xuri/aurora

Cross-platform beanstalkd queue server admin console.

aurora's Issues

Do not force config path to be relative to binary location

If I invoke aurora as follows:

/path/to/bin/aurora -c /etc/aurora.toml

I expect it to read the config from /etc/aurora.toml; however, it instead tries to read from /path/to/bin/etc/aurora.toml (that is to say, it always interprets the config path to be relative to the directory the executable is in).

how to install aurora

Is there any document,how to install aurora

How to use in linux system

As the title, I'v tried "sh aurora", but nothing happens...

Support setting ttr in samples

Hi,

It seems like inserting new job sample always setts ttr to 60 secs.
Would be useful if user can be allowed to set it as part of a sample data.

Thanks

Running Aurora on boot

Hi,

Using an Ubuntu Server to run beanstalkd and aurora - do you have any example scripts/details about writing an init.d script for aurora?

Thanks for the awesome tool!

Broken Commit - HTML Rendering

Description

The commit a935e08 breaks HTML rendering.

Steps to reproduce the issue:

1. go build/go get
2. ./aurora
3. Open the app, click on a tube

Describe the results you received:

I receive escaped HTML output, (&nbsp, < etc)

Describe the results you expected:

A pretty rendered page ;)

Output of go version:

go version go1.13.8 linux/amd64

Aurora version or commit ID:

a935e08f81115dec77ef516a7b85ca72b0a43675

Environment details (OS, physical, etc.):

DISTRIB_ID=Ubuntu
DISTRIB_RELEASE=20.04
DISTRIB_CODENAME=focal
DISTRIB_DESCRIPTION="Ubuntu 20.04.2 LTS"

Remarks / Fix:

I revered back to the commit before that one, and then it works.

Feature Request - SSO

It would be nice to see something beyond basic auth like an SSO plugin for G Suite or Git.

Search this tube not works if peeked job has the highest JID

Description

The Search this tube is not working, if the peeked job from the Tube has the highest JID - was the last one added with a higher priority than other jobs.

Steps to reproduce the issue:

1. create a new tube
2. insert 5 same tasks with normal priority, insert 1 same task with higher priority (priority number is lower)
3. search for the common text in all 6 tasks
4. you will see only 1 result, instead of 6 results

The problematic is the code here and here:

You should not start with the JID of the peeked job, but probably you should try to find the lowest JID or start from the 0. Or just peek all ready jobs, without releasing them.

Describe the results you received:
1 result returned for the search

Describe the results you expected:
6 results, of each task in the queue

Output of go version:

beanstalkd 1.10
aurora version 2.1

no go installed on the machine

Environment details (OS, physical, etc.):

Linux Debian 8, virtual machine, but not relevant.

Custom name for the server

Feature request - would be possible to set up the custom name for the server instead of displaying hostname:port (in most cases it is only ip:port)?

I would appreciate showing the server name in the list of tubes, and everywhere possible.

I could prepare a PR with the change, if you do not have the time for this.

How to use Aurora on Debian server

Hi,

I tried use Aurora on my work machine, but when I created the link, is doesn't work.

Follow my aurora.toml:

servers = ["67.205.165.199:11300"]
listen = "0.0.0.0:3333"
version = 2.1

[openpage]
enabled = false

[auth]
enabled = true
password = "password"
username = "admin"

[sample]
storage = "{}"

Regards,

Feature Request: Kick ALL jobs burried

It would be convenient to have a button that will allow a user to kick all of the jobs that were buried. Sometimes, they might pile up into the thousands because of a communication outage or some other temporary issue. Once resolved, it would be great to just kick all of the buried jobs vs. 10 at a time.

Or, offer a count selector: 1, 10, 100, All.
Or an input with a default value of 10.

Allow bypassing openPage()

On launch, aurora attempts to open a browser in the openPage function. This really only makes sense in testing, and shouldn't be attempted by default.

This also doesn't work if you're binding to a wildcard (e.g., to 0.0.0.0 or [::])

Could I use this directly on a server?

I want to run it on a server, and access it from my own computer. But it seems this acquires the browser to run. Since I do not have a Xwindow on the server, is it possible to run it on server?

REST API

Is your feature request related to a problem? Please describe.
REST API to extract tube/ system status to other custom apps.

Setting up auth section doesn't actually allow login

Upon hitting the page, it prompts for login but the username and password don't actually log me in. Authentication fails.

Has anyone actually tested authentication? Or am I missing a crucial understanding of how to use a toml to configure the app to run? There's no documentation anywhere showing how the toml config works or what it should look like. Which I find quite surprising since the home page looks so nice.

servers = ["192.168.1.1:11300"]
listen = "127.0.0.1:11400"
version = 2.1

[openpage]
enabled = false

[auth]
enabled = true
password = "myadmin"
username = "secrets"

[sample]
storage = "{}"

Feature request: permissions

Is your feature request related to a problem? Please describe.

Would be nice to have a permissions model.

Describe the solution you'd like

If there was the concept of read-only users and admin users you could allow some users to view stats but not do anything destructive like delete jobs.

Describe alternatives you've considered

None.

Additional context

It would be even better if the permissions could be tied to SSO groups (as mentioned in #9). Then you could manage the users/groups externally to Aurora and have a mapping from SSO group to permission.

No documentation on how to configure with a toml file

I can see the example toma file in the repo. I don't see any examples of how to set the values which leads to trial and error guess work. Please include a detailed writeup on how to configure the toml exactly as the innards would expect them to be.

Reorder messages

Is your feature request related to a problem? Please describe.
Is there a way to reorder messages? I'd love to be able to move some stuff that's lower in the queue up higher.

Describe the solution you'd like
In the search page, under actions, if we could add a "send to front" or a "send to back" or something similar.

Describe alternatives you've considered
No alternative that I can think of other than wait out the queue.

Additional context
I'm using beanstalkd to keep track of jobs that take a while to run (10+min each to run). Sometimes I want jobs that are lower to finish first but as far as I can tell there isn't any way to to this yet.

Unknown Brew Formulae

I was unable to find the brew formulae for the project. Can you please list in the readme?

Feature Request - Subdirectory

hi, if use nginx and proxy to app , many errors to url '/' , redirect to '/', all must be './'
plz fix this.

Update https://github.com/beanstalkd/go-beanstalk to fix indefinite client hang after connection reset

Description

We appear to have run into the issue fixed by beanstalkd/go-beanstalk#29. A job that had been created days earlier was never processed until a restart forced the connection to be re-established.

Ideally this module would update the go-beanstalk module to include the above fix.

Steps to reproduce the issue:
Unclear how the connection gets in this state to trigger the underlying issue in the go-beanstalk module.

Describe the results you received:
Job stuck indefinitely until a restart forced the connection to be re-established

Describe the results you expected:
Job should be picked up via ready state

Output of go version:

go version go1.13 linux/amd64

Aurora version or commit ID:

commit ID: 8024f7a

Environment details (OS, physical, etc.):

Ubuntu 16.04

Sort tube names?

It might be easier to read if the tube names were alphabetically (or otherwise consistently) sorted.