# Datacenter stanzas

**URL:** <https://discuss.aerospike.com/t/datacenter-stanzas/10159>\
**Category:** XDR (Cross Data Center Replication)\
**Created:** [April 21, 2023, 4:20pm UTC](https://discuss.aerospike.com/t/datacenter-stanzas/10159 "2023-04-21T16:20:56Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![Jeff.MacDonald](https://avatars.discourse-cdn.com/v4/letter/j/e9bcb4/32.png) [@Jeff.MacDonald](https://discuss.aerospike.com/u/Jeff.MacDonald)\
**Post date:** [April 21, 2023, 4:20pm UTC](https://discuss.aerospike.com/t/datacenter-stanzas/10159/1 "2023-04-21T16:20:56Z")

</div>

This is for 4.9

Lets imagine I have 2 DCs. DC1 and DC2.

Lets imagine that DC2 foobar has 20 servers.

And that DC1 has a stanza that looks like this

```auto
datacenter foobar {
  tls-node foobar-1.xx.foo.com DC1 4333
}

```

What are the implications of the stanza not containing all of DC2’s 20 nodes?

Will DC1 just reach out to foobar-1, and foobar-1 will reply with a list of it’s nodes, and thus the real issue is that if foobar-1 was down, that DC1 would not know “who” to ask about nodes?

Jeff.

---

<div class="post-metadata">

**Author:** ![Albot](https://sea1.discourse-cdn.com/flex019/user_avatar/discuss.aerospike.com/albot/32/2076_2.png) [@Albot](https://discuss.aerospike.com/u/Albot)\
**Post date:** [April 22, 2023, 6:34pm UTC](https://discuss.aerospike.com/t/datacenter-stanzas/10159/2 "2023-04-22T18:34:11Z")

</div>

> In Aersopike 5.0, superseded by [`node-address-port`](https://docs.aerospike.com/reference/configuration#node-address-port) Multiple nodes can be specified. Used as a seed list, similar to dc-node-address-port when not using TLS.

> Will DC1 just reach out to foobar-1, and foobar-1 will reply with a list of it’s nodes, and thus the real issue is that if foobar-1 was down, that DC1 would not know “who” to ask about nodes?

That’s correct.

This works the same as the client. It is a seed to connect to any node in the cluster, and if successful it will ask the node it connected to for its cluster members. If more than 1 entry is present, it should traverse the list 1-by-1 until it finds an alive node.

If no alive nodes are found I think it will treat the DC as being down, and keep trying until it is back online. In v4.x you may have data loss, since it uses a digestlog ring buffer.

---

<div class="post-metadata">

**Author:** ![system](https://sea1.discourse-cdn.com/flex019/user_avatar/discuss.aerospike.com/system/32/2274_2.png) [@system](https://discuss.aerospike.com/u/system)\
**Post date:** [April 21, 2024, 6:34pm UTC](https://discuss.aerospike.com/t/datacenter-stanzas/10159/3 "2024-04-21T18:34:40Z")

</div>

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.