Cloudify Manager installation failure with CLI bootstrapping
Luca Spinacci
I've been trying to install Cloudify on my Windows 7 based PC with no success yet:
1. installed Cloudify CLI (.exe for Windows 7)
2. Cloudify Manager installation according to option 2 (bootstrapping from CLI).
For the step 2 above I configured the file simple-manager-blueprint-inputs.yaml (attached below) and installed a SSH server for Windows on the same host.
I followed the online guide:
http://docs.getcloudify.org/4.1.0/installation/bootstrapping/#option-2-bootstrapping-a-cloudify-manager
While bootstrapping I got a recursive error message that I couldn't fix:
CFY <manager> [amqp_influx_rfu9uq.creation] Task failed 'fabric_plugin.tasks.run_scripts' -> Needed to prompt for a connection or sudo password (host: 127.0.0.1), but abort-on-prompts was set to True
Thank you for the help.
BR,
Luca.
P.S.
Follows my simple-manager-blueprint-inputs.yaml:
#############################
# Provider specific Inputs
#############################
# The public IP of the manager to which the CLI will connect.
public_ip: '127.0.0.1'
# The manager's private IP address. This is the address which will be used by the
# application hosts to connect to the Manager's fileserver and message broker.
private_ip: '127.0.0.1'
# SSH user used to connect to the manager
ssh_user: 'admin'
# SSH port used to connect to the manager
#ssh_port: 22
# SSH key path used to connect to the manager
ssh_key_filename: '/sshkey/RSAKey.cfg'
# This is the user with which the Manager will try to connect to the application hosts.
agents_user: ubuntu
#resources_prefix: ''
#############################
# Security Settings
#############################
# Enabling SSL limits communication with the server to SSL only.
# NOTE: If enabled, the certificate and private key files must reside in resources/ssl.
#ssl_enabled: false
# Username and password of the Cloudify administrator.
# This user will also be included in the simple userstore repostiroty if the
# simple userstore implementation is used.
admin_username: 'admin'
admin_password: 'admin'
#insecure_endpoints_disabled: true
#############################
# Bootstrap Validations
#############################
# Validations are performed to check that attributes like disk space and memory
# correspond with some prerequisites and that some resources are available for
# download.
# Setting to `true` will allow to ignore those validations.
#ignore_bootstrap_validations: false
# These allow to override specific validation values
# NOTE: We do not recommend changing these values unless you know exactly
# what you're doing.
#minimum_required_total_physical_memory_in_mb: 3792
#minimum_required_available_disk_space_in_gb: 5
#############################
# Manager Resources Package
#############################
#manager_resources_package: http://repository.cloudifysource.org/cloudify/4.1.1/ga-release/cloudify-manager-resources_4.1.1-ga.tar.gz
# Providing a checksum file url will allow validating the resources package.
# By default, no validation is performed. Providing a checksum file will use
# the file to validate. Note that not providing a file but changing
# `skip_checksum_validation` to false means we will try to guess the location
# of an md5 checksum file and validate against it.
# You can download our md5 checksum file by appending .md5
# to the `manager_resources_package` url.
#manager_resources_package_checksum_file: ''
#skip_checksum_validation: true
#############################
# Agent Packages
#############################
# The key names must be in the format: distro_release_agent (e.g. ubuntu_trusty_agent)
# as the key is what's used to name the file, which later allows our
# agent installer to identify it for your distro and release automatically.
# Note that the windows agent key name MUST be `cloudify_windows_agent`
#agent_package_urls:
