pcp
[Top] [All Lists]

Re: [pcp] rpm pmda

To: Nathan Scott <nathans@xxxxxxxxxx>
Subject: Re: [pcp] rpm pmda
From: Stan Cox <scox@xxxxxxxxxx>
Date: Wed, 27 Nov 2013 15:31:22 -0500
Cc: PCP <pcp@xxxxxxxxxxx>
Delivered-to: pcp@xxxxxxxxxxx
In-reply-to: <2052845457.6426836.1382325018005.JavaMail.root@xxxxxxxxxx>
References: <52608FC9.2050103@xxxxxxxxxx> <2052845457.6426836.1382325018005.JavaMail.root@xxxxxxxxxx>
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.1.0
I just wanted to give a heads up for this reboot, which caches the rpm metrics in advance as opposed to version one's accessing them just in time. This is an effort to improve the fetch performance. The rpm metrics are cached into the data structure 'cache'. A hash table is built; the key is the rpm name and the value is the pointer to the cache entry. The hash table, via hsearch_r, is used to find the cache entry for an rpm. The fetch callback accesses the cache and returns the proper info. If the rpm database changes then it is reloaded. -Currently inotify checks the entire dir to make testing easier
-The hashing sometimes fails; I am diagnosing that problem
-Threading needs to be revisited since the reboot
-A subset of available rpm metrics is being cached

committed to pcpfans scox/dev branch

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