dragonflyoss
/
d7y.io
mirror of https://github.com/dragonflyoss/d7y.io.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

docs: change kubernetes and multi-cluster-kubernetes document (#80) 

* change kubernetes and multi-cluster-kubernetes

Signed-off-by: zhaoxinxin <1186037180@qq.com>

* docs: change kubernetes and multi-cluster-kubernetes

Signed-off-by: zhaoxinxin <1186037180@qq.com>

---------

Signed-off-by: zhaoxinxin <1186037180@qq.com>
This commit is contained in:
Zhaoxinxin 2024-03-18 14:04:23 +08:00 committed by GitHub
parent 18d6c63d77
commit ba1e3fb3b9
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
5 changed files with 293 additions and 161 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

112
docs/getting-started/quick-start/kubernetes.md
View File

@ -5,10 +5,10 @@ description: Kubernetes
slug: /getting-started/quick-start/kubernetes/
---
Documentation for deploying dragonfly on kubernetes using helm.
Documentation for deploying Dragonfly on kubernetes using helm.
You can have a quick start following [Helm Charts](../installation/helm-charts.md).
We recommend to use `Containerd with CRI` and `CRI-O` client.
We recommend to use `containerd with CRI` and `CRI-O` client.
This table describes some container runtimes version and documents.
@ -16,8 +16,8 @@ This table describes some container runtimes version and documents.
| Runtime | Version | Document | CRI Support | Pull Command |
| ----------------------- | ------- | ------------------------------------------------ | ----------- | ------------------------------------------- |
| Containerd<sup>\*</sup> | v1.1.0+ | [Link](../../setup/runtime/containerd/mirror.md) | Yes | crictl pull docker.io/library/alpine:latest |
| Containerd without CRI | v1.1.0 | [Link](../../setup/runtime/containerd/proxy.md) | No | ctr image pull docker.io/library/alpine |
| containerd<sup>\*</sup> | v1.1.0+ | [Link](../../setup/runtime/containerd/mirror.md) | Yes | crictl pull docker.io/library/alpine:latest |
| containerd without CRI | v1.1.0 | [Link](../../setup/runtime/containerd/proxy.md) | No | ctr image pull docker.io/library/alpine |
| CRI-O | All | [Link](../../setup/runtime/cri-o.md) | Yes | crictl pull docker.io/library/alpine:latest |
<!-- markdownlint-restore -->
@ -51,9 +51,9 @@ Switch the context of kubectl to kind cluster:
kubectl config use-context kind-kind
```
## Kind loads dragonfly image {#kind-loads-dragonfly-image}
## Kind loads Dragonfly image {#kind-loads-dragonfly-image}
Pull dragonfly latest images:
Pull Dragonfly latest images:
```shell
docker pull dragonflyoss/scheduler:latest
@ -69,7 +69,7 @@ kind load docker-image dragonflyoss/manager:latest
kind load docker-image dragonflyoss/dfdaemon:latest
```
## Create dragonfly cluster based on helm charts {#create-dragonfly-cluster-based-on-helm-charts}
## Create Dragonfly cluster based on helm charts {#create-dragonfly-cluster-based-on-helm-charts}
Create helm charts configuration file `charts-config.yaml`, configuration content is as follows:
@ -124,7 +124,7 @@ jaeger:
enable: true
```
Create a dragonfly cluster using the configuration file:
Create a Dragonfly cluster using the configuration file:
<!-- markdownlint-disable -->
@ -161,7 +161,7 @@ NOTES:
<!-- markdownlint-restore -->
Check that dragonfly is deployed successfully:
Check that Dragonfly is deployed successfully:
```shell
$ kubectl get po -n dragonfly-system
@ -179,7 +179,7 @@ dragonfly-scheduler-0 1/1 Running 0 8m43s
dragonfly-seed-peer-0 1/1 Running 3 (5m56s ago) 8m43s
```
## Containerd pull image back-to-source for the first time through dragonfly {#containerd-pull-image-back-to-source-for-the-first-time-through-dragonfly}
## Containerd pull image back-to-source for the first time through Dragonfly {#containerd-pull-image-back-to-source-for-the-first-time-through-dragonfly}
Pull `ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5` image in `kind-worker` node:
@ -202,7 +202,53 @@ Tracing details:
![download-back-to-source-tracing](../../resource/getting-started/download-back-to-source-tracing.jpg)
When pull image back-to-source for the first time through dragonfly, it takes `5.58s` to
When pull image back-to-source for the first time through Dragonfly, it takes `5.58s` to
download the `f643e116a03d9604c344edb345d7592c48cc00f2a4848aaf773411f4fb30d2f5` layer.
## Containerd pull image hits the cache of remote peer {#containerd-pull-image-hits-the-cache-of-remote-peer}
Delete the dfdaemon whose Node is `kind-worker` to clear the cache of Dragonfly local Peer.
<!-- markdownlint-disable -->
```shell
# Find pod name.
export POD_NAME=$(kubectl get pods --namespace dragonfly-system -l "app=dragonfly,release=dragonfly,component=dfdaemon" -o=jsonpath='{.items[?(@.spec.nodeName=="kind-worker")].metadata.name}' | head -n 1 )
# Delete pod.
kubectl delete pod ${POD_NAME} -n dragonfly-system
```
<!-- markdownlint-restore -->
Delete `ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5` image in `kind-worker` node:
```shell
docker exec -i kind-worker /usr/local/bin/crictl rmi ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5
```
Pull `ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5` image in `kind-worker` node:
```shell
docker exec -i kind-worker /usr/local/bin/crictl pull ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5
```
Expose jaeger's port `16686`:
```shell
kubectl --namespace dragonfly-system port-forward service/dragonfly-jaeger-query 16686:16686
```
Visit the Jaeger page in [http://127.0.0.1:16686/search](http://127.0.0.1:16686/search), Search for tracing with Tags
`http.url="/v2/dragonflyoss/dragonfly2/scheduler/blobs/sha256:8a9fba45626f402c12bafaadb718690187cae6e5d56296a8fe7d7c4ce19038f7?ns=ghcr.io"`:
![hit-remote-peer-cache-search-tracing](../../resource/getting-started/hit-remote-peer-cache-search-tracing.jpg)
Tracing details:
![hit-remote-peer-cache-tracing](../../resource/getting-started/hit-remote-peer-cache-tracing.jpg)
When pull image hits cache of remote peer, it takes `117.98ms` to
download the `f643e116a03d9604c344edb345d7592c48cc00f2a4848aaf773411f4fb30d2f5` layer.
