Page MenuHomePhabricator
Feed Advanced Search

Jul 5 2016

ofbeaton added a comment to T10745: Emails are sent out about restricted changes.

We have started to encounter this as we use phabricator in a devops environment with many teams from a large organisation.

Jul 5 2016, 3:09 AM · Restricted Project, Bug Report
epriestley added a comment to T10745: Emails are sent out about restricted changes.

There are several classes of transaction like this. One is that all actions are restricted:

Jul 5 2016, 1:44 AM · Restricted Project, Bug Report
Danny_B added a comment to T11271: Inconsistency in project membership displaying.

Will try to get more details for you from colleagues.

Jul 5 2016, 1:32 AM · Projects, People, Bug Report
epriestley added a comment to T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.

(And thanks for providing access to an environment where the issue reproduces!)

Jul 5 2016, 1:02 AM · Daemons, Bug Report
epriestley added a comment to T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.

@Mnkras, I think I've undone everything I did and terminated all sessions on the host. You can revoke my key at your convenience.

Jul 5 2016, 1:00 AM · Daemons, Bug Report
epriestley added a comment to T11271: Inconsistency in project membership displaying.

I can't reproduce this. Here's what I did:

Jul 5 2016, 12:59 AM · Projects, People, Bug Report
eadler added a comment to T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.

I was writing my reply while you were writing the above. Just figured I'd mention it.

Jul 5 2016, 12:42 AM · Daemons, Bug Report
epriestley added a comment to T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.

Yes, that's synchronous signal handling.

Jul 5 2016, 12:42 AM · Daemons, Bug Report
eadler added a comment to T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.

note that in the above we're explicitly handling signals. The signal_dispatch call must be placed inside of some activity loop.

Jul 5 2016, 12:41 AM · Daemons, Bug Report
eadler added a comment to T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.

function onsignal($signal) {

echo "Got SIGINT.\n";
  exit;

}

Jul 5 2016, 12:40 AM · Daemons, Bug Report
epriestley added a parent task for T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands: T9640: Make Phabricator compatible with PHP7.
Jul 5 2016, 12:39 AM · Daemons, Bug Report
epriestley renamed T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands from Daemons fail to stop and must be terminated to PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.
Jul 5 2016, 12:38 AM · Daemons, Bug Report
Danny_B added projects to T11271: Inconsistency in project membership displaying: People, Projects.

Just found another account disabled couple days ago which also says that doesn't belong to any projects, but some projects list him as a member. So the issue is in current version still present.

Jul 5 2016, 12:38 AM · Projects, People, Bug Report
epriestley added a comment to T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.

Yes, this reproduces on a local build of PHP 7.0.8. Here's what I did:

Jul 5 2016, 12:38 AM · Daemons, Bug Report
Danny_B created T11271: Inconsistency in project membership displaying.
Jul 5 2016, 12:36 AM · Projects, People, Bug Report
epriestley added a comment to T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.

I'm building PHP 7.0.8 locally. In the meantime, I found these possibly similar reports:

Jul 5 2016, 12:27 AM · Daemons, Bug Report
epriestley added a comment to T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.

Here's a smaller, self-contained test case:

Jul 5 2016, 12:15 AM · Daemons, Bug Report
epriestley added a comment to T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.

I produced this test script to try to see about version issues via 3v4l.org, but you can't write files there. Requiring C as part of a different file is important, so I don't see a way to use that.

Jul 5 2016, 12:06 AM · Daemons, Bug Report

Jul 4 2016

epriestley added a comment to T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.

I put the global signal handler in run(), unresponsive.

Jul 4 2016, 11:52 PM · Daemons, Bug Report
epriestley added a comment to T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.

I no-op'd the constructor and put the logic in run(), still unresponsive.

Jul 4 2016, 11:44 PM · Daemons, Bug Report
epriestley added a comment to T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.

I disabled the entire actual daemon behavior, still unresponsive.

Jul 4 2016, 11:40 PM · Daemons, Bug Report
epriestley added a comment to T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.

Removing the SIGINT handler entirely made it responsive.

Jul 4 2016, 11:34 PM · Daemons, Bug Report
epriestley added a comment to T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.

I changed the SIGINT handler to exit immediately, still unresponsive to SIGINT:

Jul 4 2016, 11:32 PM · Daemons, Bug Report
epriestley added a comment to T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.

This trivial script works correctly:

Jul 4 2016, 11:29 PM · Daemons, Bug Report
epriestley added a comment to T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.

The PhutilNiceDaemon daemon also does not respond to ^C.

Jul 4 2016, 11:24 PM · Daemons, Bug Report
epriestley added a comment to T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.

The debug daemon does not appear to respond to an external SIGTERM (via kill <pid> either).

Jul 4 2016, 11:21 PM · Daemons, Bug Report
epriestley added a comment to T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.

With --trace, the daemon isn't doing any work, and also isn't hung (just normal queries to look at the queue).

Jul 4 2016, 11:20 PM · Daemons, Bug Report
epriestley added a comment to T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.

I ran bin/phd stop and saw similar behavior:

Jul 4 2016, 11:17 PM · Daemons, Bug Report
epriestley added a comment to T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.

@Mnkras gave me access to the host.

Jul 4 2016, 11:03 PM · Daemons, Bug Report
epriestley added a comment to T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.

I can't reproduce that either:

Jul 4 2016, 10:49 PM · Daemons, Bug Report
Mnkras added a comment to T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.

Im really not sure, when I run restart this is what my daemon page looks like:

Jul 4 2016, 10:48 PM · Daemons, Bug Report
epriestley added a comment to T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.

I can't reproduce this. Here's what I did:

Jul 4 2016, 10:43 PM · Daemons, Bug Report
eadler moved T10225: arc patch --nobranch leaves submodules in the wrong state from Restricted Project Column to Restricted Project Column on the Restricted Project board.
Jul 4 2016, 9:19 PM · Haskell.org, Restricted Project, Arcanist, Bug Report
eadler moved T10682: Applications have no profile photos when they author transactions from Restricted Project Column to Restricted Project Column on the Restricted Project board.
Jul 4 2016, 9:19 PM · Restricted Project, Harbormaster, Bug Report
eadler moved T10109: Differental does not trigger Herald for mundane updates from Restricted Project Column to Restricted Project Column on the Restricted Project board.
Jul 4 2016, 9:16 PM · Differential, Restricted Project, Herald, Bug Report
eadler moved T9957: Arcanist cpplint install information is out of date from Restricted Project Column to Restricted Project Column on the Restricted Project board.
Jul 4 2016, 9:15 PM · Restricted Project, Lint, Bug Report
eadler moved T10105: Disabled account should disable all outbound mail from Restricted Project Column to Restricted Project Column on the Restricted Project board.
Jul 4 2016, 9:15 PM · llvm, Restricted Project, Mail, Bug Report
eadler moved T11085: Prevent users from generating an ambiguous commit message by editing revisions with the web UI from Restricted Project Column to Restricted Project Column on the Restricted Project board.
Jul 4 2016, 9:04 PM · Prioritized, Restricted Project, Arcanist, Bug Report
eadler moved T11154: Build results don't always get conferred to their buildable from Restricted Project Column to Restricted Project Column on the Restricted Project board.
Jul 4 2016, 9:04 PM · Restricted Project, Harbormaster, Bug Report
eadler moved T11145: Dialog submit buttons don't disable on click from Restricted Project Column to Restricted Project Column on the Restricted Project board.
Jul 4 2016, 9:02 PM · Restricted Project, PHUI, Bug Report
Mnkras updated the task description for T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.
Jul 4 2016, 8:33 PM · Daemons, Bug Report
Mnkras updated the task description for T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.
Jul 4 2016, 8:27 PM · Daemons, Bug Report
Mnkras created T11270: PHP 7 does not reasonably support asynchronous signal handling until `async_signals` lands.
Jul 4 2016, 8:27 PM · Daemons, Bug Report
rencris created T11269: Repository update command fails on empty Git repositories.
Jul 4 2016, 7:44 PM · Diffusion, Bug Report
epriestley added a comment to T11267: Commit graph broken for Mercurial repositories.

Thanks for the report! I think this should be fixed in HEAD of master. Let us know if that doesn't work or you run into anything else.

Jul 4 2016, 5:26 PM · Diffusion, Bug Report
epriestley closed T11267: Commit graph broken for Mercurial repositories as Resolved by committing Restricted Diffusion Commit.
Jul 4 2016, 5:24 PM · Diffusion, Bug Report
epriestley added a revision to T11267: Commit graph broken for Mercurial repositories: Restricted Differential Revision.
Jul 4 2016, 5:23 PM · Diffusion, Bug Report
epriestley claimed T11267: Commit graph broken for Mercurial repositories.
Jul 4 2016, 5:15 PM · Diffusion, Bug Report
lucaro created T11267: Commit graph broken for Mercurial repositories.
Jul 4 2016, 3:00 PM · Diffusion, Bug Report
Danny_B added a comment to T11263: DarkConsole "`" keypress does not fire immediately on some systems or keyboard layouts.

I found a workaround in the meantime: Press ` and then space (or any other key). The issue obviously is that ` is being sent like dead (combining) key on en-US keyboard layout.

Jul 4 2016, 9:57 AM · Bug Report (Needs Information)
epriestley closed T11266: Help text lost, when editing Diffusion Repository branches as Invalid.

https://secure.phabricator.com/book/phabricator/article/diffusion_managing/#branches

Jul 4 2016, 9:23 AM · Bug Report
aik099 created T11266: Help text lost, when editing Diffusion Repository branches.
Jul 4 2016, 7:32 AM · Bug Report
chad added a comment to T11263: DarkConsole "`" keypress does not fire immediately on some systems or keyboard layouts.

Firefox 44 / Ubuntu 15.04
Firefox 47.0.1 (64) Windows 8.1
Firefox 47.0.1 (64) Windows 10
Firefox 47.0.1 (64) macOS

Jul 4 2016, 1:53 AM · Bug Report (Needs Information)

Jul 3 2016

epriestley moved T10612: Editing a panel causes it to duplicate in dashboard from Backlog to v2 on the Dashboards board.
Jul 3 2016, 9:17 PM · Dashboards (v2), Bug Report
epriestley closed T11265: Milestone series navigation doesn't work as Invalid.
Jul 3 2016, 9:16 PM · Projects, Bug Report
Danny_B added a comment to T11265: Milestone series navigation doesn't work.

Ah, then it will be some of our local enhancements... Sorry for bothering then.

Jul 3 2016, 9:08 PM · Projects, Bug Report
epriestley added a comment to T11265: Milestone series navigation doesn't work.

Phabricator has no such sidebar item. Where are you seeing this?

Jul 3 2016, 8:54 PM · Projects, Bug Report
Danny_B created T11265: Milestone series navigation doesn't work.
Jul 3 2016, 8:32 PM · Projects, Bug Report
chad added a comment to T11263: DarkConsole "`" keypress does not fire immediately on some systems or keyboard layouts.

I can look in more detail when I'm back home. Kids have a picnic today.

Jul 3 2016, 5:07 PM · Bug Report (Needs Information)
Danny_B added a comment to T11263: DarkConsole "`" keypress does not fire immediately on some systems or keyboard layouts.

That's what I plan. Just tested it on WXPx86 and FF 47.0.1 32bit and it works there.

Jul 3 2016, 4:28 PM · Bug Report (Needs Information)
chad added a comment to T11263: DarkConsole "`" keypress does not fire immediately on some systems or keyboard layouts.

The next step on this report would be for you to try to reproduce on a different environment.

Jul 3 2016, 4:26 PM · Bug Report (Needs Information)
Danny_B added a comment to T11263: DarkConsole "`" keypress does not fire immediately on some systems or keyboard layouts.

And FF? And 32 or 64 bit?

Jul 3 2016, 4:21 PM · Bug Report (Needs Information)
chad added a comment to T11263: DarkConsole "`" keypress does not fire immediately on some systems or keyboard layouts.

Tested 8 and 10. I don't have Win7.

Jul 3 2016, 4:20 PM · Bug Report (Needs Information)
Danny_B added a comment to T11263: DarkConsole "`" keypress does not fire immediately on some systems or keyboard layouts.

@chad Which version of Win & FF you use?

Jul 3 2016, 4:11 PM · Bug Report (Needs Information)
chad added a comment to T11263: DarkConsole "`" keypress does not fire immediately on some systems or keyboard layouts.

I can't reproduce on Windows, Chrome, FF, IE with en layouts.

Jul 3 2016, 3:58 PM · Bug Report (Needs Information)
Danny_B added a comment to T11263: DarkConsole "`" keypress does not fire immediately on some systems or keyboard layouts.

Tested in both en and cs layouts.

Jul 3 2016, 2:33 PM · Bug Report (Needs Information)
epriestley added a comment to T11263: DarkConsole "`" keypress does not fire immediately on some systems or keyboard layouts.

/ was removed, see D15976 and the changelog:

Jul 3 2016, 2:15 PM · Bug Report (Needs Information)
Danny_B created T11263: DarkConsole "`" keypress does not fire immediately on some systems or keyboard layouts.
Jul 3 2016, 2:08 PM · Bug Report (Needs Information)
nickhutchinson added a comment to T11105: ExecFuture performs blocking reads on Windows, causing deadlocks with `arc lint`.

This is solved with D11688, but it needs to be turned on globally (at the moment it's a per call option, which was added so that internally our custom unit test runner wouldn't deadlock).

Jul 3 2016, 12:06 PM · Windows, Arcanist, Lint, Bug Report
epriestley merged T11262: On slow connections, it's possible to submit the Create Task form on workboards multiple times into T11145: Dialog submit buttons don't disable on click.
Jul 3 2016, 8:37 AM · Restricted Project, PHUI, Bug Report
epriestley merged task T11262: On slow connections, it's possible to submit the Create Task form on workboards multiple times into T11145: Dialog submit buttons don't disable on click.
Jul 3 2016, 8:37 AM · Maniphest, Bug Report
hach-que added projects to T11262: On slow connections, it's possible to submit the Create Task form on workboards multiple times: Bug Report, Maniphest.
Jul 3 2016, 5:20 AM · Maniphest, Bug Report

Jul 2 2016

epriestley removed a project from T11051: "Explain Why" isn't properly explaining why revisions were closed: Prioritized.

This could still be improved, but the immediate confusion we encountered appears fixed.

Jul 2 2016, 7:18 PM · Restricted Project, Differential, Bug Report
epriestley moved T11085: Prevent users from generating an ambiguous commit message by editing revisions with the web UI from Backlog to The Queue on the Prioritized board.
Jul 2 2016, 7:12 PM · Prioritized, Restricted Project, Arcanist, Bug Report
epriestley added a comment to T11085: Prevent users from generating an ambiguous commit message by editing revisions with the web UI.

We can fix the highly-tractable web UI part, at least.

Jul 2 2016, 7:12 PM · Prioritized, Restricted Project, Arcanist, Bug Report
epriestley renamed T11085: Prevent users from generating an ambiguous commit message by editing revisions with the web UI from arcanist incorrectly parses fields inside fields to Prevent users from generating an ambiguous commit message by editing revisions with the web UI.
Jul 2 2016, 7:12 PM · Prioritized, Restricted Project, Arcanist, Bug Report
featherless added a comment to T11259: UNRECOVERABLE FATAL ERROR after upgrading to 2016 Week 27.

Got it. Thank you! I just started the reindexing operations recommended in this release so I may leave things be 'til that finishes.

Jul 2 2016, 4:41 PM · Bug Report
epriestley added a comment to T11259: UNRECOVERABLE FATAL ERROR after upgrading to 2016 Week 27.

they won't not exist

Jul 2 2016, 4:41 PM · Bug Report
epriestley added a comment to T11259: UNRECOVERABLE FATAL ERROR after upgrading to 2016 Week 27.

"Ignored" in the sense of: no publishing, no herald rules, no audit triggers. Not, like, "they won't not exist".

Jul 2 2016, 4:41 PM · Bug Report
epriestley added a comment to T11259: UNRECOVERABLE FATAL ERROR after upgrading to 2016 Week 27.

Only that any legitimate commits pushed until things catch up will also be ignored. The repository will automatically leave the mode once everything finishes.

Jul 2 2016, 4:40 PM · Bug Report
featherless added a comment to T11259: UNRECOVERABLE FATAL ERROR after upgrading to 2016 Week 27.

Ok; will let it do its thing then.

Jul 2 2016, 4:39 PM · Bug Report
epriestley added a comment to T11259: UNRECOVERABLE FATAL ERROR after upgrading to 2016 Week 27.

There was a relatively recent change to make Phabricator discover commits which are only reachable from tags and refs. As it discovers these commits, it will treat them like new commits and publish feed stories about them. This is a one-time event, but a few installs have a lot of tag-reachable commits for whatever reason. You can find a bit more discussion in T11188, and details about the change in T9028 and T6878.

Jul 2 2016, 4:35 PM · Bug Report
featherless added a comment to T11259: UNRECOVERABLE FATAL ERROR after upgrading to 2016 Week 27.

Ah, I found the problem in our upgrade script. We run our upgrade in the following order:

Jul 2 2016, 4:30 PM · Bug Report
featherless added a comment to T11259: UNRECOVERABLE FATAL ERROR after upgrading to 2016 Week 27.

Thanks for quick response; hadn't seen this before so couldn't tell how serious the errors were.

Jul 2 2016, 4:29 PM · Bug Report
epriestley closed T11259: UNRECOVERABLE FATAL ERROR after upgrading to 2016 Week 27 as Resolved.

Oh, or Apache.

Jul 2 2016, 4:27 PM · Bug Report
featherless added a comment to T11259: UNRECOVERABLE FATAL ERROR after upgrading to 2016 Week 27.

Just rebooted apache once more and things are looking better.

Jul 2 2016, 4:26 PM · Bug Report
epriestley added a comment to T11259: UNRECOVERABLE FATAL ERROR after upgrading to 2016 Week 27.

Also likely fixed by restarting PHP-FPM, that method is defined here:

Jul 2 2016, 4:26 PM · Bug Report
epriestley added a comment to T11259: UNRECOVERABLE FATAL ERROR after upgrading to 2016 Week 27.

Did you restart everything after upgrading? See Restarting Phabricator.

Jul 2 2016, 4:25 PM · Bug Report
featherless added a comment to T11259: UNRECOVERABLE FATAL ERROR after upgrading to 2016 Week 27.

Getting the following when I visit differential/:

Jul 2 2016, 4:25 PM · Bug Report
featherless created T11259: UNRECOVERABLE FATAL ERROR after upgrading to 2016 Week 27.
Jul 2 2016, 4:23 PM · Bug Report
epriestley added a comment to T11257: HTML in Diffusion not escaped in certain circumstances.

For reference, @Danny_B filed a copy of this on HackerOne here so I could award a bounty for it:

Jul 2 2016, 2:50 PM · Security, Wikimedia, Diffusion, Bug Report
epriestley added a comment to T11257: HTML in Diffusion not escaped in certain circumstances.

Also worth noting is that Content Security Policy (T4340) would have prevented this. CSP is obviously worth implementing, but I believe the last attack it would have prevented was D4534 in January 2013, so it's hard to prioritize if it only defuses one attack every 3-4 years.

Jul 2 2016, 2:49 PM · Security, Wikimedia, Diffusion, Bug Report
Danny_B updated subscribers of T11257: HTML in Diffusion not escaped in certain circumstances.
Jul 2 2016, 1:25 PM · Security, Wikimedia, Diffusion, Bug Report
epriestley added a comment to T11257: HTML in Diffusion not escaped in certain circumstances.

The Phacility cluster has been patched.

Jul 2 2016, 1:12 PM · Security, Wikimedia, Diffusion, Bug Report
epriestley added a comment to T11257: HTML in Diffusion not escaped in certain circumstances.

In the future, please report security issues via HackerOne:

Jul 2 2016, 12:27 PM · Security, Wikimedia, Diffusion, Bug Report
epriestley claimed T11257: HTML in Diffusion not escaped in certain circumstances.
Jul 2 2016, 12:27 PM · Security, Wikimedia, Diffusion, Bug Report
epriestley closed T11257: HTML in Diffusion not escaped in certain circumstances as Resolved by committing Restricted Diffusion Commit.
Jul 2 2016, 12:18 PM · Security, Wikimedia, Diffusion, Bug Report
Danny_B created T11257: HTML in Diffusion not escaped in certain circumstances.
Jul 2 2016, 7:28 AM · Security, Wikimedia, Diffusion, Bug Report

Jul 1 2016

epriestley added a comment to T11237: Mobile action menus are "undefined".

Truly our most useful project.

Jul 1 2016, 10:22 PM · Chad's House, Mobile, Safari, Bug Report
chad added a project to T11237: Mobile action menus are "undefined": Chad's House.

Wait for it...

Jul 1 2016, 10:21 PM · Chad's House, Mobile, Safari, Bug Report