From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <pve-devel-bounces@lists.proxmox.com>
Received: from firstgate.proxmox.com (firstgate.proxmox.com [212.224.123.68])
	by lore.proxmox.com (Postfix) with ESMTPS id 3165E1FF143
	for <inbox@lore.proxmox.com>; Sat, 23 May 2026 23:30:24 +0200 (CEST)
Received: from firstgate.proxmox.com (localhost [127.0.0.1])
	by firstgate.proxmox.com (Proxmox) with ESMTP id 6C20B1BDFE;
	Sat, 23 May 2026 23:29:42 +0200 (CEST)
From: Thomas Lamprecht <t.lamprecht@proxmox.com>
To: Proxmox VE development discussion <pve-devel@lists.proxmox.com>,
	Mira Limbeck <m.limbeck@proxmox.com>
Subject: Re: [PATCH v2 storage 05/15] iscsi: add helper to get local config
Date: Sat, 23 May 2026 23:26:08 +0200
Message-ID: <20260523212856.2822353-3-t.lamprecht@proxmox.com>
X-Mailer: git-send-email 2.47.3
In-Reply-To: <20260430173220.441001-6-m.limbeck@proxmox.com>
References: <20260430173220.441001-1-m.limbeck@proxmox.com>
 <20260430173220.441001-6-m.limbeck@proxmox.com>
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable
X-Bm-Milter-Handled: 55990f41-d878-4baa-be0a-ee34c49e34d2
X-Bm-Transport-Timestamp: 1779571725320
X-SPAM-LEVEL: Spam detection results:  0
	AWL                     0.005 Adjusted score from AWL reputation of From:
 address
	BAYES_00                 -1.9 Bayes spam probability is 0 to 1%
	DMARC_MISSING             0.1 Missing DMARC policy
	KAM_DMARC_STATUS         0.01 Test Rule for DKIM or SPF Failure with Strict
 Alignment
	SPF_HELO_NONE           0.001 SPF: HELO does not publish an SPF Record
	SPF_PASS               -0.001 SPF: sender matches SPF record
Message-ID-Hash: PZIYN6PFKTNO7K7JNDPJSPVCT7RPDOPD
X-Message-ID-Hash: PZIYN6PFKTNO7K7JNDPJSPVCT7RPDOPD
X-MailFrom: t.lamprecht@proxmox.com
X-Mailman-Rule-Misses: dmarc-mitigation; no-senders; approved; loop;
 banned-address; emergency; member-moderation; nonmember-moderation;
 administrivia; implicit-dest; max-recipients; max-size; news-moderation;
 no-subject; digests; suspicious-header
X-Mailman-Version: 3.3.10
Precedence: list
List-Id: Proxmox VE development discussion <pve-devel.lists.proxmox.com>
List-Help: <mailto:pve-devel-request@lists.proxmox.com?subject=help>
List-Owner: <mailto:pve-devel-owner@lists.proxmox.com>
List-Post: <mailto:pve-devel@lists.proxmox.com>
List-Subscribe: <mailto:pve-devel-join@lists.proxmox.com>
List-Unsubscribe: <mailto:pve-devel-leave@lists.proxmox.com>

On Thu, 30 Apr 2026 19:27:03 +0200, Mira Limbeck wrote:=0D
> diff --git a/src/PVE/Storage/ISCSIPlugin.pm b/src/PVE/Storage/ISCSIPlugin=
.pm=0D
> @@ -36,6 +36,49 @@ my sub assert_iscsi_support {=0D
> +my $get_local_config =3D sub {=0D
> +    my ($scfg) =3D @_;=0D
> +=0D
> +    die "neither 'target' nor 'mapping' defined\n"=0D
> +        if !defined($scfg->{target}) && !defined($scfg->{mapping});=0D
> +=0D
> +    my $res =3D {};=0D
> +    if ($scfg->{mapping}) {=0D
> +        $res->{is_mapping} =3D 1;=0D
> +=0D
> +        my $local_mappings =3D=0D
> +            PVE::Storage::Mapping::find_mapping_on_current_node($scfg->{=
mapping});=0D
> +        die "no iSCSI per-node entries found for mapping '$scfg->{mappin=
g}'\n"=0D
> +            if !$local_mappings->@*;=0D
=0D
The die when $local_mappings is empty fires for any node that legitimately=
=0D
is not in the mapping's node set, and it propagates through status() and=0D
check_connection() every pvestatd cycle. Treating that node like a=0D
`nodes=3D`-restricted storage (reporting it as inactive) would be better;=0D
worth defining how storage `nodes=3D` and per-node `map` coverage relate.=0D
=0D
> +        for my $mapping ($local_mappings->@*) {=0D
> +            $res->{targets}->{ $mapping->{target} } =3D []=0D
> +                if !defined($res->{targets}->{ $mapping->{target} });=0D
> +            my $portals =3D [PVE::Tools::split_list($mapping->{portals})=
];=0D
> +=0D
> +            my $add_port =3D sub {=0D
> +                my ($val) =3D @_;=0D
> +=0D
> +                my ($ip, $port) =3D PVE::Tools::parse_host_and_port($val=
);=0D
> +                if (defined($port)) {=0D
> +                    return $val;=0D
> +                } else {=0D
> +                    # add default port=0D
> +                    return $ip . ':3260';=0D
> +                }=0D
> +            };=0D
> +            $portals->@* =3D map { $add_port->($_) } $portals->@*;=0D
> +=0D
> +            push $res->{targets}->{ $mapping->{target} }->@*, $portals->=
@*;=0D
> +        }=0D
> +        return $res;=0D
> +    } else {=0D
> +        $res->{is_mapping} =3D 0;=0D
> +        $res->{targets}->{ $scfg->{target} } =3D iscsi_portals($scfg->{t=
arget}, $scfg->{portal});=0D
> +        return $res;=0D
> +    }=0D
> +};=0D
=0D
Further, portals get canonicalized to host:port only on the mapping branch=
=0D
(add_port); the else branch keeps whatever iscsi_portals returns. This=0D
asymmetry is what later needs the `(?::3260)?` and unanchored matches in=0D
06/08. Normalizing both branches here once would let the comparisons be=0D
exact everywhere.=0D