## Containerd pull image hits the cache of local peer {#containerd-pull-image-hits-the-cache-of-local-peer}
@ -237,32 +283,6 @@ Tracing details:
When pull image hits cache of local peer, it takes `65.24ms` to
download the `f643e116a03d9604c344edb345d7592c48cc00f2a4848aaf773411f4fb30d2f5` layer.
## Containerd pull image hits the cache of remote peer {#containerd-pull-image-hits-the-cache-of-remote-peer}
Pull `ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5` image in `kind-worker2` node:
```shell
docker exec -i kind-worker2 /usr/local/bin/crictl pull ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5
```
Expose jaeger's port `16686`:
```shell
kubectl --namespace dragonfly-system port-forward service/dragonfly-jaeger-query 16686:16686
```
Visit the Jaeger page in [http://127.0.0.1:16686/search](http://127.0.0.1:16686/search), Search for tracing with Tags
`http.url="/v2/dragonflyoss/dragonfly2/scheduler/blobs/sha256:8a9fba45626f402c12bafaadb718690187cae6e5d56296a8fe7d7c4ce19038f7?ns=ghcr.io"`:
![hit-remote-peer-cache-search-tracing](../../resource/getting-started/hit-remote-peer-cache-search-tracing.jpg)
Tracing details:
![hit-remote-peer-cache-tracing](../../resource/getting-started/hit-remote-peer-cache-tracing.jpg)
When pull image hits cache of remote peer, it takes `117.98ms` to
download the `f643e116a03d9604c344edb345d7592c48cc00f2a4848aaf773411f4fb30d2f5` layer.
## Preheat image {#preheat-image}
Expose manager's port `8080`:
@ -271,16 +291,22 @@ Expose manager's port `8080`:
kubectl --namespace dragonfly-system port-forward service/dragonfly-manager 8080:8080
```
Preheat `ghcr.io/dragonflyoss/dragonfly2/manager:v2.0.5` image:
Please create personal access Token before calling Open API, and select `job` for access scopes, refer to [personal-access-tokens](../../reference/personal-access-tokens.md).
Use Open API to preheat the image `ghcr.io/dragonflyoss/dragonfly2/manager:v2.0.5` to Seed Peer, refer to [preheat](../../reference/preheat.md).
```shell
curl --location --request POST 'http://127.0.0.1:8080/api/v1/jobs' \
curl --location --request POST 'http://127.0.0.1:8080/oapi/v1/jobs' \
--header 'Content-Type: application/json' \
--header 'Authorization: Bearer your_personal_access_token' \
--data-raw '{
"type": "preheat",
"args": {
"type": "image",
"url": "https://ghcr.io/v2/dragonflyoss/dragonfly2/manager/manifests/v2.0.5"
"url": "https://ghcr.io/v2/dragonflyoss/dragonfly2/manager/manifests/v2.0.5",
"filteredQueryParams": "Expires&Signature",
"username": "your_registry_username",
"password": "your_registry_password"
}
}'
```
@ -293,8 +319,10 @@ The command-line log returns the preheat job id:
Polling the preheating status with job id:
```bash
curl --request GET 'http://127.0.0.1:8080/api/v1/jobs/1'
```shell
curl --request GET 'http://127.0.0.1:8080/oapi/v1/jobs/1' \
--header 'Content-Type: application/json' \
--header 'Authorization: Bearer your_personal_access_token'
```
If the status is `SUCCESS`, the preheating is successful:

144
docs/getting-started/quick-start/multi-cluster-kubernetes.md
View File

@ -5,12 +5,12 @@ description: Multi-cluster kubernetes
slug: /getting-started/quick-start/multi-cluster-kubernetes/
---
Documentation for deploying dragonfly on multi-cluster kubernetes using helm. A dragonfly cluster manages cluster within
a network. If you have two clusters with disconnected networks, you can use two dragonfly clusters to manage their own clusters.
Documentation for deploying Dragonfly on multi-cluster kubernetes using helm. A Dragonfly cluster manages cluster within
a network. If you have two clusters with disconnected networks, you can use two Dragonfly clusters to manage their own clusters.
The recommended deployment in a multi-cluster kubernetes is to use a dragonfly cluster to manage a kubernetes cluster,
and use a centralized manager service to manage multiple dragonfly clusters. Because peer can only transmit data in
its own dragonfly cluster, if a kubernetes cluster deploys a dragonfly cluster, then a kubernetes cluster forms a p2p network,
The recommended deployment in a multi-cluster kubernetes is to use a Dragonfly cluster to manage a kubernetes cluster,
and use a centralized manager service to manage multiple Dragonfly clusters. Because peer can only transmit data in
its own Dragonfly cluster, if a kubernetes cluster deploys a Dragonfly cluster, then a kubernetes cluster forms a p2p network,
and internal peers can only schedule and transmit data in a kubernetes cluster.
![multi-cluster-kubernetes](../../resource/getting-started/multi-cluster-kubernetes.png)
@ -18,7 +18,7 @@ and internal peers can only schedule and transmit data in a kubernetes cluster.
## Runtime
You can have a quick start following [Helm Charts](../../installation/helm-charts).
We recommend to use `Containerd with CRI` and `CRI-O` client.
We recommend to use `containerd with CRI` and `CRI-O` client.
This table describes some container runtimes version and documents.
@ -26,8 +26,8 @@ This table describes some container runtimes version and documents.
| Runtime | Version | Document | CRI Support | Pull Command |
| ----------------------- | ------- | ------------------------------------------------ | ----------- | ------------------------------------------- |
| Containerd<sup>\*</sup> | v1.1.0+ | [Link](../../setup/runtime/containerd/mirror.md) | Yes | crictl pull docker.io/library/alpine:latest |
| Containerd without CRI | v1.1.0 | [Link](../../setup/runtime/containerd/proxy.md) | No | ctr image pull docker.io/library/alpine |
| containerd<sup>\*</sup> | v1.1.0+ | [Link](../../setup/runtime/containerd/mirror.md) | Yes | crictl pull docker.io/library/alpine:latest |
| containerd without CRI | v1.1.0 | [Link](../../setup/runtime/containerd/proxy.md) | No | ctr image pull docker.io/library/alpine |
| CRI-O | All | [Link](../../setup/runtime/cri-o.md) | Yes | crictl pull docker.io/library/alpine:latest |
<!-- markdownlint-restore -->
@ -74,9 +74,9 @@ Switch the context of kubectl to kind cluster A:
kubectl config use-context kind-kind
```
## Kind loads dragonfly image
## Kind loads Dragonfly image
Pull dragonfly latest images:
Pull Dragonfly latest images:
```shell
docker pull dragonflyoss/scheduler:latest
@ -84,7 +84,7 @@ docker pull dragonflyoss/manager:latest
docker pull dragonflyoss/dfdaemon:latest
```
Kind cluster loads dragonfly latest images:
Kind cluster loads Dragonfly latest images:
```shell
kind load docker-image dragonflyoss/scheduler:latest
@ -92,14 +92,14 @@ kind load docker-image dragonflyoss/manager:latest
kind load docker-image dragonflyoss/dfdaemon:latest
```
## Create dragonfly cluster A
## Create Dragonfly cluster A
Create dragonfly cluster A, the schedulers, seed peers, peers and centralized manager included in
Create Dragonfly cluster A, the schedulers, seed peers, peers and centralized manager included in
the cluster should be installed using helm.
### Create dragonfly cluster A based on helm charts
### Create Dragonfly cluster A based on helm charts
Create dragonfly cluster A charts configuration file `charts-config-cluster-a.yaml`, configuration content is as follows:
Create Dragonfly cluster A charts configuration file `charts-config-cluster-a.yaml`, configuration content is as follows:
```yaml
containerRuntime:
@ -160,7 +160,7 @@ jaeger:
enable: true
```
Create dragonfly cluster A using the configuration file:
Create Dragonfly cluster A using the configuration file:
<!-- markdownlint-disable -->
@ -197,7 +197,7 @@ NOTES:
<!-- markdownlint-restore -->
Check that dragonfly cluster A is deployed successfully:
Check that Dragonfly cluster A is deployed successfully:
```shell
$ kubectl get po -n cluster-a
@ -253,56 +253,56 @@ the username is `root` and password is `dragonfly`.
![clusters](../../resource/getting-started/clusters.png)
By default, Dragonfly will automatically create dragonfly cluster A record in manager when
it is installed for the first time. You can click dragonfly cluster A to view the details.
By default, Dragonfly will automatically create Dragonfly cluster A record in manager when
it is installed for the first time. You can click Dragonfly cluster A to view the details.
![cluster-a](../../resource/getting-started/cluster-a.png)
## Create dragonfly cluster B
## Create Dragonfly cluster B
Create dragonfly cluster B, you need to create a dragonfly cluster record in the manager console first,
and the schedulers, seed peers and peers included in the dragonfly cluster should be installed using helm.
Create Dragonfly cluster B, you need to create a Dragonfly cluster record in the manager console first,
and the schedulers, seed peers and peers included in the Dragonfly cluster should be installed using helm.
### Create dragonfly cluster B in the manager console
### Create Dragonfly cluster B in the manager console
Visit manager console and click the `ADD CLUSTER` button to add dragonfly cluster B record.
Visit manager console and click the `ADD CLUSTER` button to add Dragonfly cluster B record.
Note that the IDC is set to `cluster-2` to match the peer whose IDC is `cluster-2`.
![create-cluster-b](../../resource/getting-started/create-cluster-b.png)
Create dragonfly cluster B record successfully.
Create Dragonfly cluster B record successfully.
![create-cluster-b-successfully](../../resource/getting-started/create-cluster-b-successfully.png)
### Use scopes to distinguish different dragonfly clusters
### Use scopes to distinguish different Dragonfly clusters
The dragonfly cluster needs to serve the scope. It wil provide scheduler services and
seed peer services to peers in the scope. The scopes of the dragonfly cluster are configured
The Dragonfly cluster needs to serve the scope. It wil provide scheduler services and
seed peer services to peers in the scope. The scopes of the Dragonfly cluster are configured
when the console is created and updated. The scopes of the peer are configured in peer YAML config,
the fields are `host.idc`, `host.location` and `host.advertiseIP`,
refer to [dfdaemon config](../../reference/configuration/dfdaemon.md).
If the peer scopes match the dragonfly cluster scopes, then the peer will use
the dragonfly cluster's scheduler and seed peer first, and if there is no matching
dragonfly cluster then use the default dragonfly cluster.
If the peer scopes match the Dragonfly cluster scopes, then the peer will use
the Dragonfly cluster's scheduler and seed peer first, and if there is no matching
Dragonfly cluster then use the default Dragonfly cluster.
**Location**: The dragonfly cluster needs to serve all peers in the location. When the location in
the peer configuration matches the location in the dragonfly cluster, the peer will preferentially
use the scheduler and the seed peer of the dragonfly cluster. It separated by "|",
**Location**: The Dragonfly cluster needs to serve all peers in the location. When the location in
the peer configuration matches the location in the Dragonfly cluster, the peer will preferentially
use the scheduler and the seed peer of the Dragonfly cluster. It separated by "|",
for example "area|country|province|city".
**IDC**: The dragonfly cluster needs to serve all peers in the IDC. When the IDC in the peer
configuration matches the IDC in the dragonfly cluster, the peer will preferentially use the
scheduler and the seed peer of the dragonfly cluster. IDC has higher priority than location
**IDC**: The Dragonfly cluster needs to serve all peers in the IDC. When the IDC in the peer
configuration matches the IDC in the Dragonfly cluster, the peer will preferentially use the
scheduler and the seed peer of the Dragonfly cluster. IDC has higher priority than location
in the scopes.
**CIDRs**: The dragonfly cluster needs to serve all peers in the CIDRs. The advertise IP will be reported in the peer
**CIDRs**: The Dragonfly cluster needs to serve all peers in the CIDRs. The advertise IP will be reported in the peer
configuration when the peer is started, and if the advertise IP is empty in the peer configuration,
peer will automatically get expose IP as advertise IP. When advertise IP of the peer matches the CIDRs in dragonfly cluster,
the peer will preferentially use the scheduler and the seed peer of the dragonfly cluster.
peer will automatically get expose IP as advertise IP. When advertise IP of the peer matches the CIDRs in Dragonfly cluster,
the peer will preferentially use the scheduler and the seed peer of the Dragonfly cluster.
CIDRs has higher priority than IDC in the scopes.
### Create dragonfly cluster B based on helm charts
### Create Dragonfly cluster B based on helm charts
Create charts configuration with cluster information in the manager console.
@ -319,7 +319,7 @@ Create charts configuration with cluster information in the manager console.
- `externalManager.host` is host of the manager GRPC server.
- `externalRedis.addrs[0]` is address of the redis.
Create dragonfly cluster B charts configuration file `charts-config-cluster-b.yaml`,
Create Dragonfly cluster B charts configuration file `charts-config-cluster-b.yaml`,
configuration content is as follows:
```yaml
@ -395,7 +395,7 @@ jaeger:
enable: true
```
Create dragonfly cluster B using the configuration file:
Create Dragonfly cluster B using the configuration file:
<!-- markdownlint-disable -->
@ -431,10 +431,10 @@ NOTES:
<!-- markdownlint-restore -->
Check that dragonfly cluster B is deployed successfully:
Check that Dragonfly cluster B is deployed successfully:
```shell
$ kubectl get po -n dragonfly-system
$ kubectl get po -n cluster-b
NAME READY STATUS RESTARTS AGE
dragonfly-dfdaemon-q8bsg 1/1 Running 0 67s
dragonfly-dfdaemon-tsqls 1/1 Running 0 67s
@ -447,9 +447,9 @@ Create dragonfly cluster B successfully.
![install-cluster-b-successfully](../../resource/getting-started/install-cluster-b-successfully.png)
## Using dragonfly to distribute images for multi-cluster kubernetes
## Using Dragonfly to distribute images for multi-cluster kubernetes
### Containerd pull image back-to-source for the first time through dragonfly in cluster A
### Containerd pull image back-to-source for the first time through Dragonfly in cluster A
Pull `ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5` image in `kind-worker` node:
@ -472,15 +472,35 @@ Tracing details:
![cluster-a-download-back-to-source-tracing](../../resource/getting-started/cluster-a-download-back-to-source-tracing.jpg)
When pull image back-to-source for the first time through dragonfly, peer uses `cluster-a`'s scheduler and seed peer.
When pull image back-to-source for the first time through Dragonfly, peer uses `cluster-a`'s scheduler and seed peer.
It takes `1.47s` to download the `82cbeb56bf8065dfb9ff5a0c6ea212ab3a32f413a137675df59d496e68eaf399` layer.
### Containerd pull image hits the cache of remote peer in cluster A
Pull `ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5` image in `kind-worker2` node:
Delete the dfdaemon whose Node is `kind-worker` to clear the cache of Dragonfly local Peer.
<!-- markdownlint-disable -->
```shell
docker exec -i kind-worker2 /usr/local/bin/crictl pull ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5
# Find pod name.
export POD_NAME=$(kubectl get pods --namespace cluster-a -l "app=dragonfly,release=dragonfly,component=dfdaemon" -o=jsonpath='{.items[?(@.spec.nodeName=="kind-worker")].metadata.name}' | head -n 1 )
# Delete pod.
kubectl delete pod ${POD_NAME} -n cluster-a
```
<!-- markdownlint-restore -->
Delete `ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5` image in `kind-worker` node:
```shell
docker exec -i kind-worker /usr/local/bin/crictl rmi ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5
```
Pull `ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5` image in `kind-worker` node:
```shell
docker exec -i kind-worker /usr/local/bin/crictl pull ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5
```
Expose jaeger's port `16686`:
@ -524,15 +544,33 @@ Tracing details:
![cluster-b-download-back-to-source-tracing](../../resource/getting-started/cluster-b-download-back-to-source-tracing.jpg)
When pull image back-to-source for the first time through dragonfly, peer uses `cluster-b`'s scheduler and seed peer.
When pull image back-to-source for the first time through Dragonfly, peer uses `cluster-b`'s scheduler and seed peer.
It takes `4.97s` to download the `82cbeb56bf8065dfb9ff5a0c6ea212ab3a32f413a137675df59d496e68eaf399` layer.
### Containerd pull image hits the cache of remote peer in cluster B
Pull `ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5` image in `kind-worker4` node:
<!-- markdownlint-disable -->
```shell
docker exec -i kind-worker4 /usr/local/bin/crictl pull ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5
# Find pod name.
export POD_NAME=$(kubectl get pods --namespace cluster-b -l "app=dragonfly,release=dragonfly,component=dfdaemon" -o=jsonpath='{.items[?(@.spec.nodeName=="kind-worker3")].metadata.name}' | head -n 1 )
# Delete pod.
kubectl delete pod ${POD_NAME} -n cluster-b
```
<!-- markdownlint-restore -->
Delete `ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5` image in `kind-worker3` node:
```shell
docker exec -i kind-worker3 /usr/local/bin/crictl rmi ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5
```
Pull `ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5` image in `kind-worker3` node:
```shell
docker exec -i kind-worker3 /usr/local/bin/crictl pull ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5
```
Expose jaeger's port `16686`:

6
i18n/zh/docusaurus-plugin-content-docs/current/getting-started/installation/helm-charts.md
View File

@ -152,8 +152,7 @@ docker exec -i kind-worker /usr/local/bin/crictl pull alpine:3.19
```shell
# 获取 Pod Name
export POD_NAME=$(kubectl get pods --namespace dragonfly-system -l "app=dragonfly,release=dragonfly,component=
dfdaemon" -o=jsonpath='{.items[?(@.spec.nodeName=="kind-worker")].metadata.name}' | head -n 1 )
export POD_NAME=$(kubectl get pods --namespace dragonfly-system -l "app=dragonfly,release=dragonfly,component=dfdaemon" -o=jsonpath='{.items[?(@.spec.nodeName=="kind-worker")].metadata.name}' | head -n 1 )
# 获取 Peer ID
export PEER_ID=$(kubectl -n dragonfly-system exec -it ${POD_NAME} -- grep "alpine" /var/log/dragonfly/
@ -216,8 +215,7 @@ Tracing 详细内容:
```shell
# 获取 Pod Name
export POD_NAME=$(kubectl get pods --namespace dragonfly-system -l "app=dragonfly,release=dragonfly,component=
dfdaemon" -o=jsonpath='{.items[?(@.spec.nodeName=="kind-worker")].metadata.name}' | head -n 1 )
export POD_NAME=$(kubectl get pods --namespace dragonfly-system -l "app=dragonfly,release=dragonfly,component=dfdaemon" -o=jsonpath='{.items[?(@.spec.nodeName=="kind-worker")].metadata.name}' | head -n 1 )
# 删除 Pod
kubectl delete pod ${POD_NAME} -n dragonfly-system

112
i18n/zh/docusaurus-plugin-content-docs/current/getting-started/quick-start/kubernetes.md
View File

@ -9,7 +9,7 @@ slug: /getting-started/quick-start/kubernetes/
您可以根据 [Helm Charts](../installation/helm-charts.md)
文档中的内容快速搭建 Dragonfly 的 Kubernetes 集群。
我们推荐使用 `Containerd with CRI` 和 `CRI-O` 客户端。
我们推荐使用 `containerd with CRI` 和 `CRI-O` 客户端。
下表列出了一些容器的运行时、版本和文档。
@ -17,8 +17,8 @@ slug: /getting-started/quick-start/kubernetes/
| Runtime | Version | Document | CRI Support | Pull Command |
| ----------------------- | ------- | ------------------------------------------------ | ----------- | ------------------------------------------- |
| Containerd<sup>\*</sup> | v1.1.0+ | [Link](../../setup/runtime/containerd/mirror.md) | Yes | crictl pull docker.io/library/alpine:latest |
| Containerd without CRI | v1.1.0 | [Link](../../setup/runtime/containerd/proxy.md) | No | ctr image pull docker.io/library/alpine |
| containerd<sup>\*</sup> | v1.1.0+ | [Link](../../setup/runtime/containerd/mirror.md) | Yes | crictl pull docker.io/library/alpine:latest |
| containerd without CRI | v1.1.0 | [Link](../../setup/runtime/containerd/proxy.md) | No | ctr image pull docker.io/library/alpine |
| CRI-O | All | [Link](../../setup/runtime/cri-o.md) | Yes | crictl pull docker.io/library/alpine:latest |
<!-- markdownlint-restore -->
@ -29,7 +29,7 @@ slug: /getting-started/quick-start/kubernetes/
如果没有可用的 Kubernetes 集群进行测试，推荐使用 [Kind](https://kind.sigs.k8s.io/)。
创建 Kind 多节点集群配置文件 `kind-config.yaml`, 配置如下:
创建 Kind 多节点集群配置文件 `kind-config.yaml`，配置如下:
```yaml
kind: Cluster
@ -72,7 +72,7 @@ kind load docker-image dragonflyoss/dfdaemon:latest
## 基于 Helm Charts 创建 Dragonfly P2P 集群
创建 Helm Charts 配置文件 `charts-config.yaml`, 配置如下:
创建 Helm Charts 配置文件 `charts-config.yaml`，配置如下:
```yaml
containerRuntime:
@ -180,7 +180,7 @@ dragonfly-scheduler-0 1/1 Running 0 8m43s
dragonfly-seed-peer-0 1/1 Running 3 (5m56s ago) 8m43s
```
## Containerd 通过 Dragonfly 首次回源拉镜像
## containerd 通过 Dragonfly 首次回源拉镜像
在 `kind-worker` Node 下载 `ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5` 镜像:
@ -194,7 +194,7 @@ docker exec -i kind-worker /usr/local/bin/crictl pull ghcr.io/dragonflyoss/drago
kubectl --namespace dragonfly-system port-forward service/dragonfly-jaeger-query 16686:16686
```
进入 Jaeger 页面 [http://127.0.0.1:16686/search](http://127.0.0.1:16686/search)， 搜索 Tags 值为
进入 Jaeger 页面 [http://127.0.0.1:16686/search](http://127.0.0.1:16686/search)，搜索 Tags 值为
`http.url="/v2/dragonflyoss/dragonfly2/scheduler/blobs/sha256:8a9fba45626f402c12bafaadb718690187cae6e5d56296a8fe7d7c4ce19038f7?ns=ghcr.io"`
Tracing:
@ -206,9 +206,23 @@ Tracing 详细内容:
集群内首次回源时，下载 `f643e116a03d9604c344edb345d7592c48cc00f2a4848aaf773411f4fb30d2f5` 层需要消耗时间为 `5.58s`
