How to solve MD5sum version error ?

asked 2020-08-03 17:38:26 -0500

HELPme
7 ●4 ●5 ●6

updated 2020-08-12 09:27:54 -0500

I'm using Windows7: MATLAB R2020a Ubuntu 18.04: ROS melodic, UR3 in Gazebo My code is [urArm, urGoalMsg] = rosactionclient('arm_controller/follow_joint_trajectory'); Error message of ROS is how can I solve this. I tried deleting devel and build files and catkin_make again. It didn't work. Please help

Comments

Can you please post more information about your issue. I don't see a proper error message in there.

kscottz ( 2020-08-04 14:58:40 -0500 )edit

I'm using Windows7: MATLAB R2020a Ubuntu 18.04: ROS melodic, UR3 in Gazebo My code is [urArm, urGoalMsg] = rosactionclient('arm_controller/follow_joint_trajectory'); Error message of ROS is

[control_msgs/FollowJointTrajectoryActionResult/bce83d50f7bb28226801436caf0e2043], but our version has [control_msgs/FollowJointTrajectoryActionResult/c4fb3b000dc9da4fd99699380efcc5d9]. Dropping connection.

how can I solve this. I tried deleting devel and build files and catkin_make again. It didn't work. Please help

HELPme ( 2020-08-12 09:31:15 -0500 )edit

The msg definitions might be different. Easiest way to check if what you get is what you expected is to do a rosmsg info

parzival ( 2020-08-12 12:08:00 -0500 )edit

add a comment

answered 2020-08-12 10:55:47 -0500

gvdhoorn
86574 ●283 ●1432 ●1054 http://cor.tudelft.nl/

updated 2020-08-12 10:56:05 -0500

The problem here appears to be that your Matlab ROS support uses a different version of the control_msgs/FollowJointTrajectoryActionResult message than your ROS Melodic install.

The error message about the MD5 sums no matching is telling you this.

As to why this is the case I wouldn't know. control_msgs/FollowJointTrajectoryAction hasn't changed in many years, so it's unclear to me how this could happen.

As to how to fix this (or work-around it): I would suggest you ask Mathworks for support. Assuming you haven't made any changes to ros-melodic-control-msgs on your Ubuntu machine, the problem would seem to be on the side of Matlab.