I think I've found the problem. I thought I disabled a GR OOT block when I was running my tests this morning, but I either was wrong, or some process was hung in memory somewhere after I started stripping the flowgraph down. The issue was a block that was spawning a thread that was listening on a UDP socket. When the flowgraph ended, the the thread was told to close, but the socket was blocking waiting for data that was never going to come. I added in the timeout option to our socket initialization and it seems to work fine now.
________________________________ From: USRP-users <usrp-users-boun...@lists.ettus.com> on behalf of Jason Matusiak via USRP-users <usrp-users@lists.ettus.com> Sent: Wednesday, September 4, 2019 9:51 AM To: Ettus Mail List <usrp-users@lists.ettus.com> Subject: [USRP-users] flowgraphs not stopping I have a weird problem that I've seen for a while and I've just ignored until now. I've seen this with numerous pieces of hardware, but I am currently using an E320 in network mode. This seems to only happen when I am using RFNoC, so I assumed it was me, but today I was testing with only stock blocks and have the same issue. The problem is that when I run a GR flowgraph without a GUI in "prompt for exit" mode, it doesn't always seem to exit. I'll hit enter like I am supposed to, but it just hangs until I press ctrl-c. Currently I have an RFNoC:Radio -> DmaFIFO -> GR blocks. I have tested this with different combos of blocks and they all seem to have the same issue. I originally thought it was me or the split stream block, but they are both out of the equation now. Has anyone else seen this issue before? I am running: [INFO] [UHD] linux; GNU C++ version 4.8.5 20150623 (Red Hat 4.8.5-36); Boost_105300; UHD_3.14.1.HEAD-0-g5491b80e and GR: v3.7.13.4
_______________________________________________ USRP-users mailing list USRP-users@lists.ettus.com http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
