From mboxrd@z Thu Jan 1 00:00:00 1970 X-Msuck: nntp://news.gmane.io/gmane.comp.sysutils.supervision.general/1331 Path: news.gmane.org!not-for-mail From: prj@po.cwru.edu (Paul Jarc) Newsgroups: gmane.comp.sysutils.supervision.general Subject: Re: graceful restart under runit Date: Thu, 16 Nov 2006 19:48:55 -0500 Organization: What did you have in mind? A short, blunt, human pyramid? Message-ID: References: <20061115114754.GA3759@fly.srk.fer.hr> <20061115160850.GA26987@home.power> <20061116152446.GA4721@fly.srk.fer.hr> <20061117001519.GA652@home.power> NNTP-Posting-Host: main.gmane.org Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Trace: sea.gmane.org 1163724541 3999 80.91.229.2 (17 Nov 2006 00:49:01 GMT) X-Complaints-To: usenet@sea.gmane.org NNTP-Posting-Date: Fri, 17 Nov 2006 00:49:01 +0000 (UTC) Cc: Dra?en Ka?ar Original-X-From: supervision-return-1567-gcsg-supervision=m.gmane.org@list.skarnet.org Fri Nov 17 01:48:59 2006 Return-path: Envelope-to: gcsg-supervision@gmane.org Original-Received: from antah.skarnet.org ([212.85.147.14]) by ciao.gmane.org with smtp (Exim 4.43) id 1Gkrud-0002pk-HP for gcsg-supervision@gmane.org; Fri, 17 Nov 2006 01:48:59 +0100 Original-Received: (qmail 23299 invoked by uid 76); 17 Nov 2006 00:49:20 -0000 Mailing-List: contact supervision-help@list.skarnet.org; run by ezmlm List-Post: List-Help: List-Unsubscribe: List-Subscribe: List-Archive: Original-Received: (qmail 23294 invoked from network); 17 Nov 2006 00:49:20 -0000 Original-To: supervision@list.skarnet.org In-Reply-To: <20061117001519.GA652@home.power> (Alex Efros's message of "Fri, 17 Nov 2006 02:15:19 +0200") Mail-Copies-To: nobody Mail-Followup-To: supervision@list.skarnet.org, Dra?en Ka?ar Original-Lines: 15 User-Agent: Gnus/5.110003 (No Gnus v0.3) Emacs/21.4 (gnu/linux) Xref: news.gmane.org gmane.comp.sysutils.supervision.general:1331 Archived-At: Alex Efros wrote: > Another option - you can ask runsv to 'x' (Exit) instead of 't' (Term). > In this case runsv will send SIGTERM to your process, which can process it > by just closing listening socket, waiting until existing connection finish > and then exit. > After few (up to 5) seconds runsv will be started again by runsvdir, and > so start second process of that server (which will open listening socket > again). This seems worse than t. In either case, new connections are refused while the old process cleans up its current connections, but with x, new connections are also refused for up to 5 seconds more. paul