From mboxrd@z Thu Jan 1 00:00:00 1970 From: Nikolay Aleksandrov Subject: Re: [BUG] "$ ss -a" incorrectly displays raw sockets as udp sockets Date: Thu, 09 Jul 2015 17:30:39 +0200 Message-ID: <559E939F.8030307@cumulusnetworks.com> References: <559E8178.9080808@beyondsemi.com> <1436454853.24939.31.camel@edumazet-glaptop2.roam.corp.google.com> <1436455270.24939.32.camel@edumazet-glaptop2.roam.corp.google.com> <20150709152804.GA16132@angus-think.lan> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Cc: Miha Marolt , netdev@vger.kernel.org To: Vadim Kochan , Eric Dumazet Return-path: Received: from mail-wi0-f181.google.com ([209.85.212.181]:37370 "EHLO mail-wi0-f181.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751125AbbGIPb1 (ORCPT ); Thu, 9 Jul 2015 11:31:27 -0400 Received: by wiclp1 with SMTP id lp1so112751145wic.0 for ; Thu, 09 Jul 2015 08:31:26 -0700 (PDT) In-Reply-To: <20150709152804.GA16132@angus-think.lan> Sender: netdev-owner@vger.kernel.org List-ID: On 07/09/2015 05:28 PM, Vadim Kochan wrote: > On Thu, Jul 09, 2015 at 05:21:10PM +0200, Eric Dumazet wrote: >> On Thu, 2015-07-09 at 17:14 +0200, Eric Dumazet wrote: >> >>> >>> If I checkout iproute2 tree to db08bdb816d337102c5486744008db9c9faa43bf >>> (before buggy commit) we indeed had this result : >>> >>> # ./ss -an | grep 127.0.0.1 >>> raw UNCONN 213486 0 127.0.0.1:6 *:* >> >> Maybe following patch would fix the issue. >> >> diff --git a/misc/ss.c b/misc/ss.c >> index 03f92fa69270..9b7dad00290c 100644 >> --- a/misc/ss.c >> +++ b/misc/ss.c >> @@ -1552,6 +1552,8 @@ out: >> static char *proto_name(int protocol) >> { >> switch (protocol) { >> + case 0: >> + return "raw"; >> case IPPROTO_UDP: >> return "udp"; >> case IPPROTO_TCP: >> @@ -2416,7 +2418,7 @@ static int dgram_show_line(char *line, const struct filter *f, int family) >> if (n < 9) >> opt[0] = 0; >> >> - inet_stats_print(&s, IPPROTO_UDP); >> + inet_stats_print(&s, 0); >> >> if (show_details && opt[0]) >> printf(" opt:\"%s\"", opt); >> >> > > Yes, it fixes the issue too. Yeah, I think this is better than my solution because it's not really IPPROTO_RAW protocol, so I'll drop my submission. Thanks, Nik