All sites broken after 2.1.1 upgrade - MySQL errors

Support
1

I finally just got everything working great with a custom test build of 2.1, then today auto-upgraded to 2.1.1 and now my entire local dev environment is unusable.

I’m able to start Local, and turn on a site. But when I try to load the site, the connection is refused. I think this is the relevant stuff from my Local log:

Nov 2, 2017, 4:56 PM GMT-7 - warn: [main/waitForMySQL] MySQL stderr: amysqladmin: connect to server at 'localhost' failed
error: 'Can't connect to local MySQL server through socket '/var/run/mysqld/mysqld.sock' (111)'
Check that mysqld is running and that the socket: '/var/run/mysqld/mysqld.sock' exists!
Nov 2, 2017, 4:56 PM GMT-7 - info: [main/docker-network-manager] Connected a06b25cb03793717da95bca718d1877b6580e6ebce90dd1067914f1654df1241.
Nov 2, 2017, 4:56 PM GMT-7 - info: [main/waitForMySQL] Querying MySQL
Nov 2, 2017, 4:56 PM GMT-7 - warn: [main/waitForFile] /conf/php/7.1.4/etc/php-fpm.conf does not exist
Nov 2, 2017, 4:56 PM GMT-7 - warn: [main/index] Unhandled Rejection at: Promise Promise { <rejected> '' } reason:

I get this error for every site, some running mysql 5.5 and some running 5.6.

2

Okay, I’ve an update to this:

All the above happened this afternoon on my MacBook Pro.

Tonight, I got the current release working fine on my Mac Pro. Before I upgraded Local, I made backups of every site (in case everything goes sideways and I have to nuke and start over). Then, instead of using the automatic updater, I just downloaded Local 2.1.1 directly and replaced the .app file in Applications.

I may just go ahead and nuke and rebuild the sites on the laptop anyway, but this has thrown my schedule off quite a lot!

3

@chadbush,

Edit: I just remembered you were on a test build of FDVv2. That’s probably why the issues popped up. Try going to Settings » Advanced and toggling Faster Docker Volumes.

Previous reply:

Can you try going to Help » Restart Local’s Machine?

Then, restart the sites accordingly.

4

Thanks Clay!

Toggling Faster Docker Volumes got things running again, but now Local is nuking all my databases. This error occurs:

Nov 7, 2017, 10:12 AM GMT-7 - error: [main/stopSite] mydumper stderr rm: cannot remove '/app/sql': Directory not empty Nov 7, 2017, 10:12 AM GMT-7 - error: [main/stopSite] mydumper stdout

5

Hi Chad,

Whenever a site is stopped it uses mydumper to dump the database into the /app/sql folder. That folder is only used as a backup.

That error is probably happening due to an NFS (FDVv1) issue. When you were on FDVv2, did that error not show?

6

Hi Clay,

I fully toggled Faster Docker Volumes on and off a couple of times, and restored the nuked databases from the previously mentioned sites by concatenating the files in the sql folder from my latest backups and importing them via Sequel Pro. Now the problems seem to have resolved themselves. Thank you!