pcp
[Top] [All Lists]

Re: [pcp] QA Failures: 660 and 752

To: Ken McDonell <kenj@xxxxxxxxxxxxxxxx>, Dave Brolley <brolley@xxxxxxxxxx>
Subject: Re: [pcp] QA Failures: 660 and 752
From: Nathan Scott <nathans@xxxxxxxxxx>
Date: Tue, 28 Oct 2014 17:30:43 -0400 (EDT)
Cc: pcp@xxxxxxxxxxx
Delivered-to: pcp@xxxxxxxxxxx
In-reply-to: <544FEFDC.9040005@xxxxxxxxxxxxxxxx>
References: <544FDE8E.6060403@xxxxxxxxxx> <544FEFDC.9040005@xxxxxxxxxxxxxxxx>
Reply-to: Nathan Scott <nathans@xxxxxxxxxx>
Thread-index: UKqpx5HCE7ZOaeCEiZfCdLmNs6oZsQ==
Thread-topic: QA Failures: 660 and 752

----- Original Message -----
> Dave,
> 
> I'm really encouraged to see others reporting QA issues, so thanks for that.

+1

> On 29/10/14 05:21, Dave Brolley wrote:
> > Looks like 660 needs to filter IPv4 and IPv6 addresses for local host
> > into something more generic.
> 
> Yep, but isn't the "PMWEBD error, code -22: Invalid argument" instead of
> html more significant?

Yes.  This test needs to be split into several tests now, as the presence
of jsdemos below $PCP_SHARE_DIR is not guaranteed (it's possible you have
an old copy of the demos there Dave? - there's a notrun guard at the head
of the test, which must be succeeding for you).  Either way, a bit of test
refactoring is needed here - I'll push some changes in soon.  I'm seeing
also some output dependence on microhttpd versioning on some hosts, so we
need a bit more filtering here too.

> > Looks like 752 failed because today is Tuesday  :-p
> 
> I've been wrestling with 752 for sometime and not got to the bottom of
> the "fails on Tuesday" issue ... leave this one with me, unless someone
> else has a bright idea and choose to jump in.

FWIW, I'm seeing this too - took a brief look yesterday but ran outta time
before diagnosing it.

cheers.

--
Nathan

<Prev in Thread] Current Thread [Next in Thread>