From nobody Fri Oct 14 11:24:00 2022 X-Original-To: freebsd-hackers@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 4MpkYK6RSCz4fFGg for ; Fri, 14 Oct 2022 11:24:05 +0000 (UTC) (envelope-from nonesuch@longcount.org) Received: from mail-qt1-x835.google.com (mail-qt1-x835.google.com [IPv6:2607:f8b0:4864:20::835]) (using TLSv1.3 with cipher TLS_AES_128_GCM_SHA256 (128/128 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256 client-signature RSA-PSS (2048 bits) client-digest SHA256) (Client CN "smtp.gmail.com", Issuer "GTS CA 1D4" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id 4MpkYJ656Pz3sWB for ; Fri, 14 Oct 2022 11:24:04 +0000 (UTC) (envelope-from nonesuch@longcount.org) Received: by mail-qt1-x835.google.com with SMTP id z8so3428150qtv.5 for ; Fri, 14 Oct 2022 04:24:04 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=longcount.org; s=google; h=to:in-reply-to:references:message-id:date:subject:mime-version:from :content-transfer-encoding:from:to:cc:subject:date:message-id :reply-to; bh=fuShmTObryoGMHfQN5lvWBg4cKQgv9V6aQixQz9lkfI=; b=lhRfyQ0/a/pILlGqm+daStyBzmGmd13x1/8/LrH7ur5JfDVKsAEX8z4pUzOU7kZxRe nfUWguiyJ3FUxYrVDUOCuK5gGBZ36G6R2/NrWFsrTvi9uoQKszSB5ou4PzPoUIISExev zKYm8h4OOaiwZVv8qxCDyFRtKfMtqCMMtLP037URFLLx32Sp/ubiwBHOqKZLhwwFUFzF HFA7DuZfLIG1TuAKS24TIe8/0VRxm7H4j44nRac/hbO5OYW1j1sVdHkhXTWE9vLzNxw5 DeIc3omnSMVeWuD/gQCIIRrRunfy7FVoSYXfpj8KC3dXtj7f4KoDjeiYOzfNVO1ZOfJz PgXA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=to:in-reply-to:references:message-id:date:subject:mime-version:from :content-transfer-encoding:x-gm-message-state:from:to:cc:subject :date:message-id:reply-to; bh=fuShmTObryoGMHfQN5lvWBg4cKQgv9V6aQixQz9lkfI=; b=mb8x8iQgsHAIRZ+BVnwKspS2P0Np+gmIDbRPRJsBtHMkMopeF9h4uthoKitNY8bKRF XuiX2TSS+kxhW9Qd1ZguaeSki6qNK7bBt8yLwggT5p14qkcRDXMrjvE9qiFXMKcaA3oP 3LYcJ/zMhEjj+p2UhHOckqu02XriKKt2oZ5YPZ8rRLG4xldAeSyciCblsoFBgN41+UaA ZA8dUl8dAEqox5ajQ8uWwYxczsW5sd8zKOTSEsjgzNQj4c7FvUP+tWANoB/+ySyvLfvR gX1HvVM/U3uTfAUOqK6N7VRYWHOqAcYY2e3a32ZtXaTqEZqn0G440EXcyXARiuOEbzc1 i/vg== X-Gm-Message-State: ACrzQf3HYotmGgPwdhwbt6vBH3t9K1zhRuMabnzBH24MA9pZutkBEGah +V5pQUuNwoEeT1MntTJqg673PibNVidmVPPp X-Google-Smtp-Source: AMsMyM7L85QA8AsNQtUw4L5+brLbjv5Wa3o9Pn28yQTb6evXfEZaejzfffvtuW0plDcNYQg40q7h1A== X-Received: by 2002:ac8:5b10:0:b0:39c:d63a:d88 with SMTP id m16-20020ac85b10000000b0039cd63a0d88mr3560037qtw.682.1665746642031; Fri, 14 Oct 2022 04:24:02 -0700 (PDT) Received: from smtpclient.apple (pool-68-161-195-218.nycmny.fios.verizon.net. [68.161.195.218]) by smtp.gmail.com with ESMTPSA id t26-20020a37ea1a000000b006b5cc25535fsm2159967qkj.99.2022.10.14.04.24.01 for (version=TLS1_3 cipher=TLS_AES_128_GCM_SHA256 bits=128/128); Fri, 14 Oct 2022 04:24:01 -0700 (PDT) Content-Type: multipart/alternative; boundary=Apple-Mail-C7629555-65B0-4DCE-906B-F530D6D37931 Content-Transfer-Encoding: 7bit From: Mark Saad List-Id: Technical discussions relating to FreeBSD List-Archive: https://lists.freebsd.org/archives/freebsd-hackers List-Help: List-Post: List-Subscribe: List-Unsubscribe: Sender: owner-freebsd-hackers@freebsd.org Mime-Version: 1.0 (1.0) Subject: Re: is this FreeBSD problem or HP switch Date: Fri, 14 Oct 2022 07:24:00 -0400 Message-Id: References: In-Reply-To: To: freebsd-hackers@freebsd.org X-Mailer: iPhone Mail (19G82) X-Rspamd-Queue-Id: 4MpkYJ656Pz3sWB X-Spamd-Bar: -- Authentication-Results: mx1.freebsd.org; dkim=pass header.d=longcount.org header.s=google header.b="lhRfyQ0/"; dmarc=none; spf=pass (mx1.freebsd.org: domain