pcp
[Top] [All Lists]

Re: [pcp] pmcd restart race condition

To: Ken McDonell <kenj@xxxxxxxxxxxxxxxx>
Subject: Re: [pcp] pmcd restart race condition
From: "Frank Ch. Eigler" <fche@xxxxxxxxxx>
Date: Fri, 2 Oct 2015 17:20:25 -0400
Cc: "'Martins Innus'" <minnus@xxxxxxxxxxx>, "'pcp developers'" <pcp@xxxxxxxxxxx>
Delivered-to: pcp@xxxxxxxxxxx
In-reply-to: <00bd01d0fd56$189950a0$49cbf1e0$@internode.on.net>
References: <560EB93A.6020606@xxxxxxxxxxx> <y0my4flump3.fsf@xxxxxxxx> <00bd01d0fd56$189950a0$49cbf1e0$@internode.on.net>
User-agent: Mutt/1.4.2.2i
Hi, Ken -


> [...]  Rather than a packaging post-op which has to be redone for
> each of the 3 or more packaging modes we support, I think it would
> be preferable to have pmda_setup _not_ run in the background [...]

That backgrounding was probably done to hide the latency of first-time
pmcd startup doing all the pmda Installs.


> [...]  we'll need a global marker in the filesystem (or possibly
> better, in the environment) to indicate that a pmda_setup is
> running.

Yes, as long as that is preserved throughout the process, sounds good.
... and then actually:

-           _pmda_setup &
+           PMDA_SETUP_RUNNING=yes _pmda_setup

might not even need to take away the "&", as the environment variable 
should preclude reentrancy to this part.

- FChE

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