You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
According to the documentation, fields like proc.aname do not make sense in output by themselves but only in filters. So proc.aname = bash is legit, My aname is '%proc.aname' is not.
Currently, you can still write an output string like the one above, and this is what happens:
If "%proc.aname" is evaluated as-is or at the end of an output string, it'll return proc.aname[0]
If the formatting token %proc.aname is in the middle of the output string this results in an unintialized read, as m_argid is NOT set in sinsp_filter_check_thread::parse_field_name and so the behavior is undefined
Describe the bug
According to the documentation, fields like
proc.aname
do not make sense in output by themselves but only in filters. Soproc.aname = bash
is legit,My aname is '%proc.aname'
is not.Currently, you can still write an output string like the one above, and this is what happens:
"%proc.aname"
is evaluated as-is or at the end of an output string, it'll returnproc.aname[0]
%proc.aname
is in the middle of the output string this results in an unintialized read, asm_argid
is NOT set insinsp_filter_check_thread::parse_field_name
and so the behavior is undefinedHow to reproduce it
Expected behaviour
Either a NULL which is properly handled or a syntax error. Both are acceptable I think. I prefer a syntax error at this point.
Screenshots
Environment
Additional context
The text was updated successfully, but these errors were encountered: