From mboxrd@z Thu Jan 1 00:00:00 1970 From: Roger Pau Monne Subject: Re: [PATCH v8 00/15] execute hotplug scripts from libxl Date: Wed, 11 Jul 2012 10:08:05 +0100 Message-ID: <4FFD4275.1080403@citrix.com> References: <1341403176-50715-1-git-send-email-roger.pau@citrix.com> <1341772363.797.2.camel@hastur.hellion.org.uk> <4FFC1294.9040408@citrix.com> <1341935753.8586.48.camel@hastur.hellion.org.uk> <4FFC5339.4060504@citrix.com> <1341939462.8586.74.camel@hastur.hellion.org.uk> <4FFC6528.9000008@citrix.com> <1341942024.8586.97.camel@hastur.hellion.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1341942024.8586.97.camel@hastur.hellion.org.uk> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Ian Campbell Cc: "xen-devel@lists.xen.org" List-Id: xen-devel@lists.xenproject.org Ian Campbell wrote: > On Tue, 2012-07-10 at 13:23 -0400, Roger Pau Monne wrote: >>>> I'm working on making vfb/vkb async, once that is done I can take a look >>>> on this, >>> Cool. BTW is the other stuff ready? I guess it is waiting on IanJ to >>> review again? >> I've changed the series according to your comments, and I'm waiting for >> IanJ review. > > BTW, an interesting enhancement might be to read > the /local/domain/0/backend/vif/23/0/hotplug-error node and print it > out. Currently with the "hotplug: vif: fail if a duplicate vifname is > used" patch that I've just posted xl continues on despite the failure. > With this series this is improved to: > # xl cr /etc/xen//debian-x86_32p-1 > Parsing config from /etc/xen//debian-x86_32p-1 > libxl: warning: libxl_exec.c:118:libxl_report_child_exitstatus: /etc/xen/scripts/vif-bridge online [11939] exited with error status 1 > > but with xm I get: > # xm cr /etc//xen/debian-HVM-1 > Using config file "/etc//xen/debian-HVM-1". > Error: Device 0 (vif) could not be connected. Cannot rename interface vif20.0. An interface with name foo already exists. > > which is much friendlier. Ok, so on error I will read "hotplug-error" and add it to the error message. I think this is a Linux only thing, but it won't hurt to add it. > > Also it seems that despite xl reporting the error with you patches the > domain is still running, although without networking... If a hotplug > script fails can we abort and destroy the partially constructed domain > instead? Yes, I will change that and print and error and exit instead of just printing a warning. Thanks for the report.