Re: QA resource leaks

[Ken McDonell <kenj@xxxxxxxxxxxxxxxx>]

On 19/03/15 18:05, Nathan Scott wrote:
> ...Also, this test fails regularly, although
> intermittently, for me (and in the QA farm, from last releases' QA report)
> with ...
>
> $ diff /var/lib/pcp/testsuite/666.out*
> 13c13,18
> < less than three
> ---
> > count=5 archives created, expecting less than three
> > -rw-r--r--. 1 pcp pcp 22334 Mar 18 13:05 
> > /tmp/666-6686.dir/slick/archive-20150318.020547.meta
> > -rw-r--r--. 1 pcp pcp 22335 Mar 18 13:06 
> > /tmp/666-6686.dir/slick/archive-20150318.020611.meta
> > -rw-r--r--. 1 pcp pcp 22335 Mar 18 13:06 
> > /tmp/666-6686.dir/slick/archive-20150318.020637.meta
> > -rw-r--r--. 1 pcp pcp  7042 Mar 18 13:07 
> > /tmp/666-6686.dir/slick/archive-20150318.020702.meta
> > -rw-r--r--. 1 pcp pcp 22335 Mar 18 13:07 
> > /tmp/666-6686.dir/slick/archive-20150318.020720.meta
Don't think I've ever seen MORE archives created than expected ... for 
me it is passing on 92% of my QA hosts at the last round of runs ... 
which are quite a bit better than the previous QA report/landscape I posted.