Skip to content

[History server] Update history server examples and defaults #5080

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
chiayi wants to merge 14 commits into
base: master
Choose a base branch
from
Open

[History server] Update history server examples and defaults #5080

Show file tree
Hide file tree
Changes from 8 commits
Commits
Show all changes
14 commits
Select commit Hold shift + click to select a range
7895b02
Update history server examples
chiayi Aug 4, 2026
fe35cd5
update raycluster ray version for history server samples
chiayi Aug 5, 2026
65999ff
fix ci
chiayi Aug 5, 2026
4ee49e6
update ray version
chiayi Aug 5, 2026
e1501f1
Update historyserver and convery raycluster-gcs to rayjob-gcs
chiayi Aug 5, 2026
6e5efb1
Added 2.56 updated expected grafana health response
chiayi Aug 5, 2026
b308655
rebase and update
chiayi Aug 6, 2026
bb1e2a7
reorder env var
chiayi Aug 6, 2026
c9f2fd5
add back raycluster-gcs example to not break doc
chiayi Aug 6, 2026
2c63b00
Update storage root env var value to make it clear what it is for
chiayi Aug 7, 2026
aa941ee
Update gcs yaml
chiayi Aug 7, 2026
f983057
remove unused configuration
chiayi Aug 7, 2026
bda7aeb
Updated env var for event exporter
chiayi Aug 7, 2026
9ea2bde
remove storage root and use bucket as root dir
chiayi Aug 7, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion historyserver/DEVELOPMENT.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -120,7 +120,7 @@ Install Ray locally. Make sure to use at least Ray `v2.55`.

```bash
pip uninstall -y ray
pip install -U "ray[default]==2.55.0"
pip install -U "ray[default]==2.56.0"
```

Run the `ray start` command:
Expand Down
4 changes: 2 additions & 2 deletions historyserver/cmd/collector/main.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -30,7 +30,7 @@ func main() {
rayClusterNamespace := ""
rayRootDir := ""
logBatching := 1000
eventsPort := 8080
eventsPort := 8084
pushInterval := time.Minute
ownerKind := ""
ownerName := ""
Expand All @@ -53,7 +53,7 @@ func main() {
flag.StringVar(&rayClusterNamespace, "ray-cluster-namespace", "default", "")
flag.StringVar(&rayRootDir, "ray-root-dir", "", "")
flag.IntVar(&logBatching, "log-batching", 1000, "")
flag.IntVar(&eventsPort, "events-port", 8080, "")
flag.IntVar(&eventsPort, "events-port", 8084, "")
flag.StringVar(&runtimeClassConfigPath, "runtime-class-config-path", "", "")
flag.DurationVar(&pushInterval, "push-interval", time.Minute, "")
flag.StringVar(&ownerKind, "owner-kind", "", "")
Expand Down
4 changes: 4 additions & 0 deletions historyserver/cmd/historyserver/main.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -54,6 +54,10 @@ func main() {
}
runtimeClassName = strings.ToLower(runtimeClassName)

if val := os.Getenv("RAY_ROOT_DIR"); val != "" {
rayRootDir = val
}

if qps <= 0 {
logrus.Fatalf("--kube-api-qps must be > 0, got %v", qps)
}
Expand Down
13 changes: 7 additions & 6 deletions historyserver/config/historyserver-gcs.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,4 +1,5 @@
# Ensure that the env var GCS_BUCKET is set before applying
# Apply with envsubst to substitute environment variables:
# envsubst < historyserver/config/historyserver-gcs.yaml | kubectl apply -f -
apiVersion: v1
kind: Service
metadata:
Expand Down Expand Up @@ -37,12 +38,12 @@ spec:
env:
- name: GCS_BUCKET
value: "${GCS_BUCKET}"
image: ${HISTORY_SERVER_IMAGE}
- name: STORAGE_BACKEND
value: "gcs"
- name: RAY_ROOT_DIR
value: "root"

@andrewsykim andrewsykim Aug 6, 2026

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I don't think "root" will work either? Is RAY_ROOT_DIR referring to Ray's root directory or the root directory used in storage? If it's storage, this variable name is confusing.

