Hello.
Sometimes it looks like some peers using TCP protocol are not able to discovery each other.
For example, there are two types of peers.
- Server Type: Always running without termination. Both UDP & TCP protocols are configured.
- Client Type: Short-live process. Using TCP to connect to the server.
From a point of time, both of them cannot send any sample to each other.
Once the problem happens, restarting client process doesn't solve the situation,
while restarting server fix the problem right after initiating DDS entities.
This is a critical issue for me.
Please help me!
Hello yoondo,
Can you increase the verbosity of the TCP transport?
There is information on the verbosity settings here: http://community.rti.com/rti-doc/500/ndds/doc/html/tcp_transport/group__NDDS__TransportTCPPropertyQosPolicy.html
<tcp_prefix>.logging_verbosity_bitmap
ˆ -1 (0xffffffff): do not change the current verbosity
ˆ 0x00: silence
ˆ 0x01: errors
ˆ 0x02: warnings
ˆ 0x04: local
ˆ 0x08: remote
ˆ 0x10: period
ˆ 0x80: other (used for control protocol tracing)
If you do not see any messages that indicate a problem, you can turn on all verbosity with this value:
0x9F: all (errors, warnings, local, remote, period, and other)
Thank you!
Rose