# ubuntu_trusty_agent: ''
# ubuntu_precise_agent: ''
# centos_7x_agent: ''
# centos_6x_agent: ''
# redhat_7x_agent: ''
# redhat_6x_agent: ''
# cloudify_windows_agent: ''
#############################
# Cloudify Modules
#############################
# Note that you can replace rpm urls with names of packages as long as they're available in your default yum repository.
# That is, as long as they provide the exact same version of that module.
#rest_service_rpm_source_url: ''
#management_worker_rpm_source_url: ''
#amqpinflux_rpm_source_url: ''
#cloudify_resources_url: ''
#stage_source_url: ''
#composer_source_url: ''
# This is a Cloudify specific redistribution of Grafana.
#grafana_source_url: ''
#############################
# External Components
#############################
# Note that you can replace rpm urls with names of packages as long as they're available in your default yum repository.
# That is, as long as they provide the exact same version of that module.
#pip_source_rpm_url: ''
#java_source_url: ''
# RabbitMQ Distribution of Erlang
#erlang_source_url: ''
#rabbitmq_source_url: ''
#logstash_source_url: ''
#nginx_source_url: ''
#influxdb_source_url: ''
#riemann_source_url: ''
# A RabbitMQ Client for Riemann
#langohr_source_url: ''
# Riemann's default daemonizer
#daemonize_source_url: ''
#nodejs_source_url: ''
##############################
# Workflow configuration
##############################
# Sets the number of times a failed task will be retried on recoverable error.
#workflow_task_retries: 60
# Sets the interval between retry attempts in seconds.
#workflow_task_retry_interval: 15
##################################
# Management Workers configuration
##################################
# Sets the logging level to use for the management workers. This affects the logging performed
# by the manager during the execution of management tasks, such as deployment creation
# and deployment deletion.
# NOTE: specifying "debug" will result in considerable amount of logging activity. Consider
# using "info" (or a more restrictive level) for production environments.
#management_worker_log_level: debug
#############################
# RabbitMQ Configuration
#############################
# Sets the username/password to use for clients such as celery
# to connect to the rabbitmq broker.
# It is recommended that you set both the username and password
# to something reasonably secure.
#rabbitmq_username: 'cloudify'
#rabbitmq_password: 'c10udify'
# Allows to define the message-ttl for the different types of queues (in milliseconds).
# https://www.rabbitmq.com/ttl.html
#rabbitmq_events_queue_message_ttl: 60000
#rabbitmq_logs_queue_message_ttl: 60000
#rabbitmq_metrics_queue_message_ttl: 60000
# This will set the queue length limit. Note that while new messages
# will be queued in RabbitMQ, old messages will be deleted once the
# limit is reached!
# Note this is NOT the message byte length!
# https://www.rabbitmq.com/maxlength.html
#rabbitmq_events_queue_length_limit: 1000000
#rabbitmq_logs_queue_length_limit: 1000000
#rabbitmq_metrics_queue_length_limit: 1000000
# RabbitMQ File Descriptors Limit
#rabbitmq_fd_limit: 102400
#############################
# PostgreSQL Configuration
#############################
# You can configure the PostgreSQL database name for cloudify
#postgresql_db_name: 'cloudify_db'
#postgresql_host: 'localhost'
#############################
# LDAP Configuration
#############################
#ldap_server: ''
#ldap_username: ''
#ldap_password: ''
#ldap_domain: ''
#ldap_is_active_directory: true
#ldap_dn_extra: ''
#############################
# InfluxDB Configuration
#############################
# You can configure an external endpoint of an InfluxDB Cluster to use
# instead of the built in one.
# If one is provided, the built in InfluxDB cluster will not run.
# Note that the port is currently not configurable and must remain 8086.
# Also note that the database username and password are hardcoded to root:root.
#influxdb_endpoint_ip: ''
#################################
# Management Worker Configuration
#################################
# Maximum number of worker processes started by the management worker.
#management_worker_max_workers: 100
# Minimum number of worker processes maintained by the management worker.
#management_worker_min_workers: 2
#################################
# REST Configuration
#################################
# valid values: public_ip, private_ip
#rest_host_external_endpoint_type: public_ip
# valid values: public_ip, private_ip
#rest_host_internal_endpoint_type: private_ip
# The number of gunicorn worker processes for handling requests.
# If the default value (0) is set, then min((2 * cpu_count + 1 processes), 12) will be used.
#rest_service_gunicorn_worker_count: 0
# The maximum number of requests a worker will process before restarting.
# If this is set to zero then the automatic worker restarts are disabled.
#rest_service_gunicorn_max_requests: 1000
# Logging level for the REST service. Defaults to 'INFO', as 'DEBUG' may end up
# logging sensitive information.
#rest_service_log_level: INFO
# The size, in MB, that the REST service log file may grow to before it's rotated.
#rest_service_log_file_size: 100
# Number of historical log files to keep when rotating the REST service logs.
#rest_service_log_files_backup_count: 20
###############################
# Images configuration
###############################
# Sets whether the first time the manager's VM is booted from a Cloudify manager image
# its private IP will be propagated to all relevant configuration files.
#set_manager_ip_on_boot: false
#############################
# Offline Resources Upload
#############################
# You can configure a set of resources to upload at bootstrap. These resources
# will reside on the manager and enable offline deployment. `dsl_resources`
# should contain any resource needed in the parsing process (i.e. plugin.yaml files)
#dsl_resources:
# - {'source_path': 'http://www.getcloudify.org/spec/fabric-plugin/1.5/plugin.yaml', 'destination_path': '/spec/fabric-plugin/1.5/plugin.yaml'}
# - {'source_path': 'http://www.getcloudify.org/spec/script-plugin/1.5/plugin.yaml', 'destination_path': '/spec/script-plugin/1.5/plugin.yaml'}
# - {'source_path': 'http://www.getcloudify.org/spec/diamond-plugin/1.3.6/plugin.yaml', 'destination_path': '/spec/diamond-plugin/1.3.6/plugin.yaml'}
# - {'source_path': 'http://www.getcloudify.org/spec/aws-plugin/1.4.3/plugin.yaml', 'destination_path': '/spec/aws-plugin/1.4.3/plugin.yaml'}
# - {'source_path': 'http://www.getcloudify.org/spec/openstack-plugin/2.0.1/plugin.yaml', 'destination_path': '/spec/openstack-plugin/2.0.1/plugin.yaml'}
# - {'source_path': 'http://www.getcloudify.org/spec/tosca-vcloud-plugin/1.3.1/plugin.yaml', 'destination_path': '/spec/tosca-vcloud-plugin/1.3.1/plugin.yaml'}
# - {'source_path': 'http://www.getcloudify.org/spec/vsphere-plugin/2.0.1/plugin.yaml', 'destination_path': '/spec/vsphere-plugin/2.0.1/plugin.yaml'}
# - {'source_path': 'http://www.getcloudify.org/spec/cloudify/4.1.1/types.yaml', 'destination_path': '/spec/cloudify/4.1.1/types.yaml'}
###############################
# Import Resolver Configuration
###############################
# An imported URL is prefix-matched against the key in each entry. If a match is found,
# then the URL prefix is replaced with the value of the corresponding entry.
# That allows serving YAML files from within the manager, even when the imported URL
# points to the external network.
#import_resolver_rules:
# - {'http://www.getcloudify.org/spec': 'file:///opt/manager/resources/spec'}
Tadej Borovšak
> I've been trying to install Cloudify on my Windows 7 based PC with no success yet:
>
> 1. installed Cloudify CLI (.exe for Windows 7)
> 2. Cloudify Manager installation according to option 2 (bootstrapping from
> CLI).
> For the step 2 above I configured the file
> simple-manager-blueprint-inputs.yaml (attached below) and installed a SSH
> server for Windows on the same host.
>
> I followed the online guide:
> http://docs.getcloudify.org/4.1.0/installation/bootstrapping/#option-2-bootstrapping-a-cloudify-manager
stated in documentation that you linked:
Cloudify Manager must run on a 64-bit machine with a RHEL/CentOS 7.x or
higher platform, and requires at the very least 2 vCPUs, 4GB RAM and 5GB
of free disk space. These are the minimal requirements for a Cloudify
Manager to run, and are only sufficient for demos and development. You
need to provision larger machines to actually utilize the Manager’s
capabilities.
Cheers,
Tadej
--
Tadej Borovšak
tade...@gmail.com
tadej.b...@gmail.com