pcp
[Top] [All Lists]

QA failures

To: pcp@xxxxxxxxxxx
Subject: QA failures
From: Martins Innus <minnus@xxxxxxxxxxx>
Date: Mon, 05 Jan 2015 10:56:57 -0500
Delivered-to: pcp@xxxxxxxxxxx
User-agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.10; rv:31.0) Gecko/20100101 Thunderbird/31.3.0
Hi,
I'm bringing up a bunch of VMs to do QA testing, and am seeing about 50 failures per machine. So far I've found 3 that are pretty common:

1. "bc" is required by a bunch of tests but the testsuite package doesn't depend on it. The minimal VM's I'm starting with don't have this installed by default. Is it reasonable to to make the testsuite packages depend on "bc"? If so , I can add it to the build system for the platforms I have. I have seen this on various Ubuntu and Centos VMs.


2. Some QA tests, for instance 276, try to run X based tests if the gui package is installed regardless of whether an Xserver is running. Should these not run? The relevant code seems to be in common.check:

**********
_check_display()
{
    # note: non-X systems (MacOSX, Windows) must pass here unchallenged
    which xdpyinfo >/dev/null 2>&1
    if test $? -eq 0
    then
        DISPLAY=$PCPQA_CLOSE_X_SERVER xdpyinfo >/dev/null 2>&1 || \
            _notrun "Failed sanity check on DISPLAY $PCPQA_CLOSE_X_SERVER"
        export DISPLAY=$PCPQA_CLOSE_X_SERVER
    fi
}
*************

In my case, I have the gui package installed, but no xdpyinfo. So should the testsuite package depend on xdpyinfo, or should the lack of xdpyinfo trigger a "notrun"? The comment seems to indicated that the lack of xdpyinfo can't make the test fail. This is on Ubuntu 14.04.



3. Again on Ubuntu 14.04.  Probably 20 tests fail with the following:

[4%] 039 - output mismatch (see 039.out.bad)
91a92,93
> __pmBind(/var/run/pcp/pmlogger.12496.socket): Permission denied
> __pmBind(/var/lib/pcp/.pcp/run/pmlogger.12496.socket): No such file or directory


QA running as the pcpqa user. .bad file attached. I have not had any luck in figuring this one out, since the actually QA test seems to run fine, but the parsing of the log file at the end contains this error.

Thanks

Martins

Attachment: 039.out.bad
Description: Text document

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