Hi Mikhail,
I see, yes. The reason nothing happens (not out of coincidences :-)) is
that all the reads are checking whether we are in the context of a tee
plugin (which does not apply to pmacctd / uacctd). But i agree with you:
that is recipe for potential disaster so i committed a memset() ther
Hi Paolo,
The pm_pcap_cb has however its own instance of struct plugin_requests
req :
https://github.com/pmacct/pmacct/blob/d72440dc9a7d0d0a7ed9502f1dd31b90105b1d95/src/nl.c#L51
,
and noone zeroes it up before using it seems.
Mikhail
On Tue, 21 Jan 2020 at 02:25, Paolo Lucente wrote:
>
>
> Hi
Hi Mikhail,
If you see all the daemons that make use of the 'req' structure have a
memset() for 'req' shortly after its declaration. For example here in
pmacctd: https://github.com/pmacct/pmacct/blob/master/src/pmacctd.c#L360
Paolo
On Fri, Jan 17, 2020 at 07:10:13PM +0100, Mikhail Sennikovsky
Hi all,
I was running through the pm_pcap_cb code, and it looks like the "req"
passed to exec_plugins(&pptrs, &req); at
https://github.com/pmacct/pmacct/blob/d72440dc9a7d0d0a7ed9502f1dd31b90105b1d95/src/nl.c#L167
and to load_id_file at
https://github.com/pmacct/pmacct/blob/d72440dc9a7d0d0a7ed9502f