Skip to content
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.

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

Kill handler logging #206

Open
DSsoto opened this issue Apr 3, 2017 · 15 comments
Open

Kill handler logging #206

DSsoto opened this issue Apr 3, 2017 · 15 comments

Comments

@DSsoto
Copy link
Member

DSsoto commented Apr 3, 2017

Make the alarm handler for the kill alarm log the time and reason for any kills to a file.

@mattlangford
Copy link
Member

Right now it logs to ros logs right? This would just need to be making sure there are enough ros log prints in the server or client nodes. I thought it logs whenever new alarms are received currently.

@DSsoto
Copy link
Member Author

DSsoto commented Apr 3, 2017

You're right about that, but ros logs have way too much info to be useful. I think the vehicle being killed is special enough that a warrants an easy to read log of when and why.

@mattlangford
Copy link
Member

If I'm not mistaken you can get specific logs from specific nodes. The alarm server node would probably not have too much garbage right? I haven't looked too much into ros logging but I imagine getting a specific nodes output would be something that's possible.

@DSsoto
Copy link
Member Author

DSsoto commented Apr 3, 2017

Then we'd make a script that makes getting that specific log easier, the implementation is besides the point.

@duggerdufl
Copy link
Member

We need to do something about the DVL beam x not returned messages, it's impossible to see anything out of the water on that window because of all those messages.

@DSsoto
Copy link
Member Author

DSsoto commented Apr 4, 2017

@whispercoros If we make a gui and this is on it, no need to log it.

@duggerdufl
Copy link
Member

Until you kill ROS or whatever and it goes away.

@mattlangford
Copy link
Member

Rospy doesn't have a throttled logger like roscpp does (last time I checked), it'd be pretty easy to implement one in that file. I can do it unless David wants to.

@DSsoto
Copy link
Member Author

DSsoto commented Apr 4, 2017

@duggerd the logs do persist even after all ros stuff is killed. I'll look at how the ones for the alarm server node look and if they look usable then I'll just write a script to make it easier to get from the obscure folders they're usually saved to.

@sentree
Copy link
Member

sentree commented Apr 4, 2017

I was planning on writing an RQT plugin for ros_alarms at some point before I package it, but someone might want to implement a quick fix while I am working on the CI.

@DSsoto
Copy link
Member Author

DSsoto commented Apr 4, 2017

@mattlangford want's to create a python version of roscpp throttled logging.

I have writing dvl beam heartbeat listeners with a dvl_loss meta alarm on my list.

@mattlangford
Copy link
Member

Update: I was mistaken on two fronts.

  1. It looks like python throttled logging has gotten added in the new version of ros.
  2. The driver (which I thought was in python) is in a header file.

I am going to make changes to this file to reduce the logging it does without losing data.

@DSsoto
Copy link
Member Author

DSsoto commented Apr 16, 2017

Update:

@sentree
Copy link
Member

sentree commented May 13, 2017

Have we solved this with the implementation of kill bags from online_bagger?

@DSsoto
Copy link
Member Author

DSsoto commented May 21, 2017

@whispercoros in a way yes. But this should be easier to parse, and able to be done more quickly. I made a more specific issue to address this (#260).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants