aboutsummaryrefslogtreecommitdiffstats
path: root/fdio.infra.terraform/terraform-nomad-pyspark-etl
diff options
context:
space:
mode:
Diffstat (limited to 'fdio.infra.terraform/terraform-nomad-pyspark-etl')
-rw-r--r--fdio.infra.terraform/terraform-nomad-pyspark-etl/conf/nomad/etl.hcl.tftpl298
-rw-r--r--fdio.infra.terraform/terraform-nomad-pyspark-etl/fdio/variables.tf2
-rw-r--r--fdio.infra.terraform/terraform-nomad-pyspark-etl/fdio/versions.tf2
-rw-r--r--fdio.infra.terraform/terraform-nomad-pyspark-etl/variables.tf4
4 files changed, 153 insertions, 153 deletions
diff --git a/fdio.infra.terraform/terraform-nomad-pyspark-etl/conf/nomad/etl.hcl.tftpl b/fdio.infra.terraform/terraform-nomad-pyspark-etl/conf/nomad/etl.hcl.tftpl
index 361ba357fa..0d0ecfa318 100644
--- a/fdio.infra.terraform/terraform-nomad-pyspark-etl/conf/nomad/etl.hcl.tftpl
+++ b/fdio.infra.terraform/terraform-nomad-pyspark-etl/conf/nomad/etl.hcl.tftpl
@@ -86,8 +86,8 @@ job "${job_name}" {
image = "${image}"
command = "gluesparksubmit"
args = [
- "--driver-memory", "20g",
- "--executor-memory", "20g",
+ "--driver-memory", "30g",
+ "--executor-memory", "30g",
"trending.py"
]
work_dir = "/local/csit/csit.infra.etl"
@@ -172,151 +172,151 @@ job "${job_name}" {
}
}
}
- group "${job_name}-rls2206" {
- # The restart stanza configures a tasks's behavior on task failure. Restarts
- # happen on the client that is running the task.
- #
- # https://www.nomadproject.io/docs/job-specification/restart
- #
- restart {
- mode = "fail"
- }
-
- # The constraint allows restricting the set of eligible nodes. Constraints
- # may filter on attributes or client metadata.
- #
- # For more information and examples on the "volume" stanza, please see
- # the online documentation at:
- #
- # https://www.nomadproject.io/docs/job-specification/constraint
- #
- constraint {
- attribute = "$${attr.cpu.arch}"
- operator = "!="
- value = "arm64"
- }
-
- constraint {
- attribute = "$${node.class}"
- value = "builder"
- }
-
- # The "task" stanza creates an individual unit of work, such as a Docker
- # container, web application, or batch processing.
- #
- # https://www.nomadproject.io/docs/job-specification/task.html
- #
- task "${job_name}-coverage" {
- # The artifact stanza instructs Nomad to fetch and unpack a remote
- # resource, such as a file, tarball, or binary.
- #
- # https://www.nomadproject.io/docs/job-specification/artifact
- #
- artifact {
- source = "git::https://github.com/FDio/csit"
- destination = "local/csit"
- }
-
- # The "driver" parameter specifies the task driver that should be used to
- # run the task.
- driver = "docker"
-
- # The "config" stanza specifies the driver configuration, which is passed
- # directly to the driver to start the task. The details of configurations
- # are specific to each driver, so please see specific driver
- # documentation for more information.
- config {
- image = "${image}"
- command = "gluesparksubmit"
- args = [
- "--driver-memory", "20g",
- "--executor-memory", "20g",
- "--executor-cores", "2",
- "--master", "local[2]",
- "coverage_rls2206.py"
- ]
- work_dir = "/local/csit/csit.infra.etl"
- }
-
- # The env stanza configures a list of environment variables to populate
- # the task's environment before starting.
- env {
- AWS_ACCESS_KEY_ID = "${aws_access_key_id}"
- AWS_SECRET_ACCESS_KEY = "${aws_secret_access_key}"
- AWS_DEFAULT_REGION = "${aws_default_region}"
- OUT_AWS_ACCESS_KEY_ID = "${out_aws_access_key_id}"
- OUT_AWS_SECRET_ACCESS_KEY = "${out_aws_secret_access_key}"
- OUT_AWS_DEFAULT_REGION = "${out_aws_default_region}"
- ${ envs }
- }
-
- # The "resources" stanza describes the requirements a task needs to
- # execute. Resource requirements include memory, network, cpu, and more.
- # This ensures the task will execute on a machine that contains enough
- # resource capacity.
- #
- # https://www.nomadproject.io/docs/job-specification/resources
- #
- resources {
- cpu = ${cpu}
- memory = ${memory}
- }
- }
- task "${job_name}-iterative" {
- # The artifact stanza instructs Nomad to fetch and unpack a remote
- # resource, such as a file, tarball, or binary.
