From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-ie0-f170.google.com (mail-ie0-f170.google.com [209.85.223.170]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (Client CN "smtp.gmail.com", Issuer "Google Internet Authority" (verified OK)) by huchra.bufferbloat.net (Postfix) with ESMTPS id 07B0B21F11C for ; Thu, 3 Jan 2013 16:40:41 -0800 (PST) Received: by mail-ie0-f170.google.com with SMTP id k10so19149948iea.1 for ; Thu, 03 Jan 2013 16:40:41 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:date:message-id:subject:from:to:cc:content-type; bh=jR6QYCQ5gmiwQQ/FiKwxIAIx0hH+RZQuSGGxC4KedU8=; b=ttwhHx0BD0vVZnbWJUqsvC+I0eTR+ZecZ8lHm2VLxkSZ22ftGlB/lqYSAsiW6hSLM+ /NymTxk8pu1glIo6PBXi5ODlnHuXLRvtp6LyPw21G3Doztuz6YSxdUsE2rxlL4KL2/eC 3ibR/N5XKcFObDcWcF5lRV4raRxkwXBZ29Spjg1F6ZeSzJu3SPAyttRm7FnFq/GMDuL4 NGuQqd0z0FBGxHrqVEjX0gp7hBmcd3tHrEXiO9MjRiagtOHzMbYcvIp5QcefHgs9bm8S 90Cc7cpGYddLQJ7OBeG7pYbP23uloVcRS3BoEC5DTI70a4lVBuxsZd84AXRcgGr0/BRf APUA== MIME-Version: 1.0 Received: by 10.50.88.136 with SMTP id bg8mr39472760igb.96.1357260041374; Thu, 03 Jan 2013 16:40:41 -0800 (PST) Received: by 10.64.135.39 with HTTP; Thu, 3 Jan 2013 16:40:41 -0800 (PST) Date: Thu, 3 Jan 2013 16:40:41 -0800 Message-ID: From: Dave Taht To: dpreed@reed.com Content-Type: text/plain; charset=ISO-8859-1 Cc: cerowrt-devel@lists.bufferbloat.net Subject: [Cerowrt-devel] parsing tc statistical output X-BeenThere: cerowrt-devel@lists.bufferbloat.net X-Mailman-Version: 2.1.13 Precedence: list List-Id: Development issues regarding the cerowrt test router project List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 04 Jan 2013 00:40:42 -0000 I just wanted to note that fq_codel and related have a few interesting stats. First important note: fq_codel as in the linux mainline, resets the "dropped" statistic when a queue becomes new. cero preserves it. Eric was making noises about removing dropped entirely due to the effect on cacheline behavior, I'm planning on keeping it... and adding more stats, actually! But if you parse dropped, only use the summary dropped stat not the queue one. Basic stats can be had via: tc -s qdisc show dev your_device Next up, on a busy system, doing a: tc -s class show dev your_device will actually show you the depth of your individual queues (those that have any depth). I will try to write up a wiki entry on what all those stats mean. At some point. Lastly: tc's output gets rapidly more complex and hard to parse with htb and simple_qos or dan's scripts...