, log, function=None):
'''
Upload the database dataset to the cluster.
'''
# initialize the control variable
OK = True
# warn that the log window does not have to be closed
if not isinstance(log, xlib.DevStdOut):
log.write(
'This process might take several minutes. Do not close this window, please wait!\n'
)
# check the database transfer config file
log.write(f'{xlib.get_separator()}\n')
log.write('Checking the database transfer config file ...\n')
if check_database_transfer_config_file(strict=True):
log.write('The file is OK.\n')
else:
log.write(
'*** ERROR: The database transfer config file is not valid.\n')
log.write('Please correct this file or recreate the config files.\n')
OK = False
# create the SSH client connection
if OK:
(OK, error_list,
ssh_client) = xssh.create_ssh_client_connection(cluster_name)
for error in error_list:
log.write(f'{error}\n')
# create the SSH transport connection
if OK:
(OK, error_list,
ssh_transport) = xssh.create_ssh_transport_connection(cluster_name)
for error in error_list:
log.write(f'{error}\n')
# create the SFTP client
if OK:
sftp_client = xssh.create_sftp_client(ssh_transport)
# upload the database dataset
if OK:
# get the option dictionary
database_transfer_options_dict = xlib.get_option_dict(
get_database_transfer_config_file())
# get the database dataset identification and the local directory of the database files
database_dataset_id = database_transfer_options_dict['identification'][
'database_dataset_id']
local_dir = database_transfer_options_dict['identification'][
'local_dir']
# set the cluster database directory
cluster_database_dir = '{0}/{1}'.format(
xlib.get_cluster_database_dir(), database_dataset_id)
# create the data directory in the cluster
log.write(f'{xlib.get_separator()}\n')
log.write(
'The database directory {0} in the cluster is being created ...\n'.
format(cluster_database_dir))
command = 'mkdir --parents {0}'.format(cluster_database_dir)
(OK, stdout,
stderr) = xssh.execute_cluster_command(ssh_client, command)
if OK:
log.write('The directory is created.\n')
else:
log.write(f'*** ERROR: Wrong command ---> {command}\n')
# get the sections list
sections_list = []
for section in database_transfer_options_dict.keys():
sections_list.append(section)
sections_list.sort()
# for each section "file-n"
for section in sections_list:
# check than the section identification is like file-n
if re.match('^file-[0-9]+$', section):
# get the file name
file_name = database_transfer_options_dict[section][
'file_name']
# set the local path and cluster path
local_path = '{0}/{1}'.format(local_dir, file_name)
cluster_path = '{0}/{1}'.format(cluster_database_dir,
file_name)
# upload the database file to the cluster
log.write(f'{xlib.get_separator()}\n')
log.write('The file {0} is being uploaded to {1} ...\n'.format(
file_name, cluster_database_dir))
(OK, error_list) = xssh.put_file(sftp_client, local_path,
cluster_path)
if OK:
log.write('The file has been uploaded.\n')
else:
for error in error_list:
log.write(f'{error}\n')
break
# close the SSH transport connection
if OK:
xssh.close_ssh_transport_connection(ssh_transport)
# close the SSH client connection
if OK:
xssh.close_ssh_client_connection(ssh_client)
# warn that the log window can be closed
if not isinstance(log, xlib.DevStdOut):
log.write(f'{xlib.get_separator()}\n')
log.write('You can close this window now.\n')
# execute final function
if function is not None:
function()
# return the control variable
return OK
def run_gmap_process(cluster_name, log, function=None):
'''
Run a GMAP process.
