Project

Profile

Help

Issue #1674 ยป Issue #8995 - 2016-02-16T15_13_18Z.eml

rbarlow, 02/16/2016 04:13 PM

 
Date: Tue, 16 Feb 2016 16:13:18 +0100
From: rbarlow@redhat.com
To: dropbox+pulp+c71e+pulp@plan.io
Message-ID: <56c33c8ebdc4c_9573fda404dba20571d9@gemini.mail>
in-reply-to: redmine.journal-8970.20160215184919.6b171fd8e3d23ff6@plan.io
Subject: Re: [Pulp - Issue #1674] Tasks in canceled state can not be purged
Mime-Version: 1.0
Content-Type: multipart/mixed;
boundary=4tCRGCAxpeGahGVC5p2lfgsQGDEAe6fUf;
charset=UTF-8
Content-Transfer-Encoding: 7bit
X-Scanned-By: MIMEDefang 2.68 on 10.5.11.27
X-HE-Spam-Level: ------
X-HE-Spam-Score: -6.0
X-HE-Spam-Report: Content analysis details: (-6.0 points) pts rule name
description ---- ----------------------
-------------------------------------------------- -5.0 RCVD_IN_DNSWL_HI RBL:
Sender listed at http://www.dnswl.org/, high trust [209.132.183.28 listed in
list.dnswl.org] -1.0 ENCRYPTED_MESSAGE Message is encrypted, not likely to be
spam
X-HE-SPF: PASSED


--4tCRGCAxpeGahGVC5p2lfgsQGDEAe6fUf
Content-Type: text/plain;
charset=UTF-8
Content-Transfer-Encoding: quoted-printable

Pulp wrote:
> it's still not clear why purging the TaskStatus records could have a
> negative impact, even with today's code and possible task states. Once
> the cancel is handled by the webserver, it sends a celery signal to all=

> workers which are stored in a registry of cancelled tasks internally to=

> Celery. In other words once celery is told to cancel a task, removing
> the TaskStatus record should have no effect.

Hello Brian!

I think the only problem lies with tasks that don't immediately stop
when getting canceled (like a sync/publish). Those may continue writing
data to the TaskStatus. That is probably not a big deal, but something
to consider. Maybe it's fine for them to just write to nothing and error
(or maybe they would fail silently depending on how they form their
update query).

-- =

Randy Barlow
irc: bowlofeggs

--4tCRGCAxpeGahGVC5p2lfgsQGDEAe6fUf--
    (1-1/1)