@chiayi chiayi Aug 6, 2026

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It's referring to the root directory in storage. So maybe STORAGE_ROOT_DIR instead?

@andrewsykim andrewsykim Aug 7, 2026

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

yeah I would rename to STORAGE_ROOT_DIR. Also when it's empty it should just use the specified bucket's root dir

@andrewsykim andrewsykim Aug 7, 2026

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@chiayi please fix this in a fast follow-up

image: historyserver:v0.1.0
imagePullPolicy: IfNotPresent
command:
- historyserver
- --runtime-class-name=gcs
- --ray-root-dir=log
ports:
- containerPort: 8080
resources:
Expand Down
13 changes: 4 additions & 9 deletions historyserver/config/historyserver.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -50,17 +50,12 @@ spec:
value: "test"
- name: S3FORCE_PATH_STYLE
value: "true"
- name: STORAGE_BACKEND
value: "s3"
- name: RAY_ROOT_DIR
value: "log"
image: historyserver:v0.1.0
imagePullPolicy: IfNotPresent
command:
- historyserver
- --runtime-class-name=s3
- --ray-root-dir=log
# Enable proxying to live RayClusters with token authentication enabled.
# The history server reads the auth Secret and forwards requests with the
# appropriate authorization header. Requires the RBAC in
# service_account_auth_token_mode.yaml.
# - --use-auth-token-mode=true
Comment thread
machichima marked this conversation as resolved.
ports:
- containerPort: 8080
resources:
Expand Down
46 changes: 34 additions & 12 deletions historyserver/config/raycluster-azureblob.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,10 +29,8 @@ spec:
- name: RAY_DASHBOARD_AGGREGATOR_AGENT_EVENTS_EXPORT_ADDR
value: "http://localhost:8084/v1/events"
- name: RAY_DASHBOARD_AGGREGATOR_AGENT_EXPOSABLE_EVENT_TYPES
value: "TASK_DEFINITION_EVENT,TASK_LIFECYCLE_EVENT,ACTOR_TASK_DEFINITION_EVENT,
TASK_PROFILE_EVENT,DRIVER_JOB_DEFINITION_EVENT,DRIVER_JOB_LIFECYCLE_EVENT,
ACTOR_DEFINITION_EVENT,ACTOR_LIFECYCLE_EVENT,NODE_DEFINITION_EVENT,NODE_LIFECYCLE_EVENT"
image: rayproject/ray:2.52.0
value: "ALL"
image: rayproject/ray:2.56.0
imagePullPolicy: IfNotPresent
securityContext:
allowPrivilegeEscalation: true
Expand All @@ -52,20 +50,33 @@ spec:
image: collector:v0.1.0
imagePullPolicy: IfNotPresent

@machichima machichima Aug 5, 2026

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

If we use nightly, I think we need to set imagePullPolicy: Always, otherwise it will never pull the newest image. Also for other files.
Actually I think we can come back and update this after release to pin the specific version

