Hi all, This patch addresses excessive "connection lost" and "connection reset" log spam on iSCSI targets caused by Proxmox storage monitoring performing TCP connection tests every 10 seconds, even when iSCSI sessions are active.
The issue appears as continuous log entries on iSCSI targets: "ctld[PID]: 192.168.1.XX: read: connection lost" every 10 seconds on each node. Others have experienced this, as seen in forum posts: iSCSI Reconnecting every 10 seconds to FreeNAS solution | Proxmox Support Forum<https://forum.proxmox.com/threads/iscsi-reconnecting-every-10-seconds-to-freenas-solution.21205/> freenas ctld: read: connection lost | TrueNAS Community<https://www.truenas.com/community/threads/freenas-ctld-read-connection-lost.45513/> Changes: - Support hostnames in portal addresses (not just IP addresses) by updating ISCSI_TARGET_RE and session parsing regex - Skip TCP connection tests in check_connection when active sessions exist - Skip storage activation when sessions are already established This reduces iSCSI target log spam from continuous entries to only legitimate connection issues, while maintaining monitoring functionality. Tested with: - IPv4 and IPv6 portals - Hostname based portals (like nas.example.com) - Multiple Proxmox nodes in cluster 8.2.4 + 8.3.0 (Requires latest ISCSIPlugin.pm. taken from raw.githubusercontent.com/proxmox/pve-storage/823707a7ac99eafe345d9131d10837ebefbb9078/src/PVE/Storage/ISCSIPlugin.pm<https://raw.githubusercontent.com/proxmox/pve-storage/823707a7ac99eafe345d9131d10837ebefbb9078/src/PVE/Storage/ISCSIPlugin.pm>) - TrueNAS and Linux iSCSI targets - Can apply from /usr/share/perl5/PVE/Storage# patch -p0 < /path/to/iscsifix.patch --- PVE/Storage/ISCSIPlugin.pm | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) --- ISCSIPlugin.pm.new-upstream 2025-06-19 16:24:09.220675514 -0500 +++ ISCSIPlugin.pm 2025-06-19 16:24:42.265694074 -0500 @@ -33,7 +33,7 @@ } # 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*$/; +my $ISCSI_TARGET_RE = qr/^(\S+:\d+)\,\S+\s+(\S+)\s*$/; sub iscsi_session_list { assert_iscsi_support(); @@ -49,7 +49,7 @@ my $line = shift; # example: tcp: [1] 192.168.122.252:3260,1 iqn.2003-01.org.linux-iscsi.proxmox-nfs.x8664:sn.00567885ba8f (non-flash) if ($line =~ - m/^tcp:\s+\[(\S+)\]\s+((?:$IPV4RE|\[$IPV6RE\]):\d+)\,\S+\s+(\S+)\s+\S+?\s*$/ + m/^tcp:\s+\[(\S+)\]\s+(\S+:\d+)\,\S+\s+(\S+)\s+\S+?\s*$/ ) { my ($session_id, $portal, $target) = ($1, $2, $3); # there can be several sessions per target (multipath) @@ -480,6 +480,10 @@ sub activate_storage { my ($class, $storeid, $scfg, $cache) = @_; + # Skip if session already exists + my $existing_sessions = iscsi_session($cache, $scfg->{target}); + return 1 if defined($existing_sessions) && scalar(@$existing_sessions) > 0; + return if !assert_iscsi_support(1); my $sessions = iscsi_session($cache, $scfg->{target}); @@ -618,6 +622,11 @@ sub check_connection { my ($class, $storeid, $scfg) = @_; my $cache = {}; + # Check for active session first + my $sessions = iscsi_session_list(); + return 1 if defined($sessions->{$scfg->{target}}); + + # Only check portals if no active session my $portals = iscsi_portals($scfg->{target}, $scfg->{portal}); for my $portal (@$portals) { --- Regards, Stelios Vailakakis _______________________________________________ pve-devel mailing list pve-devel@lists.proxmox.com https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
