I'm seeing osqueryd segfaulting on some of our ins...
# general
z
I'm seeing osqueryd segfaulting on some of our instances, what can I check here to see why?
l
on mac os you could look for
/Library/Logs/DiagnosticReports/*osquery*.crash
z
this is on rhel itself
a
I’ve seen a bug report about the
magic
table, is there any chance it is being used in your query packs? EDIT: fixed the table name
I think we had support for debug symbol packages on *.rpm (cc @Stefano Bonicatti)
z
no, we're just getting a list of packages and processes
no
magic
table that I know of.
s
There should be a
osquery-debuginfo
package to be installed so that symbols are available when looking at stack traces
z
how do I use that?
just run it?
s
That package exists only to install the debug symbols; to get a stack trace you either have to run osquery under a debugger (gdb), or have core dumps enabled and point gdb to it.
z
it's not clear how to enable core dumps for osquery
s
You might want to look at https://access.redhat.com/solutions/56021. There’s no specific way for osquery, it’s a system feature to enable system wide or temporarily in a shell where you would then launch osquery. The procedure differs slightly from distro to distro and versions; that link should help.