ERR! Get "http://127.0.0.1:4001//status": dial tcp 127.0.0.1:4001: connect: connection refused

39 views
Skip to first unread message

Yannick G

unread,
Jan 7, 2022, 12:52:26 PM1/7/22
to rqlite
Hello folks,

I tested rqlite via docker distribution and it works nice. Then I tried to launch 3 fedora nodes in VMWare and built the cluster. I have an error when I enter the command "rqlite" :

ERR! Get "http://127.0.0.1:4001//status": dial tcp 127.0.0.1:4001: connect: connection refused

I guess i've done something stupid or forgot to do something but what ?

Philip O'Toole

unread,
Jan 7, 2022, 1:05:56 PM1/7/22
to rql...@googlegroups.com
You need to tell the CLI where to find a node, as per the manual.


  -H, --host[=127.0.0.1]
      rqlited host address

  -p, --port[=4001]
      rqlited host port

--
You received this message because you are subscribed to the Google Groups "rqlite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rqlite+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/rqlite/e45b86af-7dad-4c63-a983-be6209b0cbe4n%40googlegroups.com.

Yannick G

unread,
Jan 7, 2022, 1:22:06 PM1/7/22
to rql...@googlegroups.com
Actually, I run the command from the node CLI, sorry i did not mention. But thanks a lot, your answer lead me to my dumbness :)

I affected the nodes a bridged IP on local network and rqlite -H 10.0.4.x is working perfectly.

TY

Koray

unread,
Jan 9, 2022, 9:10:59 AM1/9/22
to rqlite
When you set up clusters with rqlite, when you reboot the servers, the cluster doesn't get up by itself and I get this error you got. I am using the following commands while removing the cluster.

Where do you think I went wrong? Can you post the commands you used while removing the cluster?

./rqlited -node-id 1 -http-addr 10.10.10.88:4001 -raft-addr 10.10.10.88:4002 ~/node

./rqlited -node-id 2 -http-addr 10.10.10.89:4001 -raft-addr 10.10.10.89:4002 -join http://10.10.10.88:4001 ~/node

./rqlited -node-id 3 -http-addr 10.10.10.90:4001 -raft-addr 10.10.10.90:4002 -join http://10.10.10.88:4001 ~/node


7 Ocak 2022 Cuma tarihinde saat 21:22:06 UTC+3 itibarıyla yannic...@gmail.com şunları yazdı:

Philip O'Toole

unread,
Jan 9, 2022, 11:59:58 AM1/9/22
to rql...@googlegroups.com
On Sun, Jan 9, 2022 at 9:11 AM Koray <koray...@gmail.com> wrote:
When you set up clusters with rqlite, when you reboot the servers, the cluster doesn't get up by itself

Restarting a cluster absolutely works. It is tested here:


If you're getting an error connecting to the cluster it's not because rqlite can't do a simple restart, something else is going on.

Philip
 

Philip O'Toole

unread,
Jan 9, 2022, 12:18:02 PM1/9/22
to rql...@googlegroups.com
Restarting a cluster where a quorum, or more, of nodes (in your case a quorum is 2 nodes) Raft IP addresses change will not start correctly, however. This is because the nodes have no way of finding each other at their new network addresses. If you must change a quorum of nodes' address, only change one address at a time, or use the recovery process.

Koray

unread,
Jan 9, 2022, 12:39:54 PM1/9/22
to rqlite
No, you didn't understand me. I'm setting up an rqlite clsuter by following the instructions on the rqlite github page and entering the commands I just mentioned in my answer. However, the commands I entered work directly in the terminal. It does not run as a service or in the background. Therefore, when the terminal is closed, the command starting with ./rqlited stops. Why is rqlite not running as a service? How do you make it run in the background?

9 Ocak 2022 Pazar tarihinde saat 20:18:02 UTC+3 itibarıyla Philip O'Toole şunları yazdı:

Philip O'Toole

unread,
Jan 10, 2022, 5:30:29 PM1/10/22
to rql...@googlegroups.com
On Sun, Jan 9, 2022 at 12:39 PM Koray <koray...@gmail.com> wrote:
No, you didn't understand me. I'm setting up an rqlite clsuter by following the instructions on the rqlite github page and entering the commands I just mentioned in my answer. However, the commands I entered work directly in the terminal. It does not run as a service or in the background. Therefore, when the terminal is closed, the command starting with ./rqlited stops. Why is rqlite not running as a service? How do you make it run in the background?

There are many ways to run rqlite -- Docker, rkt, as a service, within a VM etc. The specific way to run rqlite is left to the end-user (like yourself). If you want to run it as a service, then you need to write the service config file that best suits your needs. 


9 Ocak 2022 Pazar tarihinde saat 20:18:02 UTC+3 itibarıyla Philip O'Toole şunları yazdı:
Restarting a cluster where a quorum, or more, of nodes (in your case a quorum is 2 nodes) Raft IP addresses change will not start correctly, however. This is because the nodes have no way of finding each other at their new network addresses. If you must change a quorum of nodes' address, only change one address at a time, or use the recovery process.

On Sun, Jan 9, 2022 at 11:59 AM Philip O'Toole <oto...@google.com> wrote:


On Sun, Jan 9, 2022 at 9:11 AM Koray <koray...@gmail.com> wrote:
When you set up clusters with rqlite, when you reboot the servers, the cluster doesn't get up by itself

Restarting a cluster absolutely works. It is tested here:


If you're getting an error connecting to the cluster it's not because rqlite can't do a simple restart, something else is going on.

Philip
 

--
You received this message because you are subscribed to the Google Groups "rqlite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rqlite+un...@googlegroups.com.
Reply all
Reply to author
Forward
0 new messages