[Solved] GlusterFS - Transport endpoint is not connected

[coppolino97]

Hi folks,
I am running three virtual machine in the same LAN (Network 192.168.88.0/24)
These three virtual servers are running Debian 11 (I personally installed all updates available).
Their goal is a glusterfs deployment:

• Two GlusterFS node as storage (their hostname are docker01 and docker02)

• One GlusterFS node as arbiter (its hostname is swarm01)

GlusterFS installed is # gluster --version ==> glusterfs 11.1
I already edited /etc/hosts file to create DNS record with success.

Code: Select all

root@docker01:~# tail -f /var/log/glusterfs/mnt.log 
[2024-02-13 22:16:23.623527 +0000] W [fuse-bridge.c:1403:fuse_attr_cbk] 0-glusterfs-fuse: 1020: LOOKUP() / => -1 (Transport endpoint is not connected)
[2024-02-13 22:16:23.623571 +0000] W [fuse-bridge.c:1403:fuse_attr_cbk] 0-glusterfs-fuse: 1021: LOOKUP() / => -1 (Transport endpoint is not connected)
[2024-02-13 22:16:23.792071 +0000] W [fuse-bridge.c:1403:fuse_attr_cbk] 0-glusterfs-fuse: 1022: LOOKUP() / => -1 (Transport endpoint is not connected)
[2024-02-13 22:16:23.792118 +0000] W [fuse-bridge.c:1403:fuse_attr_cbk] 0-glusterfs-fuse: 1023: LOOKUP() / => -1 (Transport endpoint is not connected)
[2024-02-13 22:16:23.908092 +0000] W [fuse-bridge.c:1403:fuse_attr_cbk] 0-glusterfs-fuse: 1024: LOOKUP() / => -1 (Transport endpoint is not connected)
[2024-02-13 22:16:23.908136 +0000] W [fuse-bridge.c:1403:fuse_attr_cbk] 0-glusterfs-fuse: 1025: LOOKUP() / => -1 (Transport endpoint is not connected)
[2024-02-13 22:16:26.241343 +0000] E [MSGID: 114058] [client-handshake.c:946:client_query_portmap_cbk] 0-storegfs-client-1: failed to get the port number for remote subvolume. Please run gluster volume status on server to see if brick process is running [] 
[2024-02-13 22:16:29.696653 +0000] W [fuse-bridge.c:1403:fuse_attr_cbk] 0-glusterfs-fuse: 1044: LOOKUP() / => -1 (Transport endpoint is not connected)
[2024-02-13 22:16:45.654162 +0000] I [socket.c:835:__socket_shutdown] 0-storegfs-client-2: intentional socket shutdown(6)
[2024-02-13 22:17:02.259513 +0000] I [socket.c:835:__socket_shutdown] 0-storegfs-client-0: intentional socket shutdown(6)
q
^C
root@docker01:~# gluster volume status
Status of volume: storegfs
Gluster process                             TCP Port  RDMA Port  Online  Pid
------------------------------------------------------------------------------
Brick docker01:/mnt/disk1/br0               N/A       N/A        N       N/A  
Brick docker02:/mnt/disk2/br0               N/A       N/A        N       N/A  
Brick swarm01:/mnt/disk3/br0                N/A       N/A        N       N/A  
Self-heal Daemon on localhost               N/A       N/A        Y       906  
Self-heal Daemon on swarm01                 N/A       N/A        Y       940  
Self-heal Daemon on docker02                N/A       N/A        Y       1765 
 
Task Status of Volume storegfs
------------------------------------------------------------------------------
There are no active volume tasks

These three virtual server can ping each other, iptables on these machines haven't any rule set.
Gluster service is running, but I got error message concerning
I collected this data from the first of three VM in my GlusterFS deployment.

I create volume with the following command (I didn't got any error).

Code: Select all

cluster volume create storegfs replica 3  arbiter 1 docker01:/mnt/disk1/br0 docker02:/mnt/disk2/br0 swarm01:/mnt/disk3/br0

In the end I would use this GlusterFS deployment with Docker Swarm.
I never tried Docker Swarm in this deployment due to it is not properly working GlusterFS now.

Any suggestion for me?

[coppolino97]

I will share you more details.
Below output of "cluster volume info"

Code: Select all

Volume Name: storegfs
Type: Replicate
Volume ID: b03c686a-46cd-458c-8209-d738d5d2e208
Status: Started
Snapshot Count: 0
Number of Bricks: 1 x (2 + 1) = 3
Transport-type: tcp
Bricks:
Brick1: docker01:/mnt/disk1/br0
Brick2: docker02:/mnt/disk2/br0
Brick3: swarm01:/mnt/disk3/br0 (arbiter)
Options Reconfigured:
performance.client-io-threads: off
transport.address-family: inet
storage.fips-mode-rchecksum: on
cluster.granular-entry-heal: on
cluster.self-heal-daemon: enable

Thanks a lot in advance

[coppolino97]

Hello,
today I tried again concerning GlusterFS deployment.
I created new virtual machine in the same LAN (all of them are running Debian 11).
I repeated more or less same deployment steps, despite of I install using "apt install glusterfs-server" without installing before external gpg key or repo.
In this way I got GlusterFS 9 instead of GlusterFS 11.

In this moment my home lab deployment is running properly.
See you soon

[Aki]

Hello,

I repeated more or less same deployment steps, despite of I install using "apt install glusterfs-server" without installing before external gpg key or repo. In this way I got GlusterFS 9 instead of GlusterFS 11.
In this moment my home lab deployment is running properly.

Thanks for updating the discussion. So you installed the required programs from a non-Debian repository first, didn't you?

[coppolino97]

Hello @Aki
Yes I did it the first time.
I installed GlusterFS software using external repo (GlusterFS). I did this "wrong" step following a shared guide online.

Thanks