pcp
[Top] [All Lists]

Re: [pcp] QA status

To: Ken McDonell <kenj@xxxxxxxxxxxxxxxx>
Subject: Re: [pcp] QA status
From: Nathan Scott <nathans@xxxxxxxxxx>
Date: Thu, 29 Oct 2015 00:48:14 -0400 (EDT)
Cc: PCP <pcp@xxxxxxxxxxx>
Delivered-to: pcp@xxxxxxxxxxx
In-reply-to: <56318711.3050404@xxxxxxxxxxxxxxxx>
References: <56318711.3050404@xxxxxxxxxxxxxxxx>
Reply-to: Nathan Scott <nathans@xxxxxxxxxx>
Thread-index: W9EKBy0vEqJhzKwnh4etGnPmlR7YdQ==
Thread-topic: QA status

----- Original Message -----
> No whining this time ... looks pretty good.
> 
> [...]
> Summary: 21037 run, 38 failed (0.18%)
> 

Wow.  Someone buy that man a beer!  Those results are great, thanks Ken.

> ==== QA Failure (X) Map ====
> 
> Host       bl 00 01 02 04 05 08 14 18 19 21 22 27 28 29
> Test %fail                                              Test QA groups
>  892   10%     X  X                 X                    892 pmcd libpcp

Could you forward one or two of the .bad files for this one?  Its an IPv6
test, so maybe a configuration issue or situation we should be notrun-ing.

>  964   10%              X  X                       X     964 pmcd

Hmm, this one too - simple-ish test, so maybe a timing race lurking in
there still.

>  727    7%                    X  X                       727 avahi

This one's puzzling - its just like 840 and 950 (pmwebd advertising
vs pmcd/pmproxy) ... uses the same libpcp code ... should just pass;
so possibly a pmwebd buglet there, something about the way it calls
the service advertising routines.

> 666    7%  X                                   X        666 pmmgr slow
> 956    7%  X  X                                         956 pmcd

Actually, could you send sample .bad files for all 5 tests above?
(i.e. all the ones above the 3% fail mark in the farm).  Taa.

cheers.

--
Nathan

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