Paulius, I could not reproduce segfault.
Can you provide stack traces for all threads, not only for one in which segfault happened?

segfaulted on first try this time:

aspdb1:/glds/wincable/raid0-3x256gbSSD # /opt/fb25/bin/fbtracemgr -se localhost/3055:service_mgr -user sysdba -password masterkey -config /opt/fb25/fbtrace.conf -start
Trace session ID 2 started
^CSegmentation fault (core dumped)

call stack looks similar:

[New Thread 9683]
Core was generated by `/opt/fb25/bin/fbtracemgr -se localhost/3055:service_mgr -user sysdba -password'.
Program terminated with signal 11, Segmentation fault.
#0 gds__log (text=0x7ffbfb88fd18 "INET/inet_error: %s errno = %d")
    at ../src/include/../common/classes/alloc.h:408
408 ((MemoryBlock*) ((char*) block - MEM_ALIGN(sizeof(MemoryBlock))))->mbk_pool->deallocate(block);
(gdb) thread apply all bt

Thread 2 (Thread 9683):
#0 strcmp () at ../sysdeps/x86_64/strcmp.S:108
#1 0x00007ffbfbb55e2c in check_match (sym=0xfefefefefefefeff) at do-lookup.h:125
#2 0x00007ffbfbb562e4 in do_lookup_x (new_hash=1842243024, old_hash=0x7ffbf9a59aa8,
    ref=<value optimized out>, result=0x1000, scope=<value optimized out>, i=8, flags=5, skip=0x0)
    at do-lookup.h:197
#3 0x00007ffbfbb5652e in _dl_lookup_symbol_x (undef_name=0x7ffbfaa96158 "__cxa_finalize",
    undef_map=0x7ffbfbd34000, ref=0x7ffbf9a59b10, symbol_scope=0x7ffbfbd34358,
    version=0x7ffbfbd32f78, type_class=1, flags=5, skip_map=0x0) at dl-lookup.c:340
#4 0x00007ffbfbb5a3b9 in _dl_fixup (l=0xffffffffffffffff, reloc_offset=<value optimized out>)
    at dl-runtime.c:108
#5 0x00007ffbfbb602a2 in _dl_runtime_resolve () at ../sysdeps/x86_64/dl-trampoline.S:43
#6 0x00007ffbfaa96ea6 in __do_global_dtors_aux () from /lib64/libm.so.6
#7 0x0000000000000000 in ?? ()

Thread 1 (Thread 9682):
#0 gds__log (text=0x7ffbfb88fd18 "INET/inet_error: %s errno = %d")
    at ../src/include/../common/classes/alloc.h:408
#1 0x00007ffbfb7290c8 in inet_error (port=0x7ffbfbd5fd00,
    function=0x7ffbfbd5b8b8 <Address 0x7ffbfbd5b8b8 out of bounds>, operation=335544726,
    status=6433568) at ../src/remote/inet.cpp:2727
#2 0x00007ffbfb72a096 in packet_receive (port=0x7ffbfbd5fd00,
    buffer=0x7ffbfbd5bc18 <Address 0x7ffbfbd5bc18 out of bounds>,
    buffer_length=<value optimized out>, length=0x7fff3e50b746) at ../src/remote/inet.cpp:3150
#3 0x00007ffbfb72a535 in inet_getbytes (xdrs=0x7ffbfbd5fdf0, buff=0x7fff3e50b794 "ÿ\177",
    count=<value optimized out>) at ../src/remote/inet.cpp:2851
#4 0x00007ffbfb728c0c in inet_getlong (xdrs=0x7ffbfa65aa00, lp=0x7fff3e50b7c4)
    at ../src/remote/inet.cpp:2667
#5 0x00007ffbfb73300a in xdr_enum (xdrs=0x7ffbfa65aa00, ip=0x7ffbfbd64338)
    at ../src/remote/xdr.cpp:340
#6 0x00007ffbfb72feaa in xdr_protocol (xdrs=0x7ffbfa65aa00, p=0x7ffbfbd5b8b8)
    at ../src/remote/protocol.cpp:271
#7 0x00007ffbfb728cab in receive (main_port=0x7ffbfbd5fd00, packet=0x7ffbfbd64338)
    at ../src/remote/inet.cpp:2076
#8 0x00007ffbfb71d44c in receive_packet_noqueue (port=0x7ffbfbd5fd00, packet=0x7ffbfbd64338,
    user_status=<value optimized out>) at ../src/remote/interface.cpp:6207
#9 0x00007ffbfb71d966 in receive_response (rdb=0x7ffbfbd64300, packet=0x7ffbfbd64338)
    at ../src/remote/interface.cpp:6094
#10 0x00007ffbfb71e670 in info (user_status=0x7fff3e50fd20, rdb=0x7ffbfbd64300,
    operation=<value optimized out>, object=<value optimized out>, incarnation=<value optimized out>,
    item_length=<value optimized out>, items=0x7fff3e50fdc0 "@\004", recv_item_length=2,
    recv_items=0x4177c9 "?\001", buffer_length=16383, buffer=0x7fff3e50bd20 "?")
    at ../src/remote/interface.cpp:5718
#11 0x00007ffbfb726445 in REM_service_query (user_status=0x7fff3e50fd20,
---Type <return> to continue, or q <return> to quit---
    svc_handle=<value optimized out>, item_length=8, items=0x7fff3e50fdc0 "@\004",
    recv_item_length=<value optimized out>, recv_items=0x4177c9 "?\001",
    buffer_length=<value optimized out>, buffer=0x7fff3e50bd20 "?")
    at ../src/remote/interface.cpp:4024
#12 0x00007ffbfb4a8091 in isc_service_query (user_status=<value optimized out>,
    handle=0x7fff3e510a28, send_item_length=11040, send_items=0x7fff3e50fdc0 "@\004",
    recv_item_length=2, recv_items=0x4177c9 "?\001", buffer_length=<value optimized out>,
    buffer=0x7fff3e50bd20 "?") at ../src/jrd/why.cpp:4726
#13 0x0000000000403ede in Firebird::TraceSvcUtil::runService (this=<value optimized out>,
    spbSize=<value optimized out>, spb=<value optimized out>)
    at ../src/utilities/fbtracemgr/traceMgrMain.cpp:245
#14 0x0000000000404bdd in Firebird::TraceSvcUtil::startSession (this=0x7fff3e510a20,
    session=@0x7fff3e5105e0) at ../src/utilities/fbtracemgr/traceMgrMain.cpp:181
#15 0x000000000040fc0d in Firebird::fbtrace (uSvc=0x7ffbfbd562e0, traceSvc=0x7fff3e510a20)
    at ../src/jrd/trace/TraceCmdLine.cpp:459
#16 0x0000000000403cce in main (argc=<value optimized out>, argv=<value optimized out>)
    at ../src/utilities/fbtracemgr/traceMgrMain.cpp:346
Current language: auto; currently asm
Current language: auto; currently c++
(gdb)

Same might happen with fbsvcmgr.
Does not need frontporting - in FB3 there are no reasons for this segfault.