'''
# initialize the control variable
OK = True
# get the GMAP option dictionary
gmap_option_dict = xlib.get_option_dict(get_gmap_config_file())
# get the experiment identification
experiment_id = gmap_option_dict['identification']['experiment_id']
# warn that the log window must not be closed
if not isinstance(log, xlib.DevStdOut):
log.write('This process might take several minutes. Do not close this window, please wait!\n')
# validate the GMAP config file
log.write('{0}\n'.format(xlib.get_separator()))
log.write('Validating the {0} config file ...\n'.format(xlib.get_gmap_name()))
(OK, error_list) = validate_gmap_config_file(strict=True)
if OK:
log.write('The config file is OK.\n')
else:
log.write('*** ERROR: The config file is not valid.\n')
log.write('Please correct this file or recreate the config files.\n')
# create the SSH client connection
if OK:
log.write('{0}\n'.format(xlib.get_separator()))
log.write('Connecting the SSH client ...\n')
(OK, error_list, ssh_client) = xssh.create_ssh_client_connection(cluster_name, 'master')
if OK:
log.write('The SSH client is connected.\n')
else:
for error in error_list:
log.write('{0}\n'.format(error))
# create the SSH transport connection
if OK:
log.write('{0}\n'.format(xlib.get_separator()))
log.write('Connecting the SSH transport ...\n')
(OK, error_list, ssh_transport) = xssh.create_ssh_transport_connection(cluster_name, 'master')
if OK:
log.write('The SSH transport is connected.\n')
else:
for error in error_list:
log.write('{0}\n'.format(error))
# create the SFTP client
if OK:
log.write('{0}\n'.format(xlib.get_separator()))
log.write('Connecting the SFTP client ...\n')
sftp_client = xssh.create_sftp_client(ssh_transport)
log.write('The SFTP client is connected.\n')
# warn that the requirements are being verified
if OK:
log.write('{0}\n'.format(xlib.get_separator()))
log.write('Verifying process requirements ...\n')
# verify the master is running
if OK:
(master_state_code, master_state_name) = xec2.get_node_state(cluster_name, 'master')
if master_state_code != 16:
log.write('*** ERROR: The cluster {0} is not running. Its state is {1} ({2}).\n'.format(cluster_name, master_state_code, master_state_name))
OK = False
# verify the GMAP-GSNAP is setup
if OK:
(OK, error_list, is_setup) = xbioinfoapp.is_setup_bioconda_package(xlib.get_gmap_gsnap_bioconda_code(), cluster_name, True, ssh_client)
if OK:
if not is_setup:
log.write('*** ERROR: {0} is not setup.\n'.format(xlib.get_gmap_name()))
OK = False
else:
log.write('*** ERROR: The verification of {0} setup could not be performed.\n'.format(xlib.get_gmap_name()))
# warn that the requirements are OK
if OK:
log.write('Process requirements are OK.\n')
# determine the run directory in the cluster
if OK:
log.write('{0}\n'.format(xlib.get_separator()))
log.write('Determining the run directory in the cluster ...\n')
current_run_dir = xlib.get_cluster_current_run_dir(experiment_id, xlib.get_gmap_code())
command = 'mkdir --parents {0}'.format(current_run_dir)
(OK, stdout, stderr) = xssh.execute_cluster_command(ssh_client, command)
if OK:
log.write('The directory path is {0}.\n'.format(current_run_dir))
else:
log.write('*** ERROR: Wrong command ---> {0}\n'.format(command))
# build the GMAP process script
if OK:
log.write('{0}\n'.format(xlib.get_separator()))
log.write('Building the process script {0} ...\n'.format(get_gmap_process_script()))
(OK, error_list) = build_gmap_process_script(cluster_name, current_run_dir)
if OK:
log.write('The file is built.\n')
if not OK:
log.write('*** ERROR: The file could not be built.\n')
# upload the GMAP process script in the cluster
if OK:
log.write('{0}\n'.format(xlib.get_separator()))
log.write('Uploading the process script {0} in the directory {1} of the master ...\n'.format(get_gmap_process_script(), current_run_dir))
