From: Dominik Csapak <d.csapak@proxmox.com>
To: Yuri Konotopov <ykonotopov@gnome.org>, pve-devel@lists.proxmox.com
Subject: Re: [pve-devel] [PATCH v3 storage 1/1] fix #254: iscsi: add support for multipath iscsi targets
Date: Fri, 20 Oct 2023 11:23:28 +0200 [thread overview]
Message-ID: <f883722d-341d-4946-9420-314578b3a8bb@proxmox.com> (raw)
In-Reply-To: <20231017161948.139795-2-ykonotopov@gnome.org>
hi,
nice elegant solution for not needing to change the config
or introducing a new file :)
there is one caveats with this solution, but that should not
matter for most situations:
in case the configured portal is not accessible, any new node
in the cluster (or one that did not activate the storage until then)
will not be able to activate that storage
(without manually doing a discovery on another portal)
i don't think this is a huge problem though, the patch
still makes the situation a whole lot better
the patch looks mostly fine, some comments inline
On 10/17/23 18:19, Yuri Konotopov wrote:
> With this patch Proxmox now tries to login to all discovered portals in
> case some of them are not logged yet.
> In case of multipath configuration when initially configured portal is
> missing for some reason Proxmox don't lose iscsi storage now and can
> succesfully restore iscsi connection between reboots.
>
> Signed-off-by: Yuri Konotopov <ykonotopov@gnome.org>
> ---
> PVE/Storage.pm | 2 +-
> PVE/Storage/ISCSIPlugin.pm | 117 ++++++++++++++++++++++++++++---------
> 2 files changed, 89 insertions(+), 30 deletions(-)
>
> diff --git a/PVE/Storage.pm b/PVE/Storage.pm
> index cec3996..87755ac 100755
> --- a/PVE/Storage.pm
> +++ b/PVE/Storage.pm
> @@ -1433,7 +1433,7 @@ sub scan_iscsi {
> die "unable to parse/resolve portal address '${portal_in}'\n";
> }
>
> - return PVE::Storage::ISCSIPlugin::iscsi_discovery($portal);
> + return PVE::Storage::ISCSIPlugin::iscsi_discovery([ $portal ]);
> }
>
> sub storage_default_format {
> diff --git a/PVE/Storage/ISCSIPlugin.pm b/PVE/Storage/ISCSIPlugin.pm
> index a79fcb0..1d47f3a 100644
> --- a/PVE/Storage/ISCSIPlugin.pm
> +++ b/PVE/Storage/ISCSIPlugin.pm
> @@ -18,6 +18,9 @@ use base qw(PVE::Storage::Plugin);
> my $ISCSIADM = '/usr/bin/iscsiadm';
> $ISCSIADM = undef if ! -X $ISCSIADM;
>
> +# Example: 192.168.122.252:3260,1 iqn.2003-01.org.linux-iscsi.proxmox-nfs.x8664:sn.00567885ba8f
> +my $ISCSI_TARGET_RE = qr/^((?:$IPV4RE|\[$IPV6RE\]):\d+)\,\S+\s+(\S+)\s*$/m;
the '/m' here is unnecessary since we only match single lines anyway
> +
> sub check_iscsi_support {
> my $noerr = shift;
>
> @@ -45,11 +48,12 @@ sub iscsi_session_list {
> eval {
> run_command($cmd, errmsg => 'iscsi session scan failed', outfunc => sub {
> my $line = shift;
> -
> - if ($line =~ m/^tcp:\s+\[(\S+)\]\s+\S+\s+(\S+)(\s+\S+)?\s*$/) {
> - my ($session, $target) = ($1, $2);
> + # example: tcp: [1] 192.168.122.252:3260,1 iqn.2003-01.org.linux-iscsi.proxmox-nfs.x8664:sn.00567885ba8f (non-flash)
> + if ($line =~ m/^tcp:\s+\[(\S+)\]\s+((?:$IPV4RE|\[$IPV6RE\]):\d+)\,\S+\s+(\S+)(\s+\S+)?\s*$/) {
nit: the last capture group is not really necessary AFAICS (since we don't use it)
> + my ($session_id, $portal, $target) = ($1, $2, $3);
> # there can be several sessions per target (multipath)
> - push @{$res->{$target}}, $session;
> + my %session = ( session_id => $session_id, portal => $portal );
> + push @{$res->{$target}}, \%session;
> }
> });
> };
> @@ -68,42 +72,77 @@ sub iscsi_test_portal {
> return PVE::Network::tcp_ping($server, $port || 3260, 2);
> }
>
> -sub iscsi_discovery {
> - my ($portal) = @_;
> +sub iscsi_portals {
> + my ($target, $portal_in) = @_;
>
> check_iscsi_support ();
>
> - my $res = {};
> - return $res if !iscsi_test_portal($portal); # fixme: raise exception here?
> + my $res = [];
> + my $cmd = [$ISCSIADM, '--mode', 'node'];
> + eval {
> + run_command($cmd, outfunc => sub {
> + my $line = shift;
>
> - my $cmd = [$ISCSIADM, '--mode', 'discovery', '--type', 'sendtargets', '--portal', $portal];
> - run_command($cmd, outfunc => sub {
> - my $line = shift;
> + if ($line =~ $ISCSI_TARGET_RE) {
> + my ($portal, $portal_target) = ($1, $2);
> + if ($portal_target eq $target) {
> + push @{$res}, $portal;
> + }
> + }
> + });
> + };
>
> - if ($line =~ m/^((?:$IPV4RE|\[$IPV6RE\]):\d+)\,\S+\s+(\S+)\s*$/) {
> - my $portal = $1;
> - my $target = $2;
> - # one target can have more than one portal (multipath).
> - push @{$res->{$target}}, $portal;
> - }
> - });
> + if ($@) {
> + warn $@;
> + return [ $portal_in ];
> + }
> +
> + return $res;
> +}
> +
> +sub iscsi_discovery {
> + my ($portals) = @_;
> +
> + check_iscsi_support ();
> +
> + my $res = {};
> + for my $portal ($portals->@*) {
> + next if !iscsi_test_portal($portal); # fixme: raise exception here?
> +
> + my $cmd = [$ISCSIADM, '--mode', 'discovery', '--type', 'sendtargets', '--portal', $portal];
> + eval {
> + run_command($cmd, outfunc => sub {
> + my $line = shift;
> +
> + if ($line =~ $ISCSI_TARGET_RE) {
> + my ($portal, $target) = ($1, $2);
> + # one target can have more than one portal (multipath)
> + # and sendtargets should return all of them in single call
> + push @{$res->{$target}}, $portal;
> + }
> + });
> + };
> +
> + # In case of multipath we want to exit on any portal available
> + last if !$@;
i don't really like this, because if the discovery output did not include any line
that we could parse as a target, it still exits the loop
i'd rather check if the $res variable has picked up any targets, since
we need at least one?
e.g. something like this:
last if scalar(keys %$res) > 0;
that way we ensure that we have at least one target or we go to the next portal
does that make sense?
> + }
>
> return $res;
> }
>
> sub iscsi_login {
> - my ($target, $portal_in) = @_;
> + my ($target, $portals) = @_;
>
> check_iscsi_support();
>
> - eval { iscsi_discovery($portal_in); };
> + eval { iscsi_discovery($portals); };
> warn $@ if $@;
>
> run_command([$ISCSIADM, '--mode', 'node', '--targetname', $target, '--login']);
> }
>
> sub iscsi_logout {
> - my ($target, $portal) = @_;
> + my ($target) = @_;
>
> check_iscsi_support();
>
> @@ -133,7 +172,7 @@ sub iscsi_session_rescan {
> }
>
> foreach my $session (@$session_list) {
> - my $cmd = [$ISCSIADM, '--mode', 'session', '--sid', $session, '--rescan'];
> + my $cmd = [$ISCSIADM, '--mode', 'session', '--sid', $session->{session_id}, '--rescan'];
> eval { run_command($cmd, outfunc => sub {}); };
> warn $@ if $@;
> }
> @@ -379,14 +418,28 @@ sub activate_storage {
>
> return if !check_iscsi_support(1);
>
> - my $session = iscsi_session($cache, $scfg->{target});
> + my $sessions = iscsi_session($cache, $scfg->{target});
> + my $portals = iscsi_portals($scfg->{target}, $scfg->{portal});
> + my $do_login = !defined($sessions);
>
> - if (!defined ($session)) {
> - eval { iscsi_login($scfg->{target}, $scfg->{portal}); };
> + if (!$do_login) {
> + # We should check that sessions for all portals are available
> + my $session_portals = [ map { $_->{portal} } (@$sessions) ];
> +
> + for my $portal (@$portals) {
> + if (!grep(/^\Q$portal\E$/, @$session_portals)) {
> + $do_login = 1;
> + last;
> + }
> + }
> + }
> +
> + if ($do_login) {
> + eval { iscsi_login($scfg->{target}, $portals); };
> warn $@ if $@;
> } else {
> # make sure we get all devices
> - iscsi_session_rescan($session);
> + iscsi_session_rescan($sessions);
> }
> }
>
> @@ -396,15 +449,21 @@ sub deactivate_storage {
> return if !check_iscsi_support(1);
>
> if (defined(iscsi_session($cache, $scfg->{target}))) {
> - iscsi_logout($scfg->{target}, $scfg->{portal});
> + iscsi_logout($scfg->{target});
> }
> }
>
> sub check_connection {
> my ($class, $storeid, $scfg) = @_;
>
> - my $portal = $scfg->{portal};
> - return iscsi_test_portal($portal);
> + my $portals = iscsi_portals($scfg->{target}, $scfg->{portal});
> +
> + for my $portal (@$portals) {
> + my $result = iscsi_test_portal($portal);
> + return $result if $result;
> + }
> +
> + return 0;
> }
>
> sub volume_resize {
parent reply other threads:[~2023-10-20 9:23 UTC|newest]
Thread overview: expand[flat|nested] mbox.gz Atom feed
[parent not found: <20231017161948.139795-2-ykonotopov@gnome.org>]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=f883722d-341d-4946-9420-314578b3a8bb@proxmox.com \
--to=d.csapak@proxmox.com \
--cc=pve-devel@lists.proxmox.com \
--cc=ykonotopov@gnome.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox