From nobody Sun Sep 19 18:02:53 2021 X-Original-To: stable@mlmmj.nyi.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2610:1c1:1:606c::19:1]) by mlmmj.nyi.freebsd.org (Postfix) with ESMTP id 77ED217CE99D for ; Sun, 19 Sep 2021 18:03:08 +0000 (UTC) (envelope-from eugene@zhegan.in) Received: from elf.hq.norma.perm.ru (mail.norma.perm.ru [128.127.146.8]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256 client-signature RSA-PSS (2048 bits) client-digest SHA256) (Client CN "mail.norma.perm.ru", Issuer "Let's Encrypt Authority X3" (not verified)) by mx1.freebsd.org (Postfix) with ESMTPS id 4HCFsl1BB3z3t4V for ; Sun, 19 Sep 2021 18:03:06 +0000 (UTC) (envelope-from eugene@zhegan.in) Received: from [192.168.243.7] ([192.168.243.7]) by elf.hq.norma.perm.ru (8.16.1/8.15.2) with ESMTP id 18JI0Xox002201 for ; Sun, 19 Sep 2021 23:00:33 +0500 (+05) (envelope-from eugene@zhegan.in) DKIM-Signature: v=1; a=rsa-sha256; c=simple/simple; d=vivat-retail.ru; s=key; t=1632074434; bh=bLTiNB1C35k+4AduRNtDtzE3hDXr/u91bw0PCMBW9uw=; h=Subject:To:References:From:Date:In-Reply-To; b=en8muVxbkOUmv7nlDMucqWJ6QmTeUG5U6XsFN7+T5tVLJdMJmEjv4n+0lm+7Sj25o DV1OqCAFWIGH7t6L4otbcaoXlKBtMWegqTijmKdrALG6j2xTssF/yApbSsXigZWi4p aSiWsprcX37yxT6Z1x6ymrSm19IlFsU1NDufiqLo= Subject: Re: FreeBSD 12.1 vs 13.0 TCP connections count To: stable@freebsd.org References: <9cac45f1-9b81-5ed8-b3ae-c9567ac977cb@zhegan.in> From: "Eugene M. Zheganin" Message-ID: Date: Sun, 19 Sep 2021 23:02:53 +0500 User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:78.0) Gecko/20100101 Thunderbird/78.14.0 List-Id: Production branch of FreeBSD source code List-Archive: https://lists.freebsd.org/archives/freebsd-stable List-Help: List-Post: List-Subscribe: List-Unsubscribe: Sender: owner-freebsd-stable@freebsd.org X-BeenThere: freebsd-stable@freebsd.org MIME-Version: 1.0 In-Reply-To: <9cac45f1-9b81-5ed8-b3ae-c9567ac977cb@zhegan.in> Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 8bit Content-Language: ru X-Rspamd-Queue-Id: 4HCFsl1BB3z3t4V X-Spamd-Bar: - Authentication-Results: mx1.freebsd.org; dkim=fail ("headers rsa verify failed") header.d=vivat-retail.ru header.s=key header.b=en8muVxb; dmarc=none; spf=pass (mx1.freebsd.org: domain of eugene@zhegan.in designates 128.127.146.8 as permitted sender) smtp.mailfrom=eugene@zhegan.in X-Spamd-Result: default: False [-1.25 / 15.00]; ARC_NA(0.00)[]; NEURAL_HAM_MEDIUM(-1.00)[-1.000]; FROM_HAS_DN(0.00)[]; TO_MATCH_ENVRCPT_ALL(0.00)[]; R_SPF_ALLOW(-0.20)[+a]; MIME_GOOD(-0.10)[text/plain]; TO_DN_NONE(0.00)[]; PREVIOUSLY_DELIVERED(0.00)[stable@freebsd.org]; RCPT_COUNT_ONE(0.00)[1]; NEURAL_HAM_LONG(-1.00)[-1.000]; DMARC_NA(0.00)[zhegan.in]; R_DKIM_REJECT(1.00)[vivat-retail.ru:s=key]; DKIM_TRACE(0.00)[vivat-retail.ru:-]; NEURAL_SPAM_SHORT(0.05)[0.052]; FROM_EQ_ENVFROM(0.00)[]; MIME_TRACE(0.00)[0:+]; RCVD_TLS_LAST(0.00)[]; ASN(0.00)[asn:212494, ipnet:128.127.146.0/24, country:RU]; RCVD_COUNT_TWO(0.00)[2]; MID_RHS_MATCH_FROM(0.00)[] X-ThisMailContainsUnwantedMimeParts: N Hello, 13.09.2021 11:33, Eugene M. Zheganin пишет: > [...] > > I'm also graphing the balancer/nginx accepts/handled rate which is > constant at 4K/sec - means the flow itself isn't changing a lot. > > I'm also graphing the connections scoreboard from nginx and it also > shows x4 drops/rises when switching from 12.1 to 13.0. > > Is this a bug/glitch of the 13.3 statistics (seems unprobeble because > nginx shows it too) or is there some other explanation that I don't get ? > In the end this looks like a regression in the end. Considering two identical servers rented in one datacenter, I have to downgrade 13.x balancer to 12.2 - I'm unable to find other eplanation to the things described above. The difference in connection numbers isn't the major reason; the real reason is the inability to handle the load, and the most discouraging thing is that I'm unable to measure this inability using formal metrics: cpu/disk load is okay, no network errors on the driver layer, no complains in the dmesg or syslog. Eugene.