cluster_path = '{0}/{1}'.format(current_run_dir, os.path.basename(get_gmap_process_script()))
(OK, error_list) = xssh.put_file(sftp_client, get_gmap_process_script(), cluster_path)
if OK:
log.write('The file is uploaded.\n')
else:
for error in error_list:
log.write('{0}\n'.format(error))
# set run permision to the GMAP process script in the cluster
if OK:
log.write('{0}\n'.format(xlib.get_separator()))
log.write('Setting on the run permision of {0}/{1} ...\n'.format(current_run_dir, os.path.basename(get_gmap_process_script())))
command = 'chmod u+x {0}/{1}'.format(current_run_dir, os.path.basename(get_gmap_process_script()))
(OK, stdout, stderr) = xssh.execute_cluster_command(ssh_client, command)
if OK:
log.write('The run permision is set.\n')
else:
log.write('*** ERROR: Wrong command ---> {0}\n'.format(command))
# build the GMAP process starter
if OK:
log.write('{0}\n'.format(xlib.get_separator()))
log.write('Building the process starter {0} ...\n'.format(get_gmap_process_starter()))
(OK, error_list) = build_gmap_process_starter(current_run_dir)
if OK:
log.write('The file is built.\n')
if not OK:
log.write('***ERROR: The file could not be built.\n')
# upload the GMAP process starter in the cluster
if OK:
log.write('{0}\n'.format(xlib.get_separator()))
log.write('Uploading the process starter {0} in the directory {1} of the master ...\n'.format(get_gmap_process_starter(), current_run_dir))
cluster_path = '{0}/{1}'.format(current_run_dir, os.path.basename(get_gmap_process_starter()))
(OK, error_list) = xssh.put_file(sftp_client, get_gmap_process_starter(), cluster_path)
if OK:
log.write('The file is uploaded.\n')
else:
for error in error_list:
log.write('{0}\n'.format(error))
# set run permision to the GMAP process starter in the cluster
if OK:
log.write('{0}\n'.format(xlib.get_separator()))
log.write('Setting on the run permision of {0}/{1} ...\n'.format(current_run_dir, os.path.basename(get_gmap_process_starter())))
command = 'chmod u+x {0}/{1}'.format(current_run_dir, os.path.basename(get_gmap_process_starter()))
(OK, stdout, stderr) = xssh.execute_cluster_command(ssh_client, command)
if OK:
log.write('The run permision is set.\n')
else:
log.write('*** ERROR: Wrong command ---> {0}\n'.format(command))
# submit the GMAP process
if OK:
log.write('{0}\n'.format(xlib.get_separator()))
log.write('Submitting the process script {0}/{1} ...\n'.format(current_run_dir, os.path.basename(get_gmap_process_starter())))
sge_env = xcluster.get_sge_env()
command = '{0}; qsub -V -b n -cwd {1}/{2}'.format(sge_env, current_run_dir, os.path.basename(get_gmap_process_starter()))
(OK, stdout, stderr) = xssh.execute_cluster_command(ssh_client, command)
if OK:
for line in stdout:
log.write('{0}\n'.format(line))
else:
log.write('*** ERROR: Wrong command ---> {0}\n'.format(command))
# close the SSH transport connection
if OK:
log.write('{0}\n'.format(xlib.get_separator()))
log.write('Closing the SSH transport connection ...\n')
xssh.close_ssh_transport_connection(ssh_transport)
log.write('The connection is closed.\n')
# close the SSH client connection
if OK:
log.write('{0}\n'.format(xlib.get_separator()))
log.write('Closing the SSH client connection ...\n')
xssh.close_ssh_client_connection(ssh_client)
log.write('The connection is closed.\n')
# warn that the log window can be closed
if not isinstance(log, xlib.DevStdOut):
log.write('{0}\n'.format(xlib.get_separator()))
log.write('You can close this window now.\n')
# execute final function
if function is not None:
function()
# return the control variable
return OK
def install_node_infrastructure_software(cluster_name, node_name, log):
'''
Install infrastructure software in a node.
'''
# initialize the control variable
OK = True
# get the infrastructure software installation script path in local compute
local_script_path = get_infrastructure_software_installation_script()
# set the infrastructure software installation script path in node
node_script_path = f'./{os.path.basename(local_script_path)}'
# set the infrastructure software installation log path in node
node_log_path = node_script_path[:node_script_path.find('.sh')] + '.log'
# build the infrastructure software installation script
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write(
f'Building the infrastructure software installation script {local_script_path} ...\n'
)
(OK, error_list
) = build_infrastructure_software_installation_script(cluster_name)
if OK:
log.write('The file is built.\n')
else:
for error in error_list:
log.write(f'{error}\n')
# create the SSH client connection
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write(f'Connecting the SSH client to node {node_name} ...\n')
(OK, error_list, ssh_client) = xssh.create_ssh_client_connection(
cluster_name, node_name)
if OK:
log.write('The SSH client is connected.\n')
else:
for error in error_list:
log.write(f'{error}\n')
# create the SSH transport connection
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write(f'Connecting the SSH transport to node {node_name} ...\n')
(OK, error_list, ssh_transport) = xssh.create_ssh_transport_connection(
cluster_name, node_name)
if OK:
log.write('The SSH transport is connected.\n')
else:
for error in error_list:
log.write(f'{error}\n')
# create the SFTP client
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write(f'Connecting the SFTP client to node {node_name} ...\n')
sftp_client = xssh.create_sftp_client(ssh_transport)
log.write('The SFTP client is connected.\n')
# download the AWSCLI2 compressed file to local computer
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write(
f'Downloading the {xlib.get_awscli_name()} compressed file to local computer ...\n'
)
local_path = f'{xlib.get_temp_dir()}/{xlib.get_awscli_name()}.zip'
if not os.path.exists(os.path.dirname(local_path)):
os.makedirs(os.path.dirname(local_path))
try:
urllib.request.urlretrieve(xlib.get_awscli_url(), local_path)
except Exception as e:
log.write(f'*** EXCEPTION: "{e}".')
log.write(
f'*** ERROR: The file {xlib.get_awscli_url()} can not be downloaded.\n'
)
OK = False
else:
log.write('The file is downloaded.\n')
# upload the AWSCLI2 compressed file to cluster
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write(
f'Uploading the {xlib.get_awscli_name()} compressed file to the cluster ...\n'
)
cluster_path = f'./{os.path.basename(local_path)}'
(OK, error_list) = xssh.put_file(sftp_client, local_path, cluster_path)
if OK:
log.write('The file is uploaded.\n')
else:
for error in error_list:
log.write(f'{error}\n')
# upload the infraestructe software installation script to the node
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write(
f'Uploading the infraestructe software installation script to the node {node_name} ...\n'
)
(OK, error_list) = xssh.put_file(sftp_client, local_script_path,
node_script_path)
if OK:
log.write('The file is uploaded.\n')
else:
for error in error_list:
log.write(f'{error}\n')
# set run permision to the infraestructe software installation script in the cluster
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write('Setting on the run permision ...\n')
command = f'chmod u+x {node_script_path}'
(OK, stdout,
stderr) = xssh.execute_cluster_command(ssh_client, command)
if OK:
log.write('The run permision is set.\n')
else:
log.write(f'*** ERROR: Wrong command ---> {command}\n')
# submit the infraestructe software installation script
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write(
f'Submitting the infraestructe software installation script in the node {node_name} ...\n'
)
command = f'{node_script_path} &>{node_log_path} &'
(OK, stdout,
stderr) = xssh.execute_cluster_command(ssh_client, command)
if OK:
log.write('The script is submitted.\n')
else:
log.write(f'*** ERROR: Wrong command ---> {command}\n')
# close the SSH transport connection
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write('Closing the SSH transport connection ...\n')
xssh.close_ssh_transport_connection(ssh_transport)
log.write('The connection is closed.\n')
# close the SSH client connection
if OK:
log.write(f'{xlib.get_separator()}\n')
log.write('Closing the SSH client connection ...\n')
xssh.close_ssh_client_connection(ssh_client)
log.write('The connection is closed.\n')
# return the control variable
return OK
def download_result_dataset(cluster_name, log, function=None):
'''
Download the result dataset of a run from the cluster.