env:
- name: RAY_CLUSTER_NAME
valueFrom:
fieldRef:
fieldPath: metadata.labels['ray.io/cluster']
- name: RAY_CLUSTER_NAMESPACE
valueFrom:
fieldRef:
fieldPath: metadata.namespace
- name: POD_IP
valueFrom:
fieldRef:
fieldPath: status.podIP
- name: FQ_RAY_IP
value: raycluster-historyserver-head-svc.default.svc.cluster.local
value: $(RAY_CLUSTER_NAME)-head-svc.$(RAY_CLUSTER_NAMESPACE).svc.cluster.local
- name: RAY_TMP_ROOT
value: *rayTmpRoot
- name: RAY_ROLE
value: "Head"
- name: STORAGE_BACKEND
value: "azureblob"
- name: RAY_ROOT_DIR
value: "log"
# reference: https://learn.microsoft.com/en-us/azure/storage/common/storage-use-azurite#connect-to-the-emulator-by-using-the-azure-storage-explorer
- name: AZURE_STORAGE_CONNECTION_STRING
value: "DefaultEndpointsProtocol=http;AccountName=devstoreaccount1;AccountKey=Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==;BlobEndpoint=http://azurite-service.azurite-dev.svc.cluster.local:10000/devstoreaccount1;"
- name: AZURE_STORAGE_CONTAINER
value: ray-historyserver
command: [collector, --role=Head, --runtime-class-name=azureblob, --ray-cluster-name=raycluster-historyserver, --ray-root-dir=log, --events-port=8084]
volumeMounts:
- name: historyserver
mountPath: *rayTmpRoot
Expand Down Expand Up @@ -100,10 +111,8 @@ spec:
- name: RAY_DASHBOARD_AGGREGATOR_AGENT_EVENTS_EXPORT_ADDR
value: "http://localhost:8084/v1/events"
- name: RAY_DASHBOARD_AGGREGATOR_AGENT_EXPOSABLE_EVENT_TYPES
value: "TASK_DEFINITION_EVENT,TASK_LIFECYCLE_EVENT,ACTOR_TASK_DEFINITION_EVENT,
TASK_PROFILE_EVENT,DRIVER_JOB_DEFINITION_EVENT,DRIVER_JOB_LIFECYCLE_EVENT,
ACTOR_DEFINITION_EVENT,ACTOR_LIFECYCLE_EVENT,NODE_DEFINITION_EVENT,NODE_LIFECYCLE_EVENT"
image: rayproject/ray:2.52.0
value: "ALL"
image: rayproject/ray:2.56.0
imagePullPolicy: IfNotPresent
name: ray-worker
securityContext:
Expand All @@ -123,20 +132,33 @@ spec:
image: collector:v0.1.0
imagePullPolicy: IfNotPresent
env:
- name: RAY_CLUSTER_NAME
valueFrom:
fieldRef:
fieldPath: metadata.labels['ray.io/cluster']
- name: RAY_CLUSTER_NAMESPACE
valueFrom:
fieldRef:
fieldPath: metadata.namespace
- name: POD_IP
valueFrom:
fieldRef:
fieldPath: status.podIP
- name: FQ_RAY_IP
value: raycluster-historyserver-head-svc.default.svc.cluster.local
value: $(RAY_CLUSTER_NAME)-head-svc.$(RAY_CLUSTER_NAMESPACE).svc.cluster.local
- name: RAY_TMP_ROOT
value: *rayTmpRoot
- name: RAY_ROLE
value: "Worker"
- name: STORAGE_BACKEND
value: "azureblob"
- name: RAY_ROOT_DIR
value: "log"
# reference: https://learn.microsoft.com/en-us/azure/storage/common/storage-use-azurite#connect-to-the-emulator-by-using-the-azure-storage-explorer
- name: AZURE_STORAGE_CONNECTION_STRING
value: "DefaultEndpointsProtocol=http;AccountName=devstoreaccount1;AccountKey=Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==;BlobEndpoint=http://azurite-service.azurite-dev.svc.cluster.local:10000/devstoreaccount1;"
- name: AZURE_STORAGE_CONTAINER
value: ray-historyserver
command: [collector, --role=Worker, --runtime-class-name=azureblob, --ray-cluster-name=raycluster-historyserver, --ray-root-dir=log, --events-port=8084]
volumeMounts:
- name: historyserver
mountPath: *rayTmpRoot
Expand Down
177 changes: 0 additions & 177 deletions historyserver/config/raycluster-gcs.yaml
View file
Open in desktop

This file was deleted.

Loading
Loading