How log all gdb output when using 'command'

With logging enabled as described here, I am using command to log and continue at a breakpoint as described here.

Full gdb interraction:

$ sudo gdb -p 30560 

(gdb) set logging file myDbg.log
(gdb) set logging on
Copying output to myDbg.log.
(gdb) set trace-commands on
(gdb) set pagination off
(gdb) br __libc_malloc
(gdb) conti

// Once the bp is hit:
(gdb) info stack         << Output for this shows up in myDbg.log 

(gdb) command
+command
Type commands for breakpoint(s) 1, one per line.
End with a line saying just "end".
>info stack
>conti
>end
(gdb) conti

// Now on the console I get:
Continuing.
[Switching to Thread 0x7fff98949700 (LWP 30627)]

Thread 49 "triggerproc" hit Breakpoint 1, 0x00007fffecc05f70 in malloc () from /lib64/libc.so.6
+info stack
#0  0x00007fffecc05f70 in malloc () from /lib64/libc.so.6
#1  0x00007fffed79d18d in operator new(unsigned long) () from /lib64/libstdc++.so.6
...
+conti
[Switching to Thread 0x942e700 (LWP 32191)]

// While the log file only has
Continuing.
[Switching to Thread 0x7fff98949700 (LWP 30627)]

Thread 49 "triggerproc" hit Breakpoint 1, 0x00007fffecc05f70 in malloc () from /lib64/libc.so.6
+info stack
                ** expected to see stack trace here **
+conti
[Switching to Thread 0x942e700 (LWP 32191)]

I don't get the info stack output in the log. If I do info stack outside of command, i.e. regular gdb prompt, its output is written to the log file. How do I get info ... outputs from within command into the log file? Application is a multithreaded C++ running on a CentOS VM, if that matters.