ISCSI Keeps Trying Unused Portals

[uclemmer]

Hello,

PVE 9.0

We have a Dell PowerVault ME4024 with two controllers A & B. Each controller has four ports and our Volume (DevDataVault) is configured to use only the last port on each controller:

Controller A
10.0.0.1 - Not used
10.0.0.2 - Not used
10.0.0.3 - Not used
10.0.0.4 - Mapped to DevDataVault

Controller B
172.16.0.1 - Not used
172.16.0.2 - Not used
172.16.0.3 - Not used
172.16.0.4 - Mapped to DevDataVault

I have configured my PVE host to use the ISCSI drives and multipathing as in the Proxmox official documentation.

However my log files are getting cluttered with attempts to connect to the ISCSI drive through the unused IP addresses.

Oct 02 13:11:42 PVEHost1 iscsid[2605]: connect to 10.0.0.3:3260 failed (No route to host)
Oct 02 13:11:44 PVEHost1 iscsid[2605]: connect to 172.16.0.1:3260 failed (No route to host)
Oct 02 13:11:44 PVEHost1 iscsid[2605]: connect to 172.16.0.2:3260 failed (No route to host)
Oct 02 13:11:44 PVEHost1 iscsid[2605]: connect to 10.0.0.1:3260 failed (No route to host)
Oct 02 13:11:44 PVEHost1 iscsid[2605]: connect to 10.0.0.2:3260 failed (No route to host)
Oct 02 13:11:44 PVEHost1 iscsid[2605]: connect to 172.16.0.3:3260 failed (No route to host)
Oct 02 13:11:45 PVEHost1 iscsid[2605]: Connection-1:0 to [target: [*iqn*], portal: 10.0.0.3,3260] through [iface: default] is shutdown.
Oct 02 13:11:46 PVEHost1 pvestatd[2403]: command '/usr/bin/iscsiadm --mode node --targetname [*iqn*] --login' failed: exit code 15
Oct 02 13:11:46 PVEHost1 pvestatd[2403]: status update time (19.164 seconds)
Oct 02 13:11:47 PVEHost1 iscsid[2605]: Connection-1:0 to [target: [*iqn*] portal: 172.16.0.1,3260] through [iface: default] is shutdown.
Oct 02 13:11:47 PVEHost1 iscsid[2605]: Connection-1:0 to [target: [*iqn*], portal: 172.16.0.2,3260] through [iface: default] is shutdown.
Oct 02 13:11:47 PVEHost1 iscsid[2605]: Connection-1:0 to [target: [*iqn*], portal: 10.0.0.1,3260] through [iface: default] is shutdown.
Oct 02 13:11:47 PVEHost1 iscsid[2605]: Connection-1:0 to [target: [*iqn*], portal: 10.0.0.2,3260] through [iface: default] is shutdown.
Oct 02 13:11:47 PVEHost1 iscsid[2605]: Connection-1:0 to [target: [*iqn*], portal: 172.16.0.3,3260] through [iface: default] is shutdown.

I have tried multiple methods to disable this ports and get rid of these unneeded attempts. What am I missing?

Thanks!

 

[bbgeek17]

Hi @uclemmer , welcome to the forum.

You are not missing anything. The flow of the iSCSI protocol is that the Target (ME) provides a packet that contains all Portals that it wants to advertise. The Initiator (PVE) can/will attempt to login to all those. As far as Initiator knows - Target advertises valid IPs.
One way to solve it is to disable advertising invalid IPs on ME side. If you are unable to find an option for it, perhaps Dell support can assist.

This article may be helpful to you: https://kb.blockbridge.com/technote/proxmox-lvm-shared-storage/
In particular this section: https://kb.blockbridge.com/technote...nderstand-multipath-reporting-for-your-device

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox