When I'm trying to wh -f foo -s bar deploy, the packaged zip is 10mb, but I'm getting a 413 error despite the fact that I've configured my nginx reverse-proxy to allow client body size of several hundred megabytes.

http://forums.webhook.com/t/can-not-invite-collaborators/206/2

Looks like you can't invite collaborators in a self-host setup. Returns a permissions error.

Hi there,

I have a webhook server up and running, but it doesn't create any buckets when I run a deploy for a site.
Do you know what can go wrong? I double checked the site verifications already.

From the forums

http://forums.webhook.com/t/some-issues-as-a-self-hoster/307

The upload works perfectly, but when the image markdown tag gets added to the text, the url is wrong. Let's say my website is www.example.com, instead of getting http://www.example.com/webhook-uploads/xxxxxxxx.jpg I get this wrong url : http://www,1example,1com.webhook.org/webhook-uploads/xxxxxxxx.jpg.

I'm getting the error:

========================================================
# The site themetest3,1sonnetmedia,1org/ is not owned by you
========================================================
# You must be an owner on a site to delete it.

I'm an owner in Firebase, so I'm guessing there's an issue with the rules.

In the commandDelegator, a comment states that memcached is needed to prevent jobs from getting queued twice. Similarly, the jobQueue speaks of "locking".

It's not entirely clear to me, however, what exactly these locks are meant to prevent - as I understand it, the child_added event from Firebase only fires once for any given job (ie. no duplicates can occur there). Similarly, presumably beanstalkd will only hand out a job once, so no locking should be necessary there either.

Could you clarify on the purpose of memcached here, and whether it still plays a role in the current version of the codebase? I've been looking at the code for a while now, and I can't find any explanations or hints anywhere of the concrete issues that this is meant to prevent.

Logged in as owner I noticed that when trying to add settings to a self-hosted project that nothing happens, no error or anything, but the settings don't hold. I tested this against hosted sites where I can indeed add settings. I tried importing settings and got a permissions error (so there I got an error).

Once I saw that it was a permissions error, I tried rolling back to the old Firebase rules (commit 361008e) and sure enough I'm able to change settings. Looking at the diff, I couldn't tell exactly what needs to change though.

So I'm getting errors when I try to change an user from any group to an owner.

[Warning] FIREBASE WARNING: set at /management/sites/www,example,1org/owners/budparr@sonnetmedia,1net failed: permission_denied  (vendor.min.js, line 21)

I'm not getting the error in my hosted sites, just self-hosted, yet I'd think that a) the source would be a) in the security rules, and b) the same as hosted.

Any thoughts where I should be looking?

To delete a website, the wh cli publishes data to the management/commands/delete node of the Firebase. Looking through this repo, I can't find where this command gets handled. Is that something that can be pushed to the repository?

When sending an invite e-mail on a self-hosted website, the subject is :

[Webhook] You've been invited to edit www,1example,1com

The best solution would be the use the Site Name value entered in the site's Global Settings, but then we would need to validate that the user has entered a value before sending an invite email.

Upon copy/pasting the code from https://github.com/webhook/webhook-server-open/blob/master/security.json into Firebase, I get a bad string error on line 67.

Hi guys - I wrote a post on the forums about build times for larger sites and that got me thinking about use-cases. I'd love to see some of the editorially driven sites I work on capable of using Webhook, but I think even a long-running blog would start to be a drag once it got to a few thousand pages.

I may be naïve, but it seems like Grunt Newer, which says it runs "Grunt tasks with only those source files modified since the last successful run" would be a good solution.

Would love to hear what you think, particularly if this sort of thing isn't even a concern, or on the roadmap but in the far future.