vesoft-inc / nebula-dashboard Goto Github PK

Page refresh does not show

Is your feature request related to a problem? Please describe.
We have merics data visualization, it's conveniet for user, they want to get notified when some metrics are not right

Describe the solution you'd like
We can simple combine this with our baseline。

Is your feature request related to a problem? Please describe.
Currently, not all metrics are shown on the dashboard, e.g., getNeighborsLatency is not available under SUM. If I am a new user, I would think there is no such metric. Is it better to show all the available metrics? If some aggregation method does not fit for a particular metric, maybe you can just leave the data blank.

I can not confirm the num_heartbeats is right, please help me check it .

There are 3 meta nodes, but only one can get num_heartbeats.

As title, which could be a FAQ.

Additional context
https://discuss.nebula-graph.com.cn/t/topic/6329

This page can only hold at most 6 metrics at the same time:

Can we make this page slippery to show more metrics at the same time?

社区版兼容性方案改造

vesoft-inc/nebula#3098

https://hub.docker.com/r/vesoft/nebula-dashboard to avoid confusion to the user?

Or we will provide one in the future?

1. support machine monitoring, includes cpu/memory/load/disk
2. support customize monitoring metrics.
3. get a cool ui
4. support service metrics

want to ask do you have any bug bounty program or do your team provide bounty for reporting critical issues? because i found 1 critical issue

Thank you

We should leverage systemd or equivalent to have control of the dashboard processes, at least it should be auto-started after process failures.

To get rid of nohup.

I will port what had been done in https://github.com/wey-gu/nebula-up here

Before that you could try curl -fsSL nebula-up.siwei.io/all-in-one.sh | bash

or

curl -fsSL nebula-up.siwei.io/all-in-one.sh | bash -s -- v3 dashboard

To play with it!

tasks

• Create docker-compose files first relying on docker file to be built, this decouples the task of docker image build/push to docker hub before it's done. #65
• Create GH actions to build and push images to the docker hub
• Refer GH images in compose file
• Documentations to be added

Describe the bug (must be provided)
nebula dashboard only support http://ip:port pattern for work
A clear and concise description of what the bug is.

Your Environments (must be provided)

• OS: Linux
• Node-version: node v16.9.1
• Dashboard-version: v2.5.0

How To Reproduce(must be provided)

step1:
bind a domain like dashboard.xxx.com to ip:port
step2:
view dashboard.xxx.com/ would report
nodejs.ECONNRESETError:
read ECONNRESET
at TCP.onStreamRead (node:internal/stream_base_commons:220:20)
errno: -104
code: "ECONNRESET"
step3:
because this proxy file produce some route problem when view by domain url

xxx/nebula-graph-dashboard/nebula-graph-dashboard/app/middleware/proxy.js

Additional context

Provide logs and configs, or any other context to trace the problem.

Monitoring update in both enterprise edition and community edition(compare with grafana)

Is your feature request related to a problem? Please describe.
After building Dashboard, there are some slow queries/ error queries show up. i want to locate at which services or which queries happened, but there are no any method as i known to debug.

Describe the solution you'd like
currently, no

Describe alternatives you've considered
currently no

Additional context
no

Describe the bug (must be provided)

As title

Your Environments (must be provided)

• OS: macOS Intel
• Node-version: 14
• Dashboard-version: v1.0.2

How To Reproduce(must be provided)

Steps to reproduce the behavior:

1. Step 1, login
2. Step 2, click Long-Term task

Expected behavior

No error

Additional context

Nebula core in 2.6.0

如题，已经开启图空间监控，并且机器、管理两个功能是正常的。

1:

Here, "@base" is not defined, should be "@/" instead
2:

There should be another ")" here.
3: (not fixed yet)

However, when I have fixed above two little bugs which the webpack indicates directly , I still cann't "npm run dev" the project,
it told me that "ERROR in Conflict: Multiple assets emit different content to the same filename index.html".
4: (not fixed yet)
I first try to use pnpm to install the node_modules, but it turns out "it can't find some modules". I try yarn latter and find out that "can't find modules" problem disappears.

My Environments

• OS: Windows & Linux
• Node-version: node: 16.15.0
• Dashboard-version: master

It can only create a cluster in the default path now, maybe we can consider support choosing the path or using the default path when creating the cluster, it may be better.

How to get the disk monitor data in your code，i want to add some features in the dashboard.

as title :)

There are too many steps or components when deploy the dashboard, some setup steps can be combined into a shell scripts.

8

ref: https://github.com/vesoft-inc/nebula-dashboard/tree/master/vendors

Our use is asking if we can provide all the Promql we are using for node level metrics, and the corresponding raw metrics from node_exporter. The reason of the latter is because they will whitelist those raw metrics and do not need to send everything to grafana.