pcp
[Top] [All Lists]

Re: [pcp] Failing PMDA Perl example with disappearing instance domain va

To: Paul Smith <psmith@xxxxxxxxxx>
Subject: Re: [pcp] Failing PMDA Perl example with disappearing instance domain values
From: Nathan Scott <nathans@xxxxxxxxxx>
Date: Thu, 4 Dec 2014 03:51:14 -0500 (EST)
Cc: pcp@xxxxxxxxxxx
Delivered-to: pcp@xxxxxxxxxxx
In-reply-to: <30806F1D-98FB-4C8A-B626-0981F486F8FB@xxxxxxxxxx>
References: <30806F1D-98FB-4C8A-B626-0981F486F8FB@xxxxxxxxxx>
Reply-to: Nathan Scott <nathans@xxxxxxxxxx>
Thread-index: hxyFZQYzSmFiLwdRaj3GLUgDg4Omsg==
Thread-topic: Failing PMDA Perl example with disappearing instance domain values
Hi Paul,

----- Original Message -----
> Hi Nathan (& everyone), as discussed on IRC, here's the example PMDA we
> created copied, cleaned, scaled, and gutted from our activemq which shows
> what we're seeing the issue where the PMDA crashes when the refresh decides
> to inform PMCD that all of the instances in the indom have disappeared (I
> always think of the canonical case of instance domains of 'disks' where one
> has ejected all the disks leaving nothing, which should be ok.. as you say,
> not usual, but should work).

Thanks for the test case.  Yes, this is triggering an unhandled edge
condition in the C code that interfaces between libpcp_pmda and perl
scripts.  The fix is attached.  If it helps at all, its specific to
the use of array-based instances - switching to the hash-based model
will dodge this bullet.

I'll make a pcp/qa test case tomorrow from yours and commit the fix
for pcp-3.10.2 (which is scheduled for early January).

cheers.

--
Nathan

Attachment: perl.patch
Description: Text Data

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