pcp
[Top] [All Lists]

Re: [pcp] URGENT potentially serious regression in 3.7.0

To: pcp@xxxxxxxxxxx
Subject: Re: [pcp] URGENT potentially serious regression in 3.7.0
From: Ken McDonell <kenj@xxxxxxxxxxxxxxxx>
Date: Wed, 03 Apr 2013 09:00:22 +1100
Delivered-to: pcp@xxxxxxxxxxx
In-reply-to: <515B2372.90503@xxxxxxxxxx>
References: <513B99E4.7030007@xxxxxxxxxxxxxxxx> <513D9A93.6080806@xxxxxxxxxxxxxxxx> <515A7236.5080901@xxxxxxxxxxxxxxxx> <515B2372.90503@xxxxxxxxxx>
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130308 Thunderbird/17.0.4
G'day Dave.

On 03/04/13 05:29, Dave Brolley wrote:
...
My apologies for the turdlet(tm) and the subsequent fallout.

No problem. I (and I suspect others) had noticed this code (bonus points for the TODO annotation which we've found most useful across the life of PCP) but I had not appreciated the subtle implications before digging into the qa/169 failures.

Fortunately the result was benign and probably not visible to anyone outside QA land.

... There is
another potentially unnecessary reverse lookup in __pmGetAddrInfo (both
implementations call __pmGetNameInfo) that I'm looking into as well.

I'm not sure we can purge all the reverse DNS lookups, but the one that had been in the pmcd trace diagnostics code from the epoch was definitely lead in the saddle bags.

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