# Nginx.http.sock bind failed after reboot

**URL:** <https://meta.discourse.org/t/nginx-http-sock-bind-failed-after-reboot/29550>\
**Category:** Self-hosting\
**Created:** [June 2, 2015, 1:17pm UTC](https://meta.discourse.org/t/nginx-http-sock-bind-failed-after-reboot/29550 "2015-06-02T13:17:21Z")\
**Posts on this page:** 1\
**Showing post:** 6

<div class="post-metadata">

**Author:** ![rriemann](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/rriemann/32/115238_2.png) [@rriemann](https://meta.discourse.org/u/rriemann)\
**Post date:** [June 9, 2015, 1:36pm UTC](https://meta.discourse.org/t/nginx-http-sock-bind-failed-after-reboot/29550/6 "2015-06-09T13:36:04Z")

</div>

I have the same issue. First, I thought that my own nginx has to be started a while after the docker containers (yet to confirm), but I found out then that the issue is related to the sock file.

From what I found on stackoverflow 😉 this issue seems to be a nginx one that is supposed to delete the sock file on termination. The discourse container has nginx v1.7.x Maybe an update to v1.8.x + changing the init.d script would do the trick?

Have a look:

> <https://unix.stackexchange.com/questions/164866/nginx-leaves-old-socket/198477#198477>

---

_[View the full topic](https://meta.discourse.org/t/nginx-http-sock-bind-failed-after-reboot/29550)._