Some things that are maybe worthwhile (in addition to the blocked tasks):

See T5240.

If it were to be automagic, there should at least be a way to remove related tasks... For example, someone might say This is related to T123 and receive a reply of No, this is a separate issue. In this case, we should be able to remove T123 from the "automagic related tasks" field.

Would it be valuable to record the repositoryPHID for DivinerLiveSymbol / DivinerLiveAtom as well? Furthermore, are you able to explain the distinction between these two classes.

This is working as intended. You can add the address to your account in Settings → Email Addresses if you want it to work, per the instructions in the error:

Adding the PHID: I think this is quite reasonable, since the vast majority of books have a meaningful value for the field, but it should be nullable (i.e., you should be able to generate a book from a random collection of text files on your local disk if you want) and that it's likely not the best primary organization mechanism for finding/listing books.

I removed this on purpose when making the most recent round of modularization updates. My thinking was:

Overwhelmingly, the most common reason for this error is that you have the wrong remote URI or the wrong credentials. Have you double checked that those are correct?

It's expected that this will work, as long as you're using the same address as your email address in Phabricator and as a reply address.

T7604 removes the "Arcanist Project" field, but we could show the repository instead.

@hach-que, this is what you were asking about in IRC.

I've changed the test plan of D13066 now, but here is the original text that triggered the transaction being created:

The issue is that I always forget to learn how to use this functionality until I want to respond to an email and then its too late.

I have no idea why this works through the web UI...

It's intentional that these images aren't cacheable, but we should try to improve the browser behavior.

For a weekly event, I'd expect about 52. For a daily event, I'd expect about 365.

I don't want to build a hacky version of this blindly without having a future plan for solving the real package management problems we'll eventually encounter (off the top of my head: versioning, dependencies, trust, signing, channels/distribution, support for Phabricator applications and general-purpose libphutil libraries rather than just arc extensions).

We don't plan to support this.

Sorry, we do not support or accept feature requests for prototype applications like Harbormaster. See here for discussion:

Restart Apache (if you use Apache) or php-fpm (if you use nginx + php-fpm), and let me know if you still see the issue?

This happened again today (for my own future reference, see D20770 on our install).

See subtasks for details, but I've:

This is intentional in the general case because some config issues may be raised before we can load CSS, so they can't contain Remarkup.

This should have been resolved by modernizing ArcanistPylintLinter in D9109.

I set this up recently and can probably flesh out some documentation.

That sounds like a poor field name for users...

From T8343, the total corpus size warning didn't appear to fire properly.

In concrete terms, major items here that we know about specifically are:

we want to decide which emails just do not get sent, because a list will be cc'ed

ability to enforce selecting a mailing list to cc

@avivey, added your case from D13036 above.

I took a screencast watching htop over a ~1 minute period (filtered to show only PHP processes). It looks like /conpherence/panel/ and /notification/panel/ are quite CPU intensive.

If some commit, "A", is on some other commit, "B", which is on master, we consider that to be the "correct" form of "A": it's the one you tested, it's the one any local lint / unit ran against, and it's what we expect you to land.