pcp
[Top] [All Lists]

Re: [pcp] Recent QA failures

To: Ken McDonell <kenj@xxxxxxxxxxxxxxxx>, 'pcp developers' <pcp@xxxxxxxxxxx>
Subject: Re: [pcp] Recent QA failures
From: Martins Innus <minnus@xxxxxxxxxxx>
Date: Mon, 27 Apr 2015 13:10:57 -0400
Delivered-to: pcp@xxxxxxxxxxx
In-reply-to: <02d701d08064$fe50d6a0$faf283e0$@internode.on.net>
References: <026e01d07fa2$874e9cd0$95ebd670$@internode.on.net> <553D50C3.8090703@xxxxxxxxxxx> <02d701d08064$fe50d6a0$faf283e0$@internode.on.net>
User-agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.10; rv:31.0) Gecko/20100101 Thunderbird/31.5.0
Ken,

On 4/26/15 5:07 PM, Ken McDonell wrote:
-----Original Message-----
From: Martins Innus [mailto:minnus@xxxxxxxxxxx]
Sent: Monday, 27 April 2015 6:56 AM
...
982 and 800 are likely mine from trying to fix previous false positives.
Can you
send me an OS where those are failing?
G'day Martins.

The failures so far have only been on bozo-vm (PCP 3.10.5) x86_64  Debian
7.6 which is running on a VirtualBox VM  on a Windows 7 host (my laptop).


OK, I am not able to reproduce on a similar system: Deb 7.6 x86_64 on a VirtualBox VM on a Mac host. Could I ask you to try a couple things?

1. For 800 and 982 increase the wait/pmsleep, maybe double the values. If that doesn't work, increase them significantly, maybe 20 seconds.

2. Send a directory listing of some /proc/<pid>

3. Try :

sudo pmstore  hotproc.control.config "uid==0"
<wait ~20 seconds>
pminfo -f  hotproc.control.config
pminfo -f  hotproc.nprocs
pminfo -f  hotproc.id.uid

And send me the output?

Thanks

Martins

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