service-fabrik-broker job from service-fabrik/3.20.0

Github source: abe31f9 or master branch

Properties¶

broker¶

backup¶

abort_time_out¶

Timeout time for abort of backup to complete in ms (defaults to 5 mins)

Default

300000

backup_restore_status_check_every¶

Status of deployment backup/restore once in every ms

Default

120000

backup_restore_status_poller_timeout¶

Timeout for backup/restore status checker in ms

Default

8.64e+07

lock_check_delay_on_restart¶

On restart of service fabrik queries all deployments to see if there is a lock on it. This delay ensures each call is spaced with this delay (ms)

Default

5000

max_num_on_demand_backup¶

Maximum number of on-demand backups

Default

2

provider¶

IaaS-specific backup provider configuration

retention_period_in_days¶

Scheduled backups are going to be maintained during this period & cannot be deleted. Beyond this period backups are automatically deleted by backup job

Default

14

retry_delay_on_error¶

In case of unlock failure 3 retry attempts will be done with this configured delay (milliseconds) in exponential manner

Default

60000

status_check_every¶

Interval in milliseconds to check the status of service fabrik backup

Default

120000

cf¶

password¶

Admin password for the Cloud Foundry cloud controller

url¶

URL of the Cloud Foundry cloud controller

username¶

Admin username for the Cloud Foundry cloud controller

circuit_breaker¶

Overridden circuit breaker configuration

directors¶

URL of the Bosh directors

docker¶

allocate_docker_host_ports¶

Allocate Docker host ports when creating a container

Default

true

job¶

Name of the swarm/docker job

Default

swarm_manager

skip_ssl_validation¶

Determines whether the broker should verify SSL certificates when communicating with the Swarm Manager

Default

true

url¶

Docker URL (HTTP address or Unix socket)

Default

https://10.11.252.10:2376

volume_driver¶

Volume driver used for Docker containers (only local or lvm-volume-driver are currently supported)

Default

local

enable_circuit_breaker¶

Switch used to turn on/off circuit breaker

Default

false

event¶

defaults¶

cf_last_operation¶

Determines if an HTTP operation is to be additionally treated with cf-last operation semantics

Default

false

http_inprogress_codes¶

Defines the default in progress HTTP status codes

Default

- 202

http_success_codes¶

Defines the default HTTP success codes

Default

  - 200
  - 201

ignore_service_fabrik_operation¶

Determines if a broker API is being invoked via service fabrik API operation

Default

false

include_response_body¶

Determines wheter HTTP response body is to be logged or not

Default

false

log_inprogress_state¶

Determines if an in-progress event is to be logged or not

Default

true

delete¶

http_success_codes¶

Defines the HTTP success codes for Delete operation

Default

  - 200
  - 410

external¶

api_requires_admin_scope¶

Only Administrators of the Cloud Foundry cloud controller are allowed to use the Service Fabrik API endpoint

Default

false

cookie_secret¶

Unique secret key, used to sign sessions

host¶

Domain name used for external endpoints such as dashboards or the Service Fabrik API (will be registered at the router)

log_event¶

Determines whether event logging must be enabled or not

Default

true

port¶

Port used for external endpoints such as dashboards or the Service Fabrik API

Default

9292

session_expiry¶

Session expiry time of the session in seconds

Default

86400

trust_proxy¶

This is required when running an Express app behind a proxy (see http://expressjs.com/en/guide/behind-proxies.html)

Default

2

feature¶

ServiceInstanceAutoUpdate¶

Switch used to turn on/off schedule update feature

Default

false

http_timeout¶

Timeout duration for any request to broker

Default

120000

internal¶

domain_socket¶

path¶

domain socket path used to communicate with broker monitoring agent

Default

/tmp/sfevents

ip¶

IP address used for internal endpoints such as administration or the service broker API

log_event¶

Determines whether event logging must be enabled or not

Default

true

port¶

Port used for internal endpoints such as administration or the service broker API

Default

9443

ssl¶

cert¶

Signed certificate used for internal communication

key¶

Private key used for internal communication

log_level¶

Log level of the broker

Default

info

mongodb¶

agent¶

provider¶

container¶

S3/Swift container name where service fabrik’s mongodb backup blobs will be stored

backup¶

schedule_interval¶

Cron expression defining the backup interval for service fabrik’s internal mongodb’s backup job

deployment_name¶

Bosh deployment name of the internal mongo-db

Default

service-fabrik-mongodb

provision¶

network_index¶

Network segment index within the service fabrik network where the internal mongodb is to be provisioned

Default

1

plan_id¶

Mongo Plan Id from the service catalog which is to be used for provisioning internal mongodb by Fabrik

record_max_fetch_count¶

Max number of records that can be fetched at a time from DB

Default

100

retry_connect¶

max_attempt¶

Maximum retry attempts for connecting to DB on errors

Default

8

min_delay¶

Minimum delay before retry attempt to connect to DB

Default

120000

url¶

MongoDB connection URL for service-fabrik’s internal needs. ex : mongodb://user:pass@localhost:port/database

monitoring¶

event_name_prefix¶

All riemann events originating from service fabrik will have this prefix

Default

CF.service-fabrik

events_logged_in_db¶

Comma seperated list of event names that are to be logged in DB

Default

create_backup, update_instance

include_response_body¶

Determines if the riemann event should contain the HTTP method response while logging

Default

false

unauthorized¶

include_response_body¶

Determines if the unauthorized event should contain the HTTP method response while logging

Default

true

name¶

Name of the service broker as it will be registered at the Cloud Foundry cloud controller

