On 2015-07-30 at 10:55:06 +0200, Vadim Kochan <[email protected]> wrote: > On Thu, Jul 30, 2015 at 09:09:13AM +0200, Tobias Klauser wrote: > > On 2015-07-29 at 17:07:29 +0200, Vadim Kochan <[email protected]> wrote: > > > From: Vadim Kochan <[email protected]> > > > > > > In the previous version there was no panic if > > > file does not exist, so lets follow this behaviour. > > > > In my opinion the current behavior is fine. These files should exists on > > any decent system which is needed to run the netsniff-ng tools. > > Hi Tobias, > > But I got this on network namespace: > > $ ip netns exec ns0 netsniff-ng -i veth0 > > Really I dont know yet what is the difference if to run this on > separated net namespace.
Then just running it with `netsniff-ng -A' to avoid setting of the socket memory options might be the better solution. I would prefer to see if the socket memory value(s) couldn't be set instead of silently ignoring the issue, as would be the case with your patch applied. If anything we could maybe convert the panic() in set_system_socket_mem() to a simple printf() and/or already check the return value of get_system_socket_mem() inside set_system_socket_memory() What do you think? -- You received this message because you are subscribed to the Google Groups "netsniff-ng" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
