What steps will reproduce the problem?
1. Build an Android node that subscribes to a message with a fixed-size array,
such as nav_msgs/Odometry
2. Run the Android app and publish to that activity with rostopic
Example command: rostopic pub /odom nav_msgs/Odometry '{header: {frame_id: "/UTM"}}'
What is the expected output? What do you see instead?
Expected the message to be passed to the Android activity. Instead, the
following error in the terminal:
[WARN] [WallTime: 1334694631.826881] Could not process inbound connection: Client [/avatar] wants topic [/ros_odom] to have datatype/md5sum [nav_msgs/Odometry/86b740de5f66321fbd1151710f00a7a1], but our version has [nav_msgs/Odometry/cd5e73d190d741a2f92e81eda573aca7] Dropping connection.{'topic': '/ros_odom', 'message_definition': '# This represents an estimate of a position and velocity in free space. \n# The pose in this message should be specified in the coordinate frame given by header.frame_id.\n# The twist in this message should be specified in the coordinate frame given by the child_frame_id\nHeader header\nstring child_frame_id\ngeometry_msgs/PoseWithCovariance pose\ngeometry_msgs/TwistWithCovariance twist\n', 'md5sum': '86b740de5f66321fbd1151710f00a7a1', 'type': 'nav_msgs/Odometry', 'callerid': '/avatar'}
and corresponding error from LogCat:
04-19 17:00:31.840: E/scriberHandshakeHandler(1456): Subscriber handshake failed: Client [/avatar] wants topic [/ros_odom] to have datatype/md5sum [nav_msgs/Odometry/86b740de5f66321fbd1151710f00a7a1], but our version has [nav_msgs/Odometry/cd5e73d190d741a2f92e81eda573aca7] Dropping connection.
What version of the product are you using? On what operating system?
changeset: 171:f0b683aa1f8a
tag: tip
OS: Ubuntu 10.10, Android 2.3.3 (API 10)
Please provide any additional information below.
I have tested many messages (including all of those in the nav_msgs and
geometry_msgs packages and several custom messages), and it appears that this
hash mismatch occurs with only messages that include fixed-size array
components (e.g., PoseWithCovariance has a float64[36] component). Removing the
size definition (i.e., modifying to float64[]) fixes the issue, and adding a
size definition to another message that already works introduces the issue.
Original issue reported on code.google.com by jatow...@gmail.com on 18 Apr 2012 at 2:57
Original issue reported on code.google.com by
jatow...@gmail.com
on 18 Apr 2012 at 2:57