I don't believe this is related to E being set in variables_order.

Is it possible this is related to the E argument mentioned in T12071, but for arc clients?

Just spent some time tracking down what ended up being this issue. I wonder if adding a check in this block to query for the given revision would be sufficient enough to detect this case and return a more sensible error message. I can submit a diff for that if that'd be acceptable

Does datasource work? I've never tried it personally, but seems to be what you're asking for.
https://secure.phabricator.com/book/phabricator/article/custom_fields/

Is there any way to sort Differential diffs by how simple the change is? Is there a tag for small bug fixes so newbs can come in and see how a small problem was fixed?

Thanks!

There is a Badge Awarded tag for easy-ish contributions.

Unfortunately if it's a 5-10 minute bug fix, we'll just fix it.

Poking around the source code was fun. Please find time to put up minor issues like these for newbs. Thanks :)

Thanks, glad this all worked out.

It took me a little, but I found the source and fixed it. Sorry, this was probably a bit hairier than expected.

I don't know where the bug is, so I'm not sure how to guide you to fix it without fixing it myself.

Is it in PhabricatorFeedStory or PhabricatorFeedBuilder from the folder src/applications/feed. If I'm looking in the wrong place, could you please clarify what you meant as "level or two up".

UPDATE: This issue seems to have been addressed by upgrading to the latest codebase.

You can enforce the setting in /etc/security/limits.d/<user>.conf, example :

<user> hard nofile 8192
<user> soft nofile 8192

Replace <user> with the service username

Can you update your install and provide more detailed steps for us to follow? See Contributing Bug Reports for more information.

We don't accept bug reports from out of date installs, nor without steps the upstream can follow and reproduce.

Nothing set in stone, It just popped into my head as a possible solution

I'd probably use it in Dashboard creation as well for finding Panels.

I'd say it's a level or two up from the view controller, since setImageHref exists in PHUIFeedStory, but isn't being called.

Not really.

Generally makes sense to me.

@epriestley I think for Dashboards, and probably other applications, having some sort of compound typeahead/select would be helpful. I think maybe we can abstract this into using ApplicationSearch as the filter, and the tokenizer for the final results. Thoughts?

I just noticed this task as a beginner task. Shall I try to fix this?

I think T4316 is probably the best solution here, without knowing more about your user's workflow.

OK, I understand, I tried it several times... thanks for your time.

See Describing Root Problems if you need help defining the root problem. We don't take "feature does not exist" as a reason to build something.

The root problem is that I have created a custom field and I cannot modify it's value from the "Add Action" menu.

Well it's not a real problem if you see like that. It was only a suggestion. Of course I can live without that.

You have yet to provide a root problem

Sure I read them before posting anything!. Sorry I can't see what's wrong?

This doesn't explain the root problem we require for feature requests. Please see Contributing Feature Requests and Describing Root Problems. We don't expect people to need to make so many forms, we don't know why you're creating so many forms, we don't know what number that is on say a weekly basis, and so forth. We also have upcoming products like Nuance that might be better suited for your workflow. Please please please... only file feature requests stating the problem (and not the solution) you are having with Phabricator.

Sorry I'll try to describe the problem better.

Have you read the articles I linked?

Sorry I'll try to describe the problem better.

I don't think the upstream should provide this yet. You're already able to write your own custom field and since it's your code and your rules, will be 10x better for your users. Mostly, we need to abstract out other similar top level problems (input that drives an input?) and devise a single solution that may or may not perfectly fit your needs. It also might take years.

The information provided in this feature request doesn't provide the upstream with enough information to move forward. Please read Contributing Feature Requests and Describing Root Problems carefully and be sure follow all required steps before filing a task. We always require a root problem with every task.

The information provided in this feature request doesn't provide the upstream with enough information to move forward. Please read Contributing Feature Requests and Describing Root Problems carefully and be sure follow all required steps before filing a task. We always require a root problem with every task.

storage.local-disk.path is the location where persisted files are stored by the local-disk storage engine., not temporary files. See https://secure.phabricator.com/book/phabricator/article/configuring_file_storage/#engine-local-disk.

The chunks engine is not writable - this may be a problem. Moreover, both local-disk and amazon-s3 are not writable either. I have not set storage.local-disk.path - what should this typically be set to? Just some local path where phabricator will store temporary files?

Issue was: no more disk space

I know I can create a custom field with the select type, but the problem is that I have 10 different products (project tags) and each one has 100 milestones (2.0.1, 2.0.2, ...). You can imagine why I can not use select for this...

Custom fields already support select type, and also allows searching by it:

{
	"company.foundin": {
		"name": "Found In",
		"type": "select",
		"search": true,
		"options": ["version 1", "version 2"]
	}
}

search interface:

Thank for your answer!

Thank you for your quick reply. Indeed, this is tricky. Apparently, max-width has to be set on cells for overflow to work.

That said, if it's open, we do plan to address it eventually.

It sounds counter-intuitive, be the upstream prefers not to spend time on theoretical issues. Basically we have thousands of open tasks, and we complete them based on overall impact. Something hypothetically an issue isn't going to receive much attention.

What about triple backticks in code blocks? Code blocks can contain
anything, so I consider not being able to put three consecutive backticks
in them a bug.

Per T12218, the Phacility cluster is reaching a scale (~50 devices) where we need to improve tooling over "open a bunch of terminal windows". I want to look at implementing Phage for this use case and see how far I can get.

This summarizes the situation

What exactly is the user trying to do and how does it fail?

Can't the same tricks used to make a single web server handle multiple instances be used for the daemons?

I think that the problem is the content itself. This is evident by thumbnail fails to be generated when the second one is ok. On my OS/browser combo (Gentoo/firefox) first video loads very slowly while the second is perfectly ok. On Chrome first video refuses to load.
After downloading and testing locally, everything is fine. My guess is that chrome and firefox can guess proper values for decoders for "normal" video, while for "weird" video firefox tries it's best and chrome just gives up.

Yeah, same here. Both videos are of similar size (1.6MB vs 2.1MB) and both appear to legitimately be MP4 / H264 videos (neither is some kind of obviously spooky magic like an AVI named .mp4).

I can't play the top one in safari either

Here's what I see in Chrome at time of writing:

I was joking.

(Maybe the real user report was something only indirectly related, like putting the query on a Dashboard? That would make sense given that the name talks about dashboards...)