Default

service-fabrik-broker

password¶

Broker’s basic auth password

Default

secret

quota¶

enabled¶

If Quota Management Service is enabled

Default

false

oauthDomain¶

Oauth domain for quota service

password¶

Client secret for quota service

serviceDomain¶

Onboarding Service domain for quota service

username¶

clientId for quota service

whitelist¶

List of whitelisted organitions for which quota check should not happen

riemann¶

host¶

Riemann Host IP

Default

10.1.3.1

log_additional_event¶

Boolean configuration to log additional event to Riemann

Default

true

port¶

Riemann Port

Default

5555

prefix¶

Riemann Prefix

Default

CF

show_errors¶

Determines whether Riemann errors are to be logged or not

Default

true

skip_http_status_40x¶

Boolean configuration to skip logging events to riemann when the HTTP response from broker indicates a client error

Default

true

scheduler¶

agenda_collection¶

Name of the collection in mongodb which is to be used by agendaJS to store the scheduled job meta info

Default

agendaJobs

default_concurrency¶

A number which specifies the default number of a specific job that can be running at any given moment

Default

20

default_lock_lifetime¶

Specifies the default lock lifetime of a batchjob in milliseconds. A job cannot be re-run when within this locktime period (in case of errors, crash etc.)

Default

180000

job_types¶

Comma seperated list of batch job types that are enabled in service fabrik. Jobs can be enabled disabled by changing this list.

jobs¶

reschedule_delay¶

Jobs when found to be conflicting with each other are rescheduled after this delay (human-interval)

Default

20 minutes from now

scheduled_backup¶

max_attempts¶

If an update is in-progess while backup is to be run, then the backup will be rescheduled after a delay. This will be repeated for the below configured times if update is long running

Default

3

service_instance_update¶

max_attempts¶

In case of an error while trigerring an update, max retry attempts will be done based on the below configured value (note errors due to running backups are not considered)

Default

3

run_every_xdays¶

Runs the service instance update job once every configured number of days

Default

7

should_start_after_hr¶

Service instance update job schedule is randomized but this param specifies the start time (hrs) after which the randomization must be done

Default

8

should_start_after_min¶

Service instance update job schedule is randomized but this param specifies the start time (mins) after which the randomization must be done

Default

0

should_start_before_hr¶

Service instance update job schedule is randomized but this param specifies the start time (hrs) before which the randomization must be done

Default

17

should_start_before_min¶

Service instance update job schedule is randomized but this param specifies the start time (mins) before which the randomization must be done

Default

59

time_zone¶

Timezone in which the service instance update job runs

Default

Asia/Kolkata

maintenance_check_interval¶

Once scheduler is placed in maintenace mode, it checks every configured (ms) for status of maitenance mode

Default

90000

maintenance_mode_time_out¶

(ms) - wait for sf deployment to complete. Beyond this timeout out time scheduler starts

Default

3.6e+07

max_concurrency¶

A number which specifies the max number of batch jobs that can be running at any given moment

Default

30

max_workers¶

Max number of Job process workers

Default

2

process_every¶

Interval at which scheduler will query the database looking for jobs that need to be processed

Default

1 minute

run_with_web_process¶

Flag which indicates if scheduler can run in the same process as that of service fabrik broker web process

Default

false

start_delay¶

Scheduler start must be delayed for DB connections to happen

Default

120000

system_jobs¶

backup_reaper¶

interval¶

Cron expression defining schedule interval for backup reaper job

Default

0 1 * * *

job_data¶

delete_delay¶

delay in processing between two tasks (in millisecond)

Default

1000

dbcollection_reaper¶

event_detail¶

retention_in_days¶

Retention period in days for the event details (create_backup/update_instance) stored in SF. Data older than this is deleted

Default

90

interval¶

Cron expression defining schedule interval for db collection reaper job

Default

0 2 * * *

jobrun_detail¶

retention_in_days¶

Retention period in days for the agenda job run details. Data older than this is deleted

Default

60

services¶

Services and plans offered by the broker

skip_ssl_validation¶

Determines whether the broker should verify SSL certificates when communicating with other endpoints such as the cloud controller or the UAA

Default

true

sys_log_level¶

Log level of the messages that are to be forwarded to ELK via syslog transport

Default

info

syslog¶

host¶

Syslog ingestor host IP of ELK stack

Default

10.1.4.3

port¶

Syslog ingestor Port

Default

5514

username¶

Broker’s basic auth username

Default

broker

common¶

tls_cacert¶

Trust only remotes providing a certificate signed by the CA given here

tls_client_cert¶

TLS certificate file

tls_client_key¶

TLS key file

Templates¶

Templates are rendered and placed onto corresponding instances during the deployment process. This job's templates will be placed into /var/vcap/jobs/service-fabrik-broker/ directory (learn more).

• bin/health_check (from bin/health_check.erb)
• bin/job_properties.sh (from bin/job_properties.sh.erb)
• bin/service-fabrik-broker_ctl (from bin/service-fabrik-broker_ctl.erb)
• config/circuit-breaker-config.yml (from config/circuit-breaker-config.yml.erb)
• config/eventlog-config-external.yml (from config/eventlog-config-external.yml.erb)
• config/eventlog-config-internal.yml (from config/eventlog-config-internal.yml.erb)
• config/settings.yml (from config/settings.yml.erb)
• config/syslog_forwarding.conf (from config/syslog_forwarding.conf.erb)

Packages¶

Packages are compiled and placed onto corresponding instances during the deployment process. Packages will be placed into /var/vcap/packages/ directory.

• bosh-helpers
• node
• service-fabrik-broker