## Containerd 下载镜像命中 Dragonfly 本地 Peer 的缓存
## containerd 下载镜像命中 Dragonfly 远程 Peer 的缓存
删除 `kind-worker` Node 的 Containerd 中镜像 `ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5` 的缓存:
删除 Node 为 `kind-worker` 的 dfdaemon，为了清除 Dragonfly 本地 Peer 的缓存。
<!-- markdownlint-disable -->
```shell
# 获取 Pod Name
export POD_NAME=$(kubectl get pods --namespace dragonfly-system -l "app=dragonfly,release=dragonfly,component=dfdaemon" -o=jsonpath='{.items[?(@.spec.nodeName=="kind-worker")].metadata.name}' | head -n 1 )
# 删除 Pod
kubectl delete pod ${POD_NAME} -n dragonfly-system
```
<!-- markdownlint-restore -->
删除 `kind-worker` Node 的 containerd 中镜像 `ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5` 的缓存:
```shell
docker exec -i kind-worker /usr/local/bin/crictl rmi ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5
@ -226,33 +240,7 @@ docker exec -i kind-worker /usr/local/bin/crictl pull ghcr.io/dragonflyoss/drago
kubectl --namespace dragonfly-system port-forward service/dragonfly-jaeger-query 16686:16686
```
进入 Jaeger 页面 [http://127.0.0.1:16686/search](http://127.0.0.1:16686/search)， 搜索 Tags 值为
`http.url="/v2/dragonflyoss/dragonfly2/scheduler/blobs/sha256:8a9fba45626f402c12bafaadb718690187cae6e5d56296a8fe7d7c4ce19038f7?ns=ghcr.io"`
Tracing:
![hit-local-peer-cache-search-tracing](../../resource/getting-started/hit-local-peer-cache-search-tracing.jpg)
Tracing 详细内容:
![hit-local-peer-cache-tracing](../../resource/getting-started/hit-local-peer-cache-tracing.jpg)
命中本地 Peer 缓存时，下载 `f643e116a03d9604c344edb345d7592c48cc00f2a4848aaf773411f4fb30d2f5` 层需要消耗时间为 `65.24ms`
## Containerd 下载镜像命中 Dragonfly 远程 Peer 的缓存
在 `kind-worker2` Node 下载 `ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5` 镜像:
```shell
docker exec -i kind-worker2 /usr/local/bin/crictl pull ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5
```
暴露 Jaeger `16686` 端口:
```shell
kubectl --namespace dragonfly-system port-forward service/dragonfly-jaeger-query 16686:16686
```
进入 Jaeger 页面 [http://127.0.0.1:16686/search](http://127.0.0.1:16686/search)， 搜索 Tags 值为
进入 Jaeger 页面 [http://127.0.0.1:16686/search](http://127.0.0.1:16686/search)，搜索 Tags 值为
`http.url="/v2/dragonflyoss/dragonfly2/scheduler/blobs/sha256:8a9fba45626f402c12bafaadb718690187cae6e5d56296a8fe7d7c4ce19038f7?ns=ghcr.io"`
Tracing:
@ -264,6 +252,38 @@ Tracing 详细内容:
命中远程 Peer 缓存时，下载 `f643e116a03d9604c344edb345d7592c48cc00f2a4848aaf773411f4fb30d2f5` 层需要消耗时间为 `117.98ms`
## containerd 下载镜像命中 Dragonfly 本地 Peer 的缓存
删除 `kind-worker` Node 的 containerd 中镜像 `ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5` 的缓存:
```shell
docker exec -i kind-worker /usr/local/bin/crictl rmi ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5
```
在 `kind-worker` Node 下载 `ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5` 镜像:
```shell
docker exec -i kind-worker /usr/local/bin/crictl pull ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5
```
暴露 Jaeger `16686` 端口:
```shell
kubectl --namespace dragonfly-system port-forward service/dragonfly-jaeger-query 16686:16686
```
进入 Jaeger 页面 [http://127.0.0.1:16686/search](http://127.0.0.1:16686/search)，搜索 Tags 值为
`http.url="/v2/dragonflyoss/dragonfly2/scheduler/blobs/sha256:8a9fba45626f402c12bafaadb718690187cae6e5d56296a8fe7d7c4ce19038f7?ns=ghcr.io"`
Tracing:
![hit-local-peer-cache-search-tracing](../../resource/getting-started/hit-local-peer-cache-search-tracing.jpg)
Tracing 详细内容:
![hit-local-peer-cache-tracing](../../resource/getting-started/hit-local-peer-cache-tracing.jpg)
命中本地 Peer 缓存时，下载 `f643e116a03d9604c344edb345d7592c48cc00f2a4848aaf773411f4fb30d2f5` 层需要消耗时间为 `65.24ms`
## 预热镜像
暴露 Manager 8080 端口:
@ -272,16 +292,22 @@ Tracing 详细内容:
kubectl --namespace dragonfly-system port-forward service/dragonfly-manager 8080:8080
```
预热镜像 `ghcr.io/dragonflyoss/dragonfly2/manager:v2.0.5`:
使用 Open API 之前请先申请 Personal Access Token，并且 Access Scopes 选择为 `job`，参考文档 [personal-access-tokens](../../reference/personal-access-tokens.md)。
使用 Open API 预热镜像 `ghcr.io/dragonflyoss/dragonfly2/manager:v2.0.5`，参考文档 [preheat](../../reference/preheat.md)。
```shell
curl --location --request POST 'http://127.0.0.1:8080/api/v1/jobs' \
curl --location --request POST 'http://127.0.0.1:8080/oapi/v1/jobs' \
--header 'Content-Type: application/json' \
--header 'Authorization: Bearer your_personal_access_token' \
--data-raw '{
"type": "preheat",
"args": {
"type": "image",
"url": "https://ghcr.io/v2/dragonflyoss/dragonfly2/manager/manifests/v2.0.5"
