# 2 server set up using manual installation of YugaByte with replication factor of 2

**URL:** <https://forum.yugabyte.com/t/2-server-set-up-using-manual-installation-of-yugabyte-with-replication-factor-of-2/2702>\
**Category:** General\
**Created:** [June 19, 2024, 12:55pm UTC](https://forum.yugabyte.com/t/2-server-set-up-using-manual-installation-of-yugabyte-with-replication-factor-of-2/2702 "2024-06-19T12:55:34Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![re4nightwing](https://avatars.discourse-cdn.com/v4/letter/r/5daacb/32.png) [@re4nightwing](https://forum.yugabyte.com/u/re4nightwing)\
**Post date:** [June 19, 2024, 12:55pm UTC](https://forum.yugabyte.com/t/2-server-set-up-using-manual-installation-of-yugabyte-with-replication-factor-of-2/2702/1 "2024-06-19T12:55:34Z")

</div>

Hi, I’m new to Yugabytes. Please help me to figure this out.

I have 2 CentOS servers, where I wanted to deploy YugaByteDB using the generic binary v2.14 on both servers and setup as a 2-node cluster with replication as 2.  
First of all is it possible to do this with only 2 servers because the leader election needs 3 instances?  
I tried deploying the following commands,

```bash
# VM 1
yb-master \
   --master_addresses <vm-01>:7100,<vm-02>:7100 \
   --rpc_bind_addresses <vm-01> \
   --fs_data_dirs "/home/yugabyte/yb-data" > /home/yugabyte/yb-master.out 2>&1 &

# VM 2 
yb-master \
    --master_addresses <vm-01>:7100,<vm-02>:7100 
    --rpc_bind_addresses <vm-02> 
    --fs_data_dirs "/home/yugabyte/yb-data" > /home/yugabyte/yb-master.out 2>&1 &

# VM 1
yb-tserver 
    --tserver_master_addrs <vm-01>:7100,<vm-02>:7100 
    --rpc_bind_addresses <vm-01> 
    --fs_data_dirs "/home/yugabyte/yb-data" > /home/yugabyte/yb-tserver.out 2>&1 &

# VM 2
yb-tserver 
    --tserver_master_addrs <vm-01>:7100,<vm-02>:7100 
    --rpc_bind_addresses <vm-02> 
    --fs_data_dirs "/home/yugabyte/yb-data" > /home/yugabyte/yb-tserver.out 2>&1 &

```

But for some reason, VM1’s yb-master fails on VM1’s yb-tserver deployment. And again VM2 can run **both** at once.

VM2’s UI is accessible at port 7000 and shows this error.

 ![image](https://canada1.discourse-cdn.com/flex027/uploads/yugabyte/original/1X/a7e6e134de9451946ac37d909e32c83b829beea9.png)

Any help regarding this 2 node deployment is highly appriciated.

---

<div class="post-metadata">

**Author:** ![dorian\_yugabyte](https://yyz1.discourse-cdn.com/flex027/user_avatar/forum.yugabyte.com/dorian_yugabyte/32/206_2.png) [@dorian\_yugabyte](https://forum.yugabyte.com/u/dorian_yugabyte)\
**Post date:** [June 19, 2024, 2:23pm UTC](https://forum.yugabyte.com/t/2-server-set-up-using-manual-installation-of-yugabyte-with-replication-factor-of-2/2702/2 "2024-06-19T14:23:01Z")

</div>

Hi @re4nightwing

2 nodes deployment is not supported. The minimum is 3. Please read [Deployment checklist for YugabyteDB clusters | YugabyteDB Docs](https://docs.yugabyte.com/preview/deploy/checklist/#replication)

---

<div class="post-metadata">

**Author:** ![dorian\_yugabyte](https://yyz1.discourse-cdn.com/flex027/user_avatar/forum.yugabyte.com/dorian_yugabyte/32/206_2.png) [@dorian\_yugabyte](https://forum.yugabyte.com/u/dorian_yugabyte)\
**Post date:** [June 19, 2024, 2:56pm UTC](https://forum.yugabyte.com/t/2-server-set-up-using-manual-installation-of-yugabyte-with-replication-factor-of-2/2702/3 "2024-06-19T14:56:22Z")

</div>

> [@re4nightwing](#):
>
> But for some reason, VM1’s yb-master fails on VM1’s yb-tserver deployment.

In this case we have to look at logs of yb-master2 to know why it’s not starting [Inspect logs | YugabyteDB Docs](https://docs.yugabyte.com/preview/troubleshoot/nodes/check-logs/)

> [@re4nightwing](#):
>
> where I wanted to deploy YugaByteDB using the generic binary v2.14

It’s usually better to use a newer stable release like 2.20.

---

<div class="post-metadata">

**Author:** ![re4nightwing](https://avatars.discourse-cdn.com/v4/letter/r/5daacb/32.png) [@re4nightwing](https://forum.yugabyte.com/u/re4nightwing)\
**Post date:** [June 20, 2024, 3:42am UTC](https://forum.yugabyte.com/t/2-server-set-up-using-manual-installation-of-yugabyte-with-replication-factor-of-2/2702/4 "2024-06-20T03:42:00Z")

</div>

I understand but is it possible to deploy a yugabytedb cluster using 2 nodes + 1 witness node (container)?

---

<div class="post-metadata">

**Author:** ![dorian\_yugabyte](https://yyz1.discourse-cdn.com/flex027/user_avatar/forum.yugabyte.com/dorian_yugabyte/32/206_2.png) [@dorian\_yugabyte](https://forum.yugabyte.com/u/dorian_yugabyte)\
**Post date:** [June 20, 2024, 10:03am UTC](https://forum.yugabyte.com/t/2-server-set-up-using-manual-installation-of-yugabyte-with-replication-factor-of-2/2702/5 "2024-06-20T10:03:16Z")

</div>

No, it is not possible. The 3rd node has to be a normal node. When using the default synchronous replication model.