[Development] Handling of Needs More Info issues in Jira

Alex Blasche alexander.blasche at qt.io
Mon Nov 23 09:07:43 CET 2020

As some of you might have noticed on Saturday night, I enabled automatic closure of "Need more Info" issues. Unfortunately I screwed the process up with a slightly wrong JQL querry when sedding the sets. This caused about 750 issues to be transitioned out of NMI. Essentially I failed to set the NMI flag.

If you are interested in the details of the triggering see:


Some of you recovered their tasks already and I didn't touch them again if you fixed the state up already. For everythin else I recovered the transitions this morning. Note that those recoveries were silent.

Note that this might trigger mass closure in 4 wks again and those are not silent (hopefully for the right reasons by then).


From: Development <development-bounces at qt-project.org> on behalf of Alex Blasche <alexander.blasche at qt.io>
Sent: Tuesday, 3 November 2020 10:23
To: development
Subject: Re: [Development] Handling of Needs More Info issues in Jira

Turns out I switched it on in some projects but not all. I simply forgot about it.

I will rectify this during the next days. Please note that this will trigger quite a few warnings about upcoming closure and it would give Jason a limited window to resolve. Therefore this has to be in stages. Likely I will start with 2018 issues.


From: Development <development-bounces at qt-project.org> on behalf of Jason McDonald <macadder1 at gmail.com>
Sent: Tuesday, 3 November 2020 04:36
To: development
Subject: Re: [Development] Handling of Needs More Info issues in Jira

On Tue, 3 Nov 2020 at 05:48, Oswald Buddenhagen <oswald.buddenhagen at gmx.de<mailto:oswald.buddenhagen at gmx.de>> wrote:
On Mon, Nov 02, 2020 at 08:01:13AM -0800, Thiago Macieira wrote:
>On Monday, 2 November 2020 05:24:45 PST Jason McDonald wrote:
>> There may be such a tool, but if there is it presumably isn't functioning
>> anymore (unless it's definition of "too long" is more than two years).
>I think it was supposed to be run manually every 6 months or so, closing
>entries that were more than 3 months without interaction (or so). The "run
>manually" might be the problem...
iirc, said tool was implemented as two (?) "cron jobs" running daily (?)
entirely inside jira. the timeouts for the heads-up and subsequent
closure were measured in weeks.

alex' announcement is here:

dunno what went wrong ...

Thanks Ossi.  That's useful information (and I like the other changes mentioned there too).  I will file a bug in the QTJIRA project asking for those cronjobs to be reinstated and a human to be notified if they break again.

Dr. Jason McDonald
(macadder on FreeNode)
Development mailing list
Development at qt-project.org

More information about the Development mailing list