Forum upgrade crashing at Upgrade `docker_manager`

[modius (Geoff Bowers)]

My forum is crashing at this point in a web based upgrade; sends CPU to 100% and I have to reboot the server to recover; standard vanilla 2GB Droplet at DigitalOcean.

In Upgrade docker_manager

***  Bundling assets. This might take a while *** 
$ bundle exec rake assets:precompile
Purging temp files

Are there any manual steps to force an upgrade from the command line that might fare better?

Note, I realise it says it may take a while but its only a small forum and it never seems to recover.

[brunoedigital (Bruno)]

Mine seems to be stuck as well. But on a different part. Last one that shows up is:

'367 /var/www/discourse/vendor/bundle/ruby/2.3.0/gems/tzinfo-1.2.2/lib/tzinfo/country_index_definition.rb

Memory usage is high right now. Because of upgrade I believe.

Let’s see if it goes through or not.

If it doesn’t, what are the recommended steps? Reset upgrade and then manually rebuild via terminal?

What are the recommended commands to do that?

EDIT: in the end I had to reset it and then it went through.

BTW, is there a “normal” memory usage?

Mine seems pretty high. Only 84 free.

[Steven]

I think the best way is to reset upgrade then rebuild from the server.

cd /var/discourse
git pull
./launcher rebuild app

But before that, you may want to reboot the droplet manually using this command : shutdown -r now

The rebuild will upgrade everything and in 10-15 minutes, it will be over

[codinghorror (Jeff Atwood)]

Make absolutely sure you have 2GB of swap space set up.

[modius (Geoff Bowers)]

@codinghorror

Make absolutely sure you have 2GB of swap space set up.

Just thought I’d update on my fix. Turns out you were right to check if there was a swap file in place – there was not.

Create a swapfile for your Linux server

For servers with <= 2GB of RAM, running ./discourse-setup will prompt for and automatically create a 2GB swapfile. Most cloud virtual machine providers do not set up swapfiles as part of their server provisioning. In particular, upgrading Discourse produces a lot of memory pressure. With a swap file, rather than randomly terminating processes with an out of memory error, things will slow down instead. Having a swap file is a cheap insurance policy that protects you from many other lo…

We use to have a swap file but…

This is how i got to this point, and it might be useful for others:

• was upgrading discourse when postgres upgrade failed
• rather than troubleshoot directly, thought we would simply spin up a new droplet, fresh install and restore from back-up
• used the DigitalOcean instant install for Discourse (this was my MISTAKE)

Looks like the standard Digital Ocean jumpstart thing for Discourse does not set up a swapfile. Probably should have just run through a standard installation from a vanilla Ubuntu install.