From owner-freebsd-scsi@FreeBSD.ORG Mon Mar 9 21:53:16 2009 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id DEC231065674; Mon, 9 Mar 2009 21:53:16 +0000 (UTC) (envelope-from jtn@jtn.cx) Received: from securemail.jtn.cx (hindenburg.jtn.cx [208.65.173.122]) by mx1.freebsd.org (Postfix) with ESMTP id AA00E8FC0A; Mon, 9 Mar 2009 21:53:16 +0000 (UTC) (envelope-from jtn@jtn.cx) Received: from localhost (localhost [127.0.0.1]) by securemail.jtn.cx (Postfix) with ESMTP id 084198A1C5; Mon, 9 Mar 2009 17:27:21 -0400 (EDT) Received: from securemail.jtn.cx ([127.0.0.1]) by localhost (hindenburg.jtn.cx [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id uAh3rmSQhE2v; Mon, 9 Mar 2009 17:27:20 -0400 (EDT) Received: by securemail.jtn.cx (Postfix, from userid 1001) id 448808A16D; Mon, 9 Mar 2009 17:27:20 -0400 (EDT) Date: Mon, 9 Mar 2009 17:27:20 -0400 From: "Jason T. Nelson" To: freebsd-questions@freebsd.org Message-ID: <20090309212720.GA49294@jtn.cx> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="ikeVEW9yuYc//A+q" Content-Disposition: inline User-Agent: Mutt/1.5.13 (2006-08-11) Cc: freebsd-scsi@freebsd.org Subject: iSCSI initiator lockups X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: jtn@jtn.cx List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 09 Mar 2009 21:53:17 -0000 --ikeVEW9yuYc//A+q Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable I'm running into some odd headaches regarding what looks like iSCSI initiat= ors going to sleep for approximately 30 seconds before returning to life and pumping a ton of information back to the target. While this is happening, system load climbs up alarmingly fast. Looking at tcpdumps in Wireshark, it shows what appears to be a nearly exact 30 second delay where the initiator stops talking to the target server, then abruptly restarts. Currently 8 machines are talking to 2 servers with 4 targets a piece, and while its= =20 working, we get good throughput. Activity is moderately high, as we are=20 using the iSCSI targets as spool disks in an email cluster. As it appears that iscsi-target is a single-threaded process, would it be valuable to put each target in its own process on its own port? At any rate, this is causing serious problems on the mail processing machines. --=20 Jason T. Nelson GPG key 0xFF676C9E --ikeVEW9yuYc//A+q Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.9 (FreeBSD) iEYEARECAAYFAkm1ibcACgkQ+rtZnf9nbJ4V6wCeJr2pr9s77kOH+PalLDQWfeu+ NuoAoL/8JEzJ0n00W1r5H+TZsNmFzSbz =YS7i -----END PGP SIGNATURE----- --ikeVEW9yuYc//A+q--