Jump to content
thirty bees forum
  • 0

Question

Posted

Hey hey guys.

First up, thank you for putting thirty bees together. It's great to have the option to move from 1.6 and not have the wall ahead that seems to be 1.7.

Just two things I want to raise with you with regard to your migration module from 1.6. The first is that I can't use directory as a chanel. The second is that in modules/psonesixmigrator/classes/Tools.php on line or around 2234, you have a GuzzleHttp Client instance that you instantiate with a hard timeout value of 60 seconds.

This means that if the migration module doesn't download the file in less than that time, the request times out, and the module says that it downloaded the core files, but the checksums don't match, which they don't as we end up with an incomplete download.

I bring this to your attention, because I am currently migrating a version of a live store to thirty bees on top of a docker instance from a location where the connectivity is not as good as say on a hosted server.

I can't use the directory option, i.e. download the files and point to them, and the download times out over my connection. I've got around this by changing the value of that timeout in the file, but if I hadn't done that, I would be kind of stuck at the upgrade point.

Once again, thank you for this initiative, hope this finds you well.

7 answers to this question

Recommended Posts

  • 0
Posted

Hey hey, thanks for putting in that fix.

Just wondering if it wouldn't be better to throw an exception on a timeout which the migrator shows as an error, so that someone can see that that's the cause of the failure, or setting that timeout to never timeout so it's never an issue.

It was more just a case of having to dig down to find out why it wasn't downloading that had me confused.

I thought initially it was a php/nginx config issue, and it took a while before I figured out that was where things were going wrong. Thanks for getting back to me so quickly!

Hope you're having a good day:)

  • 0
Posted

IIRC the download isn't even tried if the requested file already exists. One solution to the problem discussed here is to give the user instructions on how to download and where to place it manually as part of the timeout message. This way he can use his preferred download manager or whatever means to get the task solved.

  • 0
Posted

@mdekker yes, it does remove the file and clean out the directory every time you start the migration process. Then checksums the new downloaded file and the add-on, and if that passes, then extracts and proceeds :)

  • 0
Posted

@mdekker Thanks for putting in that quick fix, as you said it addresses the problem quickly and should avoid it happening for most people. Sure you guys have a ton of more pressing things to get out the door, thank you again, and thank you guys for being so responsive.

  • 0
Posted

@mdekker can't be human ;) Or never sleeps!

Imagine if he's on this full time with the help of patreon! https://www.patreon.com/thirtybees

  • 0
Posted

@mdekker

IIRC the download isn’t even tried if the requested file already exists. Have you tried it? I did not succeed :(

Actually I'm not sure. It's well possible I did a change locally to make this possible. As the file is checksummed anyways, there's not much reason to delete it, just to download it again. Downloading it after a checksum mismatch is entirely sufficient.

As I'm currently working on this module anyways, I'll put such a change into my TODO list.

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...