'''
# initialize the control variable
OK = True
# get the read transfer config file
result_transfer_config_file = get_result_transfer_config_file()
# warn that the log window must not be closed
if not isinstance(log, xlib.DevStdOut):
log.write('This process might take several minutes. Do not close this window, please wait!\n')
# get and validate the result transfer config file
log.write('{0}\n'.format(xlib.get_separator()))
log.write('The result transfer config file is been validating ...\n')
if validate_result_transfer_config_file(strict=True):
log.write('The config file is OK.\n')
else:
log.write('*** ERROR: The result transfer config file is not valid.\n')
log.write('Please correct this file or recreate the config files.\n')
OK = False
# create the SSH client connection
if OK:
(OK, error_list, ssh_client) = xssh.create_ssh_client_connection(cluster_name, 'master')
for error in error_list:
log.write('{0}\n'.format(error))
# create the SSH transport connection
if OK:
(OK, error_list, ssh_transport) = xssh.create_ssh_transport_connection(cluster_name, 'master')
for error in error_list:
log.write('{0}\n'.format(error))
# create the SFTP client
if OK:
sftp_client = xssh.create_sftp_client(ssh_transport)
# get the options dictionary
if OK:
result_transfer_options_dict = xlib.get_option_dict(result_transfer_config_file)
# download the result dataset
if OK:
# get the sections list
sections_list = []
for section in result_transfer_options_dict.keys():
sections_list.append(section)
sections_list.sort()
# get the experiment identification, run identification and local directory from the section "identification"
experiment_id = result_transfer_options_dict['identification']['experiment_id']
result_dataset_id = result_transfer_options_dict['identification']['result_dataset_id']
status = result_transfer_options_dict['identification']['status'].lower()
local_dir = result_transfer_options_dict['identification']['local_dir']
# download files when the status is uncompressed
if status == 'uncompressed':
# for each section "file-n"
for section in sections_list:
# verify than the section identification is like file-n
if re.match('^file-[0-9]+$', section):
# get the dataset subdirectory and file name
dataset_subdirectory = result_transfer_options_dict[section]['dataset_subdirectory']
file_name = result_transfer_options_dict[section]['file_name']
# verify if the dataset subdirectory is created
pathlib.Path(os.path.normpath('{0}/{1}'.format(local_dir, dataset_subdirectory))).mkdir(parents=True, exist_ok=True)
# assign the cluster path and local path
cluster_path = '{0}/{1}/{2}/{3}/{4}'.format(xlib.get_cluster_result_dir(), experiment_id, result_dataset_id, dataset_subdirectory, file_name)
local_path = os.path.normpath('{0}/{1}/{2}'.format(local_dir, dataset_subdirectory, file_name))
# download the result file from the cluster
log.write('{0}\n'.format(xlib.get_separator()))
log.write('Downloading the file {0} to {1} ...\n'.format(cluster_path, local_dir))
(OK, error_list) = xssh.get_file(sftp_client, cluster_path, local_path)
if OK:
log.write('The file has been downloaded.\n')
else:
for error in error_list:
log.write('{0}\n'.format(error))
break
# download files when the status is compressed
elif status == 'compressed':
# assign the cluster path and local path
cluster_path = '{0}/{1}/{2}'.format(xlib.get_cluster_result_dir(), experiment_id, result_dataset_id)
local_path = '{0}/{1}'.format(local_dir, result_dataset_id)
# download the result file from the cluster
log.write('{0}\n'.format(xlib.get_separator()))
log.write('Downloading the file {0} to {1} ...\n'.format(cluster_path, local_dir))
(OK, error_list) = xssh.get_file(sftp_client, cluster_path, local_path)
if OK:
log.write('The file has been downloaded.\n')
else:
for error in error_list:
log.write('{0}\n'.format(error))
# close the SSH transport connection
if OK:
xssh.close_ssh_transport_connection(ssh_transport)
# close the SSH client connection
if OK:
xssh.close_ssh_client_connection(ssh_client)
# warn that the log window can be closed
if not isinstance(log, xlib.DevStdOut):
log.write('{0}\n'.format(xlib.get_separator()))
log.write('You can close this window now.\n')
# execute final function
if function is not None:
function()
# return the control variable
return OK
def form_view_cluster_experiment_process_log():
'''
View the log of an experiment process in the cluster.
'''
# initialize the control variable
OK = True
# print the header
clib.clear_screen()
clib.print_headers_with_environment('Logs - View an experiment process log in the cluster')
# get the clustner name
if OK:
print(xlib.get_separator())
if xec2.get_running_cluster_list(volume_creator_included=False) != []:
cluster_name = cinputs.input_cluster_name(volume_creator_included=False, help=True)
else:
print('WARNING: There is not any running cluster.')
