Path: csiph.com!news.redatomik.org!news.neodome.net!news.mixmin.net!news.unit0.net!newsfeed.CARNet.hr!news.spin.it!bofh.it!news.nic.it!robomod From: Reco Newsgroups: linux.debian.user Subject: Re: Still unable to restart networking on Debian 9 text mode only Date: Fri, 23 Nov 2018 13:50:01 +0100 Message-ID: References: X-Mailbox-Line: From debian-user-request@lists.debian.org Fri Nov 23 12:46:03 2018 Old-Return-Path: X-Amavis-Spam-Status: No, score=-6.1 tagged_above=-10000 required=5.3 tests=[BAYES_00=-2, DIGITS_LETTERS=1, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, FOURLA=0.1, LDO_WHITELIST=-5] autolearn=ham autolearn_force=no X-Policyd-Weight: using cached result; rate: -6.1 Dkim-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=enotuniq.net; s=dkim; h=Message-ID:User-Agent:In-Reply-To: Content-Transfer-Encoding:Content-Type:MIME-Version:References:Subject:To: From:Date:Sender:Cc:Content-ID:Content-Description; bh=e+sk4cwR9dF2VZUBs/A1XuJOdKOPAChcMciLqDzhsyU=; b=b7LHvIgqFZ6T9XJTsiPcthifv7 zIWt5dpSrE6j3czLazTqVtfGEm09QVz/l5q7zp5SLcq5HrZncAx3ahFo+B0tWrvDz8ESY8qANGF3P RfOT7qaXYws+sTjBYGtc6bCdz4o7Aqv144aN9enKI7S149fI13nxNHf0z7gE/L5GLpZ7QfWuExvB2 nCsNfn7VnNItvaYI5TEcxmHvurowa8/wo+7+qbeM7UBezmjut5h+R4nETO6ZNBFGE+HGvu7AE75Xm o/a+FSy0VwZER7UoAUvrQcqsHRdIZIIa463fMu7X0lBhlgNwIiZsrLHhMA8CPmtorJWflpJpxTZqX Qw3MBkkQ==; MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: 8bit User-Agent: NeoMutt/20170113 (1.7.2) X-Mailing-List: archive/latest/742679 List-ID: List-URL: List-Archive: https://lists.debian.org/msgid-search/E1gQAqF-0007D5-Hu@enotuniq.net Approved: robomod@news.nic.it Lines: 87 Organization: linux.* mail to news gateway Sender: robomod@news.nic.it X-Original-Date: Fri, 23 Nov 2018 15:45:51 +0300 X-Original-Message-ID: X-Original-References: Xref: csiph.com linux.debian.user:202715 Hi. On Fri, Nov 23, 2018 at 10:31:28AM -0200, Luciano Andress Martini wrote: > Good morning, > > IMPORTANT: Network Manager was not being installed automatically by > the Debian DVD when I do not mark the graphical interface. An excellent news. Let's hope that Debian continues to behave this way. > Then I edited /etc/network/interfaces and restart the system, to see > if the ip was changed. And yes, it worked. So this file is still being > used in some way - humm that seems to be the correct place to > configure my interface isn't? So, in plain English, you have ifupdown installed. > Well, I edited it again, now to try change the ip but without > restarting the system - like in all servers real situation, you don't > want to go after the server. First of all I tried all this solutions: > > #systemctl daemon-reload > (nothing happens ip didnt change) And it should not. ifupdown provides networking service, and "daemon-reload" forces systemd to re-read own config only, not to restart services. > #systemctl networking restart > (the ip did not change and the connection breaks you need to go > phisicial and ip address shows the same ip configured) "networking restart" invokes "ifdown" first, and "ifup" next. And systemd restarts most of the network services by dependency. The problem is - once they invoke ifdown, there's no network connection to invoke "ifup", hence you have no network-listening services. Do it from BMC/ILOM/HMC (it's not a server if it does not have one), or install screen/tmux and do it from there. > #nmcli connection reload > # nmtui Please leave these Finely Designed Executables™ to a desktop installation. > The solution that i found is this (but i think that is not a real solution): > > ip address delete theoldip/mask dev enp0s3 > ifdown enp0s3 > ifup enp0s3 ifdown/ifup combo should be sufficient. Just don't do it via plain ssh connection. > But if for example i do it partially, it does not work... for example: > > ip address delete theoldip/mask dev enp0s3 > ifup enp0s3 > The interface is already... blabla ifup/ifdown have their own special way to track which interface was touched by them. ifup will rightfully refuse to bring network interface UP if it considers that it's UP already. > > ifdown enp0s3 > ifup enp0s3 > (a new ip is added to the interface, but now two ips are configured > and the network is broken again) You mean, two new IPv6 ips? Can you provide a 'correct' and a 'broken' example via 'ip a l' please? And, while we're at it, the contents of /etc/network/interfaces? > Is that a bug? Short of the last part everything else can be classified as an operator's error, or a lack of knowledge. Reco