"url": "https://ghcr.io/v2/dragonflyoss/dragonfly2/manager/manifests/v2.0.5",
"filteredQueryParams": "Expires&Signature",
"username": "your_registry_username",
"password": "your_registry_password"
}
}'
```
@ -294,8 +320,10 @@ curl --location --request POST 'http://127.0.0.1:8080/api/v1/jobs' \
使用预热任务 ID 轮训查询任务是否成功:
```bash
curl --request GET 'http://127.0.0.1:8080/api/v1/jobs/1'
```shell
curl --request GET 'http://127.0.0.1:8080/oapi/v1/jobs/1' \
--header 'Content-Type: application/json' \
--header 'Authorization: Bearer your_personal_access_token'
```
如果返回预热任务状态为 `SUCCESS`，表示预热成功:
@ -316,7 +344,7 @@ docker exec -i kind-worker /usr/local/bin/crictl pull ghcr.io/dragonflyoss/drago
kubectl --namespace dragonfly-system port-forward service/dragonfly-jaeger-query 16686:16686
```
进入 Jaeger 页面 [http://127.0.0.1:16686/search](http://127.0.0.1:16686/search)， 搜索 Tags 值为
进入 Jaeger 页面 [http://127.0.0.1:16686/search](http://127.0.0.1:16686/search)，搜索 Tags 值为
`http.url="/v2/dragonflyoss/dragonfly2/manager/blobs/sha256:ceba1302dd4fbd8fc7fd7a135c8836c795bc3542b9b134597eba13c75d2d2cb0?ns=ghcr.io"`
Tracing:

80
i18n/zh/docusaurus-plugin-content-docs/current/getting-started/quick-start/multi-cluster-kubernetes.md
View File

@ -19,7 +19,7 @@ Peers 只能在当前 Dragonfly 集群内 P2P 传输数据，所以一定要保
您可以根据 [Helm Charts](../installation/helm-charts.md)
文档中的内容快速搭建 Dragonfly 的 Kubernetes 集群。
我们推荐使用 `Containerd with CRI` 和 `CRI-O` 客户端。
我们推荐使用 `containerd with CRI` 和 `CRI-O` 客户端。
下表列出了一些容器的运行时、版本和文档。
@ -27,8 +27,8 @@ Peers 只能在当前 Dragonfly 集群内 P2P 传输数据，所以一定要保
| Runtime | Version | Document | CRI Support | Pull Command |
| ----------------------- | ------- | ------------------------------------------------ | ----------- | ------------------------------------------- |
| Containerd<sup>\*</sup> | v1.1.0+ | [Link](../../setup/runtime/containerd/mirror.md) | Yes | crictl pull docker.io/library/alpine:latest |
| Containerd without CRI | v1.1.0 | [Link](../../setup/runtime/containerd/proxy.md) | No | ctr image pull docker.io/library/alpine |
| containerd<sup>\*</sup> | v1.1.0+ | [Link](../../setup/runtime/containerd/mirror.md) | Yes | crictl pull docker.io/library/alpine:latest |
| containerd without CRI | v1.1.0 | [Link](../../setup/runtime/containerd/proxy.md) | No | ctr image pull docker.io/library/alpine |
| CRI-O | All | [Link](../../setup/runtime/cri-o.md) | Yes | crictl pull docker.io/library/alpine:latest |
<!-- markdownlint-restore -->
@ -39,7 +39,7 @@ Peers 只能在当前 Dragonfly 集群内 P2P 传输数据，所以一定要保
如果没有可用的 Kubernetes 集群进行测试，推荐使用 [Kind](https://kind.sigs.k8s.io/)。
创建 Kind 多节点集群配置文件 `kind-config.yaml`, 配置如下:
创建 Kind 多节点集群配置文件 `kind-config.yaml`，配置如下:
```yaml
kind: Cluster
@ -99,7 +99,7 @@ kind load docker-image dragonflyoss/dfdaemon:latest
### 基于 Helm Charts 创建 Dragonfly 集群 A
创建 Helm Charts 的 Dragonfly 集群 A 的配置文件 `charts-config-cluster-a.yaml`, 配置如下:
创建 Helm Charts 的 Dragonfly 集群 A 的配置文件 `charts-config-cluster-a.yaml`，配置如下:
```yaml
containerRuntime:
@ -245,7 +245,7 @@ kubectl apply -f manager-rest-svc.yaml -n cluster-a
### 访问 Manager 控制台
使用默认用户名 `root`, 密码 `dragonfly` 访问 `localhost:8080` 的 Manager 控制台地址，并且进入控制台。
使用默认用户名 `root`，密码 `dragonfly` 访问 `localhost:8080` 的 Manager 控制台地址，并且进入控制台。
![signin](../../resource/getting-started/signin.png)
@ -309,7 +309,7 @@ CIDR 在 Scopes 内的优先级高于 IDC。
- `externalManager.host` 是 Manager 的 GRPC 服务的 Host。
- `externalRedis.addrs[0]` 是 Redis 的服务地址。
创建 Helm Charts 的 Dragonfly 集群 B 的配置文件 `charts-config-cluster-b.yaml`, 配置如下:
创建 Helm Charts 的 Dragonfly 集群 B 的配置文件 `charts-config-cluster-b.yaml`，配置如下:
```yaml
containerRuntime:
@ -423,7 +423,7 @@ NOTES:
检查 Dragonfly 集群 B 是否部署成功:
```shell
$ kubectl get po -n dragonfly-system
$ kubectl get po -n cluster-b
NAME READY STATUS RESTARTS AGE
dragonfly-dfdaemon-q8bsg 1/1 Running 0 67s
dragonfly-dfdaemon-tsqls 1/1 Running 0 67s
@ -438,7 +438,7 @@ dragonfly-seed-peer-0 1/1 Running 0 67s
## 使用 Dragonfly 在多集群环境下分发镜像
### 集群 A 中 Containerd 通过 Dragonfly 首次回源拉镜像
### 集群 A 中 containerd 通过 Dragonfly 首次回源拉镜像
在 `kind-worker` Node 下载 `ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5` 镜像:
@ -452,7 +452,7 @@ docker exec -i kind-worker /usr/local/bin/crictl pull ghcr.io/dragonflyoss/drago
kubectl --namespace cluster-a port-forward service/dragonfly-jaeger-query 16686:16686
```
进入 Jaeger 页面 [http://127.0.0.1:16686/search](http://127.0.0.1:16686/search)， 搜索 Tags 值为
进入 Jaeger 页面 [http://127.0.0.1:16686/search](http://127.0.0.1:16686/search)，搜索 Tags 值为
`http.url="/v2/dragonflyoss/dragonfly2/scheduler/blobs/sha256:82cbeb56bf8065dfb9ff5a0c6ea212ab3a32f413a137675df59d496e68eaf399?ns=ghcr.io"`