of nonesuch@longcount.org designates 2607:f8b0:4864:20::835 as permitted sender) smtp.mailfrom=nonesuch@longcount.org X-Spamd-Result: default: False [-3.00 / 15.00]; NEURAL_HAM_MEDIUM(-1.00)[-1.000]; NEURAL_HAM_LONG(-1.00)[-1.000]; NEURAL_HAM_SHORT(-1.00)[-1.000]; MV_CASE(0.50)[]; R_DKIM_ALLOW(-0.20)[longcount.org:s=google]; R_SPF_ALLOW(-0.20)[+ip6:2607:f8b0:4000::/36]; MIME_GOOD(-0.10)[multipart/alternative,text/plain]; DKIM_TRACE(0.00)[longcount.org:+]; RCVD_IN_DNSWL_NONE(0.00)[2607:f8b0:4864:20::835:from]; FROM_EQ_ENVFROM(0.00)[]; MIME_TRACE(0.00)[0:+,1:+,2:~]; MLMMJ_DEST(0.00)[freebsd-hackers@freebsd.org]; RCVD_TLS_LAST(0.00)[]; DMARC_NA(0.00)[longcount.org]; RCVD_VIA_SMTP_AUTH(0.00)[]; ASN(0.00)[asn:15169, ipnet:2607:f8b0::/32, country:US]; FROM_HAS_DN(0.00)[]; ARC_NA(0.00)[]; RCVD_COUNT_THREE(0.00)[3]; TO_MATCH_ENVRCPT_ALL(0.00)[]; TO_DN_NONE(0.00)[]; PREVIOUSLY_DELIVERED(0.00)[freebsd-hackers@freebsd.org]; RCPT_COUNT_ONE(0.00)[1]; MID_RHS_MATCH_FROM(0.00)[] X-ThisMailContainsUnwantedMimeParts: N --Apple-Mail-C7629555-65B0-4DCE-906B-F530D6D37931 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable > On Oct 13, 2022, at 10:11 PM, Zhenlei Huang wrote: >=20 > =EF=BB=BFSorry for missing CC >=20 >> Begin forwarded message: >>=20 >> From: Zhenlei Huang >> Subject: Re: is this FreeBSD problem or HP switch >> Date: October 14, 2022 at 10:08:13 AM GMT+8 >> To: Jason Bacon >>=20 >>=20 >>=20 >>> On Oct 14, 2022, at 8:11 AM, Jason Bacon wrote: >>>=20 >>>> On 10/13/22 17:48, Wojciech Puchar wrote: >>>> i have HP-2530-24G switch in home and 3 VLANs. >>>> there was no traffic on 2 of them, one vlan have full bandwitch traffic= from computer A to computer B - at 1000Mbit/s >>>> At this time ping from computer C to computer A (computer C have 100Mbi= t/s card) was random between 1 and 1000ms. >>>> Just limiting artifically speed to 950Mbit/s solved the problem. >>>> Is this because switch behaves that way or can it be a FreeBSD problem (= all computers runs FreeBSD) >>>=20 >>> I saw an issue some years ago where one of my FreeBSD servers was overwh= elming a Cisco switch while pushing files out to HTCondor nodes, causing it t= o drop packets. Our networks guru confirmed the problem by monitoring the s= witch. Since no handshaking was possible, I worked around it by throttling t= he interface on the FreeBSD box, using ipfw if I recall correctly. >>>=20 >>> --=20 >>> Life is a game. Play hard. Play fair. Have fun. >>>=20 >>>=20 >>=20 >>=20 >> Modern x86 hardware can easily saturate Gigabit network. Probably it is n= ot the issue of the box, or the OS ( FreeBSD in the context ). >> There's an old post about ICMP packet drops on the HPE community, https:/= /community.hpe.com/t5/aruba-provision-based/hp-procuvre-2530-icmp-packet-los= t/td-p/6347333 . Although I can not conclude it is the problem of switch, bu= t you can try connection the two FreeBSD box directly with crossover etherne= t cable and repeat the test. >>=20 >> Also the statistics reported by switch / OS are extremely useful in produ= ction ( when excluding some suspicious objects is not an option ). >>=20 >> Best regards, >> Zhenlei One other thing I stopped using hp switches in favor of fiber store=E2=80=99= s fs.com white box switches a few years ago . They are inexpensive and dece= nt enough , and Cisco like .=20 --- Mark Saad | nonesuch@longcount.org >=20 --Apple-Mail-C7629555-65B0-4DCE-906B-F530D6D37931 Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: quoted-printable

