On Wed, Sep 10, 2025 at 10:34 AM Robert Sander <r.sander@heinlein-support.de> wrote:
Am 9/10/25 um 10:27 schrieb Robert Sander:
So there must be a slight difference in the config of the subsystems / namespaces between v19 and v20.
After removing both namespaces (RBD ns01 and ns02) from the old subsystem and adding them again the initiator sees them:
# nvme list Node Generic SN Model Namespace Usage Format FW Rev --------------------- --------------------- -------------------- ---------------------------------------- ---------- -------------------------- ---------------- -------- /dev/nvme1n1 /dev/ng1n1 Ceph30844860597194 Ceph bdev Controller 0x1 107.37 GB / 107.37 GB 512 B + 0 B 25.05 /dev/nvme1n2 /dev/ng1n2 Ceph30844860597194 Ceph bdev Controller 0x2 214.75 GB / 214.75 GB 512 B + 0 B 25.05 /dev/nvme5n1 /dev/ng5n1 Ceph44670645393042 Ceph bdev Controller 0x1 161.06 GB / 161.06 GB 512 B + 0 B 25.05
# lsblk NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINTS sda 8:0 0 60G 0 disk ├─sda1 8:1 0 1G 0 part /boot/efi ├─sda2 8:2 0 2G 0 part /boot └─sda3 8:3 0 56.9G 0 part ├─system-swap 252:0 0 100M 0 lvm [SWAP] └─system-root 252:1 0 56.8G 0 lvm /var/lib/containers/storage/overlay / sr0 11:0 1 4M 0 rom nvme1n1 259:4 0 100G 0 disk └─nvme1n1p1 259:5 0 100G 0 part nvme5n1 259:9 0 150G 0 disk nvme1n2 259:14 0 200G 0 disk └─nvme1n2p1 259:16 0 200G 0 part
Hi Robert, I'm forwarding some questions on behalf of Aviv as the reproduction attempt wasn't fruitful: - Were you using quay.ceph.io/ceph-ci/ceph:squid-nvmeof container image before upgrading from squid to tentacle? - If so, what GW version was deployed? Was it 1.4.21? - Can you provide the output of "nvme-gw show" command? - Can you provide ceph-mon log from the leader monitor/node? - Can you provide some GW logs? Thanks, Ilya