# Are there issue with mass deployment related to SSH?

**URL:** <https://forum.ansible.com/t/are-there-issue-with-mass-deployment-related-to-ssh/26732>\
**Category:** Ansible Project\
**Created:** [February 14, 2018, 4:23pm UTC](https://forum.ansible.com/t/are-there-issue-with-mass-deployment-related-to-ssh/26732 "2018-02-14T16:23:49Z")\
**Posts on this page:** 6\
**Page:** 1

<div class="post-metadata">

**Author:** ![cmonty14](https://avatars.discourse-cdn.com/v4/letter/c/77aa72/32.png) [@cmonty14](https://forum.ansible.com/u/cmonty14)\
**Post date:** [February 14, 2018, 4:23pm UTC](https://forum.ansible.com/t/are-there-issue-with-mass-deployment-related-to-ssh/26732/1 "2018-02-14T16:23:49Z")

</div>

Hello!

I have a question that derives from the comparison Ansible vs. Salt.

As Ansible relies on SSH connection some people say that this has a downside for a scenario like

- mass deployment of 1000 servers

Can you confirm that there’s a severe performance impact when Ansible master must establish SSH connection to 1000+ servers at the same time (for mass deployment)?

THX

---

<div class="post-metadata">

**Author:** ![Andrew\_Latham](https://avatars.discourse-cdn.com/v4/letter/a/ed8c4c/32.png) [@Andrew\_Latham](https://forum.ansible.com/u/Andrew_Latham)\
**Post date:** [February 14, 2018, 6:35pm UTC](https://forum.ansible.com/t/are-there-issue-with-mass-deployment-related-to-ssh/26732/2 "2018-02-14T18:35:55Z")

</div>

I suggest you start with reading about [http://docs.ansible.com/ansible/latest/intro\_configuration.html#forks](http://docs.ansible.com/ansible/latest/intro_configuration.html#forks)

---

<div class="post-metadata">

**Author:** ![cmonty14](https://avatars.discourse-cdn.com/v4/letter/c/77aa72/32.png) [@cmonty14](https://forum.ansible.com/u/cmonty14)\
**Post date:** [February 15, 2018, 10:22am UTC](https://forum.ansible.com/t/are-there-issue-with-mass-deployment-related-to-ssh/26732/3 "2018-02-15T10:22:06Z")

</div>

This means there’s no drawback with regards to performance when using Ansible in a landscape with +1000 servers?

---

<div class="post-metadata">

**Author:** ![Manuel\_Schmidt](https://avatars.discourse-cdn.com/v4/letter/m/ea666f/32.png) [@Manuel\_Schmidt](https://forum.ansible.com/u/Manuel_Schmidt)\
**Post date:** [February 15, 2018, 11:15am UTC](https://forum.ansible.com/t/are-there-issue-with-mass-deployment-related-to-ssh/26732/4 "2018-02-15T11:15:48Z")

</div>

Hey,

combined with the aforementioned forks configuration it all depends on your Controll-Hosts capability to handle that many connections at the same time. I dont think that ansible would be the bottleneck here but the saturation of your network.

Regards

---

<div class="post-metadata">

**Author:** ![Andrew\_Latham](https://avatars.discourse-cdn.com/v4/letter/a/ed8c4c/32.png) [@Andrew\_Latham](https://forum.ansible.com/u/Andrew_Latham)\
**Post date:** [February 15, 2018, 3:37pm UTC](https://forum.ansible.com/t/are-there-issue-with-mass-deployment-related-to-ssh/26732/5 "2018-02-15T15:37:30Z")

</div>

Correct, Ansible is not the limit. Saltstack has guidelines around the same bottle necks for its ssh only solution.

---

<div class="post-metadata">

**Author:** ![cmonty14](https://avatars.discourse-cdn.com/v4/letter/c/77aa72/32.png) [@cmonty14](https://forum.ansible.com/u/cmonty14)\
**Post date:** [February 16, 2018, 2:19pm UTC](https://forum.ansible.com/t/are-there-issue-with-mass-deployment-related-to-ssh/26732/6 "2018-02-16T14:19:33Z")

</div>

Could you please share some info how to mitigate the risk of a bottleneck caused by SSH and/or the underlying network?