Date: Tue, 11 Feb 2014 11:43:29 +0100 From: Marko Zec <zec@fer.hr> To: Vijay Singh <vijju.singh@gmail.com> Cc: FreeBSD Net <net@freebsd.org> Subject: Re: netstat for vnets Message-ID: <20140211114329.43bda628@x23> In-Reply-To: <C8467D00-8CF0-4737-B261-DCB4C3A87146@gmail.com> References: <C8467D00-8CF0-4737-B261-DCB4C3A87146@gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 10 Feb 2014 21:24:52 -0800 Vijay Singh <vijju.singh@gmail.com> wrote: > How does "jexec <jid> netstat -an" get its data from the kernel? I > know that netstat uses kvm, but I'm not sure how it works with > vnet-jails. Any pointers would be much appreciated. I think these days libkvm first tries to find "native" symbols via kldsym(), and if it fails, attempts to prepend a "vnet_entry_" prefix in front of the search key, which then get resolved to the proper addres in _kvm_vnet_validaddr() in lib/libkvm/kvm_vnet.c. In the early days of VIMAGE it was the kernel who was responsible for resolving the symbol address in the proper vnet, honestly I do not recall when the magic moved to the userland and why... Btw. Vijay - did the patch I posted here: http://lists.freebsd.org/pipermail/freebsd-net/2014-February/037769.html resolve the panics you complained about? Cheers, Marko
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20140211114329.43bda628>