Site down after upgrade

Hi Guys,

during trying to upgrade to recent version my site went completely down.

We can see the following error messages in browser:

Corrupted Content Error

The site at has experienced a network protocol violation that cannot be repaired.

The page you are trying to view cannot be shown because an error in the data transmission was detected.

** Please contact the web site owners to inform them of this problem.**

Please can you help.


Can you give us more information?

  • How was the instance built?
  • How long since the last upgrade
  • Any error messages?

I can see you’re using CloudFlare, do you have any rules in place? Is Railgun or Minification enabled?


Adding to this, how were you upgrading? GUI? Command line?

Details on these sorts of reports allow anyone trying to assist to be much faster at sussing out issues

1 Like

Hi Stephen,

Thank you for your quick response.

  • The instance was built with digitalocean droplet since more than a year and has always been working nicely but this time during the upgrade, i dont know what went wrong.

  • Since the last upgrade i think about 2 months.

  • No error message can be seen. But when we try at explorer, we see this message: has experienced a network protocol violation that cannot be repaired.


Hi Jeff,

I was trying to do this through Commandline by following the same instructions I used to do since last one year. But this time, I really dont know what has went wrong.

We receive this error in browser now: has experienced a network protocol violation that cannot be repaired.


Is it a one-click install, or did you build it following the official install guide?

If it’s the latter is this Ubuntu 16.04 LTS? Which version of Docker is on the machine?

Do you still have the command line open? It’s very likely there’s an error message in there which will help immensely.

Yes Stephen, it was a one click install. However the install was done a year ago.This was the attempt to upgrade to the latest version that had critical updates. Unfortunately the commandline is not open. and I am unable to connect to putty as well since its shows Network error - connection timed out.

If you can’t connect to SSH it’s not a Discourse error, you need to log into DigitalOcean and restart the VPS. Discourse is an application which runs inside Docker, isolated from the underlying operating system. It can’t block something as fundamental as SSH.

Are you using a strong password for your SSH account?


ok. Sure. I can give it a go now. and will let you know shortly. Cheers


Hi Steph, I just restarted the droplet but it is still same I am afraid. I am using just a normal password for SSH account.

1 Like

Open the console for your droplet on - do you see any messages there?

Yes there are some messages for the NYC DC connecity issues and my droplet is in NYC3. Could that be a reason. But my site is down since last 8 hours now. Cheers

Look at the blue message at the top. Your server is in NYC3 and they have connectivity issues there.

During this time, some customers may experience connection issues with corresponding Droplets.

Until the network problems are resolved it’s hard to say if there is a Discourse problem at all.

For now you can click on the blue name of your server then click ‘access’ and ‘launch console’ it will at least let you see if the sever has restarted correctly.


Yes, i logged in as you suggested into console and can see the attached view. Please can you also take a look.

Ok, and if you do:

docker ps

what do you see?

If Discourse is up and running it will look something like:

CONTAINER ID        IMAGE                 COMMAND             CREATED             STATUS              PORTS                                      NAMES
358eb862c311        local_discourse/app   "/sbin/boot"        3 days ago          Up 3 days >80/tcp,>443/tcp   app
1 Like

I tried this. Please see attached.

So yes, Discourse looks to be fine, the problem is that the NYC3 zone of DigitalOcean has network issues.

When that issue has been resolved I’m pretty sure your community will reappear. If it doesn’t please update the thread and we can take a deeper look.


Great. Thank you very much Stephen. Your help is really appreciated at the time of need when I was bit stressed. I will keep an eye on network issue in NYC3 and will let you know soon i get an updated. :slight_smile:

Thanks again. Cheers :slight_smile:


Up 15 minutes

your server might have restarted on you here too.

Check it with uptime if you’re curious

Once all the outages here stabilize you should be free to upgrade as usual

1 Like

That ones on me, I asked him to reboot it from the web interface when he couldn’t SSH in!