- #
- # https://www.nomadproject.io/docs/job-specification/artifact
- #
- artifact {
- source = "git::https://github.com/FDio/csit"
- destination = "local/csit"
- }
-
- # The "driver" parameter specifies the task driver that should be used to
- # run the task.
- driver = "docker"
-
- # The "config" stanza specifies the driver configuration, which is passed
- # directly to the driver to start the task. The details of configurations
- # are specific to each driver, so please see specific driver
- # documentation for more information.
- config {
- image = "${image}"
- command = "gluesparksubmit"
- args = [
- "--driver-memory", "20g",
- "--executor-memory", "20g",
- "--executor-cores", "2",
- "--master", "local[2]",
- "iterative_rls2206.py"
- ]
- work_dir = "/local/csit/csit.infra.etl"
- }
-
- # The env stanza configures a list of environment variables to populate
- # the task's environment before starting.
- env {
- AWS_ACCESS_KEY_ID = "${aws_access_key_id}"
- AWS_SECRET_ACCESS_KEY = "${aws_secret_access_key}"
- AWS_DEFAULT_REGION = "${aws_default_region}"
- OUT_AWS_ACCESS_KEY_ID = "${out_aws_access_key_id}"
- OUT_AWS_SECRET_ACCESS_KEY = "${out_aws_secret_access_key}"
- OUT_AWS_DEFAULT_REGION = "${out_aws_default_region}"
- ${ envs }
- }
-
- # The "resources" stanza describes the requirements a task needs to
- # execute. Resource requirements include memory, network, cpu, and more.
- # This ensures the task will execute on a machine that contains enough
- # resource capacity.
- #
- # https://www.nomadproject.io/docs/job-specification/resources
- #
- resources {
- cpu = ${cpu}
- memory = ${memory}
- }
- }
- }
+# group "${job_name}-rls2206" {
+# # The restart stanza configures a tasks's behavior on task failure. Restarts
+# # happen on the client that is running the task.
+# #
+# # https://www.nomadproject.io/docs/job-specification/restart
+# #
+# restart {
+# mode = "fail"
+# }
+#
+# # The constraint allows restricting the set of eligible nodes. Constraints
+# # may filter on attributes or client metadata.
+# #
+# # For more information and examples on the "volume" stanza, please see
+# # the online documentation at:
+# #
+# # https://www.nomadproject.io/docs/job-specification/constraint
+# #
+# constraint {
+# attribute = "$${attr.cpu.arch}"
+# operator = "!="
+# value = "arm64"
+# }
+#
+# constraint {
+# attribute = "$${node.class}"
+# value = "builder"
+# }
+#
+# # The "task" stanza creates an individual unit of work, such as a Docker
+# # container, web application, or batch processing.
+# #
+# # https://www.nomadproject.io/docs/job-specification/task.html
+# #
+# task "${job_name}-coverage" {
+# # The artifact stanza instructs Nomad to fetch and unpack a remote
+# # resource, such as a file, tarball, or binary.
+# #
+# # https://www.nomadproject.io/docs/job-specification/artifact
+# #
+# artifact {
+# source = "git::https://github.com/FDio/csit"
+# destination = "local/csit"
+# }
+#
+# # The "driver" parameter specifies the task driver that should be used to
+# # run the task.
+# driver = "docker"
+#
+# # The "config" stanza specifies the driver configuration, which is passed
+# # directly to the driver to start the task. The details of configurations
+# # are specific to each driver, so please see specific driver
+# # documentation for more information.
+# config {
+# image = "${image}"
+# command = "gluesparksubmit"
+# args = [
+# "--driver-memory", "20g",
+# "--executor-memory", "20g",
+# "--executor-cores", "2",
+# "--master", "local[2]",
+# "coverage_rls2206.py"
+# ]
+# work_dir = "/local/csit/csit.infra.etl"
+# }
+#
+# # The env stanza configures a list of environment variables to populate
+# # the task's environment before starting.
+# env {
+# AWS_ACCESS_KEY_ID = "${aws_access_key_id}"
+# AWS_SECRET_ACCESS_KEY = "${aws_secret_access_key}"
+# AWS_DEFAULT_REGION = "${aws_default_region}"
+# OUT_AWS_ACCESS_KEY_ID = "${out_aws_access_key_id}"
+# OUT_AWS_SECRET_ACCESS_KEY = "${out_aws_secret_access_key}"
+# OUT_AWS_DEFAULT_REGION = "${out_aws_default_region}"
+# ${ envs }
+# }
+#
+# # The "resources" stanza describes the requirements a task needs to
+# # execute. Resource requirements include memory, network, cpu, and more.
+# # This ensures the task will execute on a machine that contains enough
+# # resource capacity.
+# #
+# # https://www.nomadproject.io/docs/job-specification/resources
+# #
+# resources {
+# cpu = ${cpu}
+# memory = ${memory}
+# }
+# }
+# task "${job_name}-iterative" {
+# # The artifact stanza instructs Nomad to fetch and unpack a remote
+# # resource, such as a file, tarball, or binary.
+# #
+# # https://www.nomadproject.io/docs/job-specification/artifact
+# #
+# artifact {
+# source = "git::https://github.com/FDio/csit"
+# destination = "local/csit"
+# }
+#
+# # The "driver" parameter specifies the task driver that should be used to
+# # run the task.
+# driver = "docker"
+#
+# # The "config" stanza specifies the driver configuration, which is passed
+# # directly to the driver to start the task. The details of configurations
+# # are specific to each driver, so please see specific driver
+# # documentation for more information.
+# config {
+# image = "${image}"
+# command = "gluesparksubmit"
+# args = [
+# "--driver-memory", "20g",
+# "--executor-memory", "20g",
+# "--executor-cores", "2",
+# "--master", "local[2]",
+# "iterative_rls2206.py"
+# ]
+# work_dir = "/local/csit/csit.infra.etl"
+# }
+#
+# # The env stanza configures a list of environment variables to populate
+# # the task's environment before starting.
+# env {
+# AWS_ACCESS_KEY_ID = "${aws_access_key_id}"
+# AWS_SECRET_ACCESS_KEY = "${aws_secret_access_key}"
+# AWS_DEFAULT_REGION = "${aws_default_region}"
+# OUT_AWS_ACCESS_KEY_ID = "${out_aws_access_key_id}"
+# OUT_AWS_SECRET_ACCESS_KEY = "${out_aws_secret_access_key}"
+# OUT_AWS_DEFAULT_REGION = "${out_aws_default_region}"
+# ${ envs }
+# }
+#
+# # The "resources" stanza describes the requirements a task needs to
+# # execute. Resource requirements include memory, network, cpu, and more.
+# # This ensures the task will execute on a machine that contains enough
+# # resource capacity.
+# #
+# # https://www.nomadproject.io/docs/job-specification/resources
+# #
+# resources {
+# cpu = ${cpu}
+# memory = ${memory}
+# }
+# }
+# }
}
diff --git a/fdio.infra.terraform/terraform-nomad-pyspark-etl/fdio/variables.tf b/fdio.infra.terraform/terraform-nomad-pyspark-etl/fdio/variables.tf
index 976fbf849e..31c462632b 100644
--- a/fdio.infra.terraform/terraform-nomad-pyspark-etl/fdio/variables.tf
+++ b/fdio.infra.terraform/terraform-nomad-pyspark-etl/fdio/variables.tf
@@ -7,7 +7,7 @@ variable "nomad_acl" {
variable "nomad_provider_address" {
description = "FD.io Nomad cluster address."
type = string
- default = "http://10.32.8.14:4646"
+ default = "http://nomad.service.consul:4646"
}
variable "nomad_provider_ca_file" {
diff --git a/fdio.infra.terraform/terraform-nomad-pyspark-etl/fdio/versions.tf b/fdio.infra.terraform/terraform-nomad-pyspark-etl/fdio/versions.tf
index 526e1d0df0..24e7c94564 100644
--- a/fdio.infra.terraform/terraform-nomad-pyspark-etl/fdio/versions.tf
+++ b/fdio.infra.terraform/terraform-nomad-pyspark-etl/fdio/versions.tf
@@ -1,6 +1,6 @@
terraform {
backend "consul" {
- address = "10.32.8.14:8500"
+ address = "vault.service.consul:8500"
scheme = "http"
path = "terraform/etl"
}
diff --git a/fdio.infra.terraform/terraform-nomad-pyspark-etl/variables.tf b/fdio.infra.terraform/terraform-nomad-pyspark-etl/variables.tf
index 3c6c12a943..9357c096f3 100644
--- a/fdio.infra.terraform/terraform-nomad-pyspark-etl/variables.tf
+++ b/fdio.infra.terraform/terraform-nomad-pyspark-etl/variables.tf
@@ -27,7 +27,7 @@ variable "aws_default_region" {
variable "cpu" {
description = "Specifies the CPU required to run this task in MHz."
type = number
- default = 10000
+ default = 20000
}
variable "cron" {
@@ -57,7 +57,7 @@ variable "job_name" {
variable "memory" {
description = "Specifies the memory required in MB."
type = number
- default = 20000
+ default = 60000
}
variable "out_aws_access_key_id" {