Failed to start gluster-blockd.service: Unit gluster-block-target.service has a bad unit file setting

[Flavio2005]

Hello,
I am new to gluster 9 and also gluster-block. I have no idea if this is a bug or just an issue on my side.
My general goal is to build a volume to be exposed in iSCSI in HA.
The failover management will entrust it to the double connection that ESXi allows to make.

With three hosts, Fedora 34, fresh from installation, following the list of commands below, I created three nodes: two in replicate and one in arbiter.

Under clean Fedora34 i run this command

dnf -y install git
git clone https://github.com/gluster/gluster-block.git
cd gluster-block/
dnf install gcc autoconf automake make file libtool libuuid-devel json-c-devel glusterfs-api-devel glusterfs-server tcmu-runner targetcli
dnf install rpcgen libtirpc-devel
./autogen.sh && ./configure && make -j install
gluster-blockd --help
dnf -y install glusterfs-server
dnf -y install glusterfs glusterfs-fuse
systemctl enable --now glusterd

Version

glusterfs 9.2
gluster-block (0.5.1)
NAME=Fedora
VERSION="34 (Server Edition)"
Kernel: Linux 5.11.12-300.fc34.x86_64

gluster peer status

gluster peer status
Number of Peers: 2

Hostname: GluNodo13
Uuid: 678e1de0-95e2-446c-b888-2ebb17b407aa
State: Peer in Cluster (Connected)

Hostname: GluNodo11
Uuid: 8bec5d5c-cf07-4580-b174-eea837514e21
State: Peer in Cluster (Connected)

gluster volume info

Volume Name: hosting-volume
Type: Replicate
Volume ID: a6026b8f-aac0-4bd6-8404-c1c516fc7022
Status: Started
Snapshot Count: 0
Number of Bricks: 1 x (2 + 1) = 3
Transport-type: tcp
Bricks:
Brick1: GluNodo11:/data/brick1/gv0
Brick2: GluNodo12:/data/brick1/gv0
Brick3: GluNodo13:/data/brick1/gv0 (arbiter)

Kind of issue

If I try to create a volume with gluster-block with this command
# gluster-block create hosting-volume/block-volume ha 3 192.168.11.151,192.168.11.152,192.168.11.153 1GiB client create failed: RPC: Remote system error - No such file or directory, unix addr /var/run/gluster-blockd.socket

# systemctl start gluster-blockd Failed to start gluster-blockd.service: Unit gluster-block-target.service has a bad unit file setting. See system logs and 'systemctl status gluster-blockd.service' for details.

I have tried in many ways but I cannot overcome the obstacle.
Can you please tell me where I'm wrong?

# systemctl status gluster-blockd
○ gluster-blockd.service - Gluster block storage utility
Loaded: loaded (/usr/local/lib/systemd/system/gluster-blockd.service; disabled; vendor preset: disabled)
Active: inactive (dead)

If I try to tun
# systemctl start gluster-blockd
Failed to start gluster-blockd.service: Unit gluster-block-target.service has a bad unit file setting.
See system logs and 'systemctl status gluster-blockd.service' for details.
gluster-block-cli.log
gluster-blockd.log

# gluster vol set hosting-volume group gluster-block
volume set: success

follows one of the many unsuccessful attempts
# gluster-block create hosting-volume/gluster-block ha 3 192.168.11.151,192.168.11.152,192.168.11.153 5GiB
client create failed: RPC: Remote system error - No such file or directory, unix addr /var/run/gluster-blockd.socket

Any idea what can be the problem and how I can solve it?
What else can I check for the problem?
Thank you !

[lxbsz]

It seems the gluster-block-target.service's setting is not compatible with the Fedora 34:

# systemctl start gluster-blockd Failed to start gluster-blockd.service: Unit gluster-block-target.service has a bad unit file setting. See system logs and 'systemctl status gluster-blockd.service' for details.

I have never run this on Fedora34 yet.
@pkalever Any idea ?

[Flavio2005]

Dear @lxbsz
based on your experiences, which operating system / release, would you choose for a fresh Gluster 9.2 installation with gluster-block / targetcli-fb / tcmu-runner?

[pkalever]

Dear @lxbsz
based on your experiences, which operating system / release, would you choose for a fresh Gluster 9.2 installation with gluster-block / targetcli-fb / tcmu-runner?

Just using the support version of gluster is good. But I don't think that is related to this issue here.

BTW - did you fix this issue or are you still waiting on help?

[Flavio2005]

I have not solved That's why I was looking for direct contact I also signed up on your site and wrote to you but, I have not had any feedback :-( In these days I am also without connectivity but as soon as they fix it for me, I will try and post what I launch, what I get in response thanks for your attention Flavio G Il giorno gio 10 giu 2021 alle ore 13:39 Prasanna Kumar Kalever < ***@***.***> ha scritto:

…

Dear @lxbsz <https://github.com/lxbsz> based on your experiences, which operating system / release, would you choose for a fresh Gluster 9.2 installation with gluster-block / targetcli-fb / tcmu-runner? Just using the support version of gluster is good. But I don't think that is related to this issue here. BTW - did you fix this issue or are you still waiting on help? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#287 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ARIONMLA6XRBP2FKCYPBRZ3TSCP6HANCNFSM45G4OW4Q> .

[pkalever]

@Flavio2005 By now you must be knowing that the project is in a maintenance-only state.

The logs you pointed didn't give any lead. To understand the issue better, I will consider reproducing this on fedora 34, but please expect that to take some time (maybe by next week)

Thanks!

[rudraitkelwar]

Hi any update on this facing same issue
using ubuntu 22.04