On Oct 13, 2022, at 10:11 PM, Zhenlei Huang <zlei.huang@gmail.com&= gt; wrote:

=EF=BB=BFSorry for missing CC

Begin forwarded message:
<= br class=3D"Apple-interchange-newline">
From: Zhenlei Huang <zlei.huang@gmail.com>
Su= bject: Re: is this FreeBSD p= roblem or HP switch
<= span style=3D"font-family: -webkit-system-font, Helvetica Neue, Helvetica, s= ans-serif; color:rgba(0, 0, 0, 1.0);" class=3D"">Date: October 14, 2022 at 10:08:13 AM GMT+8
To: Jason Bac= on <bacon4000@gmail.com= >



On Oct 14, 2022, at 8= :11 AM, Jason Bacon <ba= con4000@gmail.com> wrote:

On 10/13/22 17:48, Wojciech Puchar wrote:<= br class=3D"">
i have HP-2530-24G switch= in home and 3 VLANs.
there was no traffic on 2 of them, one v= lan have full bandwitch traffic from computer A to computer B - at 1000Mbit/= s
At this time ping from computer C to computer A (computer C h= ave 100Mbit/s card) was random between 1 and 1000ms.
Just limi= ting artifically speed to 950Mbit/s solved the problem.
Is thi= s because switch behaves that way or can it be a FreeBSD problem (all comput= ers runs FreeBSD)

I saw an issue s= ome years ago where one of my FreeBSD servers was overwhelming a Cisco switc= h while pushing files out to HTCondor nodes, causing it to drop packets. &nb= sp;Our networks guru confirmed the problem by monitoring the switch.  S= ince no handshaking was possible, I worked around it by throttling the inter= face on the FreeBSD box, using ipfw if I recall correctly.
--
Life is a game.  Play hard.  Play fa= ir.  Have fun.


=

Mode= rn x86 hardware can easily saturate Gigabit network. Probably it is not the i= ssue of the box, or the OS ( FreeBSD in the context ).
= There's an old post about ICMP packet drops on the HPE community, https://community.hpe.com/t5/aruba-p= rovision-based/hp-procuvre-2530-icmp-packet-lost/td-p/6347333 . Alt= hough I can not conclude it is the problem of switch, but you can try connec= tion the two FreeBSD box directly with crossover ethernet cable and repeat t= he test.

Also the s= tatistics reported by switch / OS are extremely useful in production ( when e= xcluding some suspicious objects is not an option ).
Best regards,
Zhenle= i

One other thing I stopped using hp switches in favor of fiber store=E2=80=99= s fs.com white box switches a few years ago .  They are inexpensive and= decent enough , and Cisco like . 


-= --
Mark Saad | nonesuch@longcount.org

= --Apple-Mail-C7629555-65B0-4DCE-906B-F530D6D37931--