OK = False
# create the SSH client connection
if OK:
(OK, error_list, ssh_client) = xssh.create_ssh_client_connection(cluster_name, 'master')
for error in error_list:
log.write('{0}\n'.format(error))
# create the SSH transport connection
if OK:
(OK, error_list, ssh_transport) = xssh.create_ssh_transport_connection(cluster_name, 'master')
for error in error_list:
log.write('{0}\n'.format(error))
# create the SFTP client
if OK:
sftp_client = xssh.create_sftp_client(ssh_transport)
# get the experiment identification
if OK:
experiment_id = cinputs.input_experiment_id(ssh_client, help=True)
if experiment_id == '':
print('WARNING: The cluster has not experiment data.')
OK = False
# get the result_dataset identification
if OK:
result_dataset_id = cinputs.input_result_dataset_id('uncompressed', ssh_client, experiment_id, help=True)
if result_dataset_id == '':
print('WARNING: The experiment {0} has not result datasets.'.format(experiment_id))
OK = False
# create the local path
if not os.path.exists(xlib.get_temp_dir()):
os.makedirs(xlib.get_temp_dir())
# get the log file name and build local and cluster paths
if OK:
log_file = xlib.get_cluster_log_file()
local_path = '{0}/{1}'.format(xlib.get_temp_dir(), log_file)
cluster_path = '{0}/{1}/{2}'.format(xlib.get_cluster_experiment_result_dir(experiment_id), result_dataset_id, log_file)
# download the log file from the cluster
if OK:
print(xlib.get_separator())
print('The file {0} is being downloaded from {1} ...'.format(log_file, cluster_path))
OK = xssh.get_file(sftp_client, cluster_path, local_path)
if OK:
print('The file has been uploaded.')
# close the SSH transport connection
if OK:
xssh.close_ssh_transport_connection(ssh_transport)
# close the SSH client connection
if OK:
xssh.close_ssh_client_connection(ssh_client)
# view the log file
if OK:
text = 'Logs - View an experiment process log in the cluster'
OK = clib.view_file(local_path, text)
# show continuation message
print(xlib.get_separator())
input('Press [Intro] to continue ...')
def form_view_cluster_start_log():
'''
View the cluster start log.
'''
# initialize the control variable
OK = True
# print the header
clib.clear_screen()
clib.print_headers_with_environment('Logs - View the cluster start log')
# get the clustner name
if OK:
print(xlib.get_separator())
if xec2.get_running_cluster_list(only_environment_cluster=True,
volume_creator_included=False) != []:
cluster_name = cinputs.input_cluster_name(
volume_creator_included=False, help=True)
else:
print('WARNING: There is not any running cluster.')
OK = False
# create the SSH client connection
if OK:
(OK, error_list,
ssh_client) = xssh.create_ssh_client_connection(cluster_name)
for error in error_list:
print(error)
# create the SSH transport connection
if OK:
(OK, error_list,
ssh_transport) = xssh.create_ssh_transport_connection(cluster_name)
for error in error_list:
print(error)
# create the SFTP client
if OK:
sftp_client = xssh.create_sftp_client(ssh_transport)
# create the local path
if not os.path.exists(xlib.get_temp_dir()):
os.makedirs(xlib.get_temp_dir())
# get the log file name and build local and cluster paths
if OK:
local_path = f'{xlib.get_temp_dir()}/{os.path.basename(xinstance.get_infrastructure_software_installation_log())}'
cluster_path = f'/home/ubuntu/{os.path.basename(xinstance.get_infrastructure_software_installation_log())}'
# download the log file from the cluster
if OK:
print(xlib.get_separator())
print(
f'The file {os.path.basename(xinstance.get_infrastructure_software_installation_log())} is being downloaded from {cluster_path} ...'
)
OK = xssh.get_file(sftp_client, cluster_path, local_path)
if OK:
print('The file has been uploaded.')
# close the SSH transport connection
if OK:
xssh.close_ssh_transport_connection(ssh_transport)
# close the SSH client connection
if OK:
xssh.close_ssh_client_connection(ssh_client)
# view the log file
if OK:
text = 'Logs - View the cluster start log'
OK = clib.view_file(local_path, text)
# show continuation message
input('Press [Intro] to continue ...')