Tracing:
@ -464,12 +464,32 @@ Tracing 详细内容:
集群 A 内首次回源时，下载 `82cbeb56bf8065dfb9ff5a0c6ea212ab3a32f413a137675df59d496e68eaf399` 层需要消耗时间为 `1.47s`。
### 集群 A 中 Containerd 下载镜像命中 Dragonfly 远程 Peer 的缓存
### 集群 A 中 containerd 下载镜像命中 Dragonfly 远程 Peer 的缓存
在 `kind-worker2` Node 下载 `ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5` 镜像:
删除 Node 为 `kind-worker` 的 dfdaemon，为了清除 Dragonfly 本地 Peer 的缓存。
<!-- markdownlint-disable -->
```shell
docker exec -i kind-worker2 /usr/local/bin/crictl pull ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5
# 获取 Pod Name
export POD_NAME=$(kubectl get pods --namespace cluster-a -l "app=dragonfly,release=dragonfly,component=dfdaemon" -o=jsonpath='{.items[?(@.spec.nodeName=="kind-worker")].metadata.name}' | head -n 1 )
# 删除 Pod
kubectl delete pod ${POD_NAME} -n cluster-a
```
<!-- markdownlint-restore -->
删除 `kind-worker` Node 的 containerd 中镜像 `ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5` 的缓存:
```shell
docker exec -i kind-worker /usr/local/bin/crictl rmi ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5
```
在 `kind-worker` Node 下载 `ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5` 镜像:
```shell
docker exec -i kind-worker /usr/local/bin/crictl pull ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5
```
暴露 Jaeger `16686` 端口:
@ -478,7 +498,7 @@ docker exec -i kind-worker2 /usr/local/bin/crictl pull ghcr.io/dragonflyoss/drag
kubectl --namespace cluster-a port-forward service/dragonfly-jaeger-query 16686:16686
```
进入 Jaeger 页面 [http://127.0.0.1:16686/search](http://127.0.0.1:16686/search)， 搜索 Tags 值为
进入 Jaeger 页面 [http://127.0.0.1:16686/search](http://127.0.0.1:16686/search)，搜索 Tags 值为
`http.url="/v2/dragonflyoss/dragonfly2/scheduler/blobs/sha256:82cbeb56bf8065dfb9ff5a0c6ea212ab3a32f413a137675df59d496e68eaf399?ns=ghcr.io"`
Tracing:
@ -490,7 +510,7 @@ Tracing 详细内容:
集群 A 中命中远程 Peer 缓存时，下载 `82cbeb56bf8065dfb9ff5a0c6ea212ab3a32f413a137675df59d496e68eaf399` 层需要消耗时间为 `37.48ms`。
### 集群 B 中 Containerd 通过 Dragonfly 首次回源拉镜像
### 集群 B 中 containerd 通过 Dragonfly 首次回源拉镜像
在 `kind-worker3` Node 下载 `ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5` 镜像:
@ -504,7 +524,7 @@ docker exec -i kind-worker3 /usr/local/bin/crictl pull ghcr.io/dragonflyoss/drag
kubectl --namespace cluster-b port-forward service/dragonfly-jaeger-query 16686:16686
```
进入 Jaeger 页面 [http://127.0.0.1:16686/search](http://127.0.0.1:16686/search)， 搜索 Tags 值为
进入 Jaeger 页面 [http://127.0.0.1:16686/search](http://127.0.0.1:16686/search)，搜索 Tags 值为
`http.url="/v2/dragonflyoss/dragonfly2/scheduler/blobs/sha256:82cbeb56bf8065dfb9ff5a0c6ea212ab3a32f413a137675df59d496e68eaf399?ns=ghcr.io"`
Tracing:
@ -516,12 +536,32 @@ Tracing 详细内容:
集群 B 中命中远程 Peer 缓存时，下载 `82cbeb56bf8065dfb9ff5a0c6ea212ab3a32f413a137675df59d496e68eaf399` 层需要消耗时间为 `4.97s`。
### 集群 B 中 Containerd 下载镜像命中 Dragonfly 远程 Peer 的缓存
### 集群 B 中 containerd 下载镜像命中 Dragonfly 远程 Peer 的缓存
在 `kind-worker4` Node 下载 `ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5` 镜像:
删除 Node 为 `kind-worker3` 的 dfdaemon，为了清除 Dragonfly 本地 Peer 的缓存。
<!-- markdownlint-disable -->
```shell
docker exec -i kind-worker4 /usr/local/bin/crictl pull ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5
# 获取 Pod Name
export POD_NAME=$(kubectl get pods --namespace cluster-b -l "app=dragonfly,release=dragonfly,component=dfdaemon" -o=jsonpath='{.items[?(@.spec.nodeName=="kind-worker3")].metadata.name}' | head -n 1 )
# 删除 Pod
kubectl delete pod ${POD_NAME} -n cluster-b
```
<!-- markdownlint-restore -->
删除 `kind-worker3` Node 的 containerd 中镜像 `ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5` 的缓存:
```shell
docker exec -i kind-worker3 /usr/local/bin/crictl rmi ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5
```
在 `kind-worker3` Node 下载 `ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5` 镜像:
```shell
docker exec -i kind-worker3 /usr/local/bin/crictl pull ghcr.io/dragonflyoss/dragonfly2/scheduler:v2.0.5
```
暴露 Jaeger `16686` 端口:
@ -530,7 +570,7 @@ docker exec -i kind-worker4 /usr/local/bin/crictl pull ghcr.io/dragonflyoss/drag
kubectl --namespace cluster-b port-forward service/dragonfly-jaeger-query 16686:16686
```
进入 Jaeger 页面 [http://127.0.0.1:16686/search](http://127.0.0.1:16686/search)， 搜索 Tags 值为
进入 Jaeger 页面 [http://127.0.0.1:16686/search](http://127.0.0.1:16686/search)，搜索 Tags 值为
`http.url="/v2/dragonflyoss/dragonfly2/scheduler/blobs/sha256:82cbeb56bf8065dfb9ff5a0c6ea212ab3a32f413a137675df59d496e68eaf399?ns=ghcr.io"`
Tracing: