[Assert failed rc >= 0 l1ctl.c:83] cell_log crashes (mostly) when monitoring specific ARFCN
Upon execution of ./cell_log -A 117 -i 127.0.0.1, the program crashes after a while with:
Assert failed rc >= 0 l1ctl.c:83 backtrace() returned 13 addresses /usr/local/lib/libosmocore.so.17(osmo_generate_backtrace+0x18) [0x7f10d726684f] /usr/local/lib/libosmocore.so.17(+0x24557) [0x7f10d7266557] /usr/local/lib/libosmocore.so.17(osmo_panic+0xdc) [0x7f10d7266638] ./cell_log() [0x407f4a] ./cell_log() [0x40821d] /usr/local/lib/libosmocore.so.17(osmo_wqueue_bfd_cb+0x3a) [0x7f10d7259e3d] /usr/local/lib/libosmocore.so.17(+0xf2c4) [0x7f10d72512c4] /usr/local/lib/libosmocore.so.17(+0xf3ba) [0x7f10d72513ba] /usr/local/lib/libosmocore.so.17(osmo_select_main+0x15) [0x7f10d72513d8] ./cell_log() [0x402f2c] /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf0) [0x7f10d66db840] ./cell_log() [0x4030b9] Aborted
Line 83 in l1ctl.c:
/* if FN is negative, we were called for something that's not CCCH! */ OSMO_ASSERT(rc >= 0);
The problem exists mostly when specifying an ARFCN manually, although i have seen the same type of crash even when no ARFCN is specified and an active scan is in place, but it is a very rare occasion and takes much more time to occur.
I have tried with different cells of different network carriers, so this is very unlikely to be a network issue.
Using the latest revision to date of master branch.