[gentoo-user] OT: BATMAN vs frr/ospf
Hi all, has anyone had experience using the batman-adv protocol and can comment on its use instead of ospf? The recommended "drop in" replacement for quagga/ospf based routing with the frr/ospf package has proven to be a less than stellar replacement in my case (not really frr's fault, but it is not identical to quagga and my requirements are complex) so I am looking to jump ship to batman. I am currently building kernels and vm's to test but I would appreciate comments from someone who has done this already. My networks include ~10-15 vlans that extend across (open)vpn tunnels and multiple wifi SSID's and have a number of potential looping scenarios that ospf manages. I use zeroconf (for homeassistant) and have lxc based instances using veth interfaces for services (asterisk, web, dns, ...). There is a moosefs data store on its own switch and two dedicated vlans. I have in excess of 30 devices on the network and ESP IoT devices are multiplying like rabbits (!) All non-esp or android phone systems use gentoo on arm32/arm64/intel, run shorewall, have multiple vlans via trunking or multiple interfaces in different vlans or in some cases up to 4 interfaces bonded for throughput. I am using d-link managed switches and a homebrew AP using hostapd in the 2.4 and 5g bands. Using quagga/ospf was mostly stable and just worked. While I could try tuning frr to work more reliably (worst problems are not staying converged, convergence time (which sometimes kills vm's via the moosefs data store disappearing off the network for minutes at a time), fighting frr's interference in ip forwarding across multiple interfaces and excessive overhead as it never seems to settle for long). I am thinking the effort might be better spent on batman - I am attracted to the supposedly fast convergence, minimal overhead and the potential of meshes (IoT) using the flat routing overlay it implements. Questions I have are: 1. easily works with shorewall 2. it actually does have fast and glitch free convergence 3. internetworking across a VPN based WAN with batman at either end 4. mesh hot spot control 5. any other gotchas? BillK
Re: [gentoo-user] eudev/udev changeover: a warning to Linode customers
On Wed, Dec 1, 2021 at 8:05 PM Rich Freeman wrote: > > On Wed, Dec 1, 2021 at 7:15 PM Dale wrote: > > > > > > root@fireball / # ls -al /etc/udev/rules.d/ > > -rw-r--r-- 1 root root 1903 Apr 4 2012 70-persistent-cd.rules > > -rw-r--r-- 1 root root 814 Jan 1 2008 70-persistent-net.rules > > -rw-r--r-- 1 root root0 Mar 22 2015 80-net-name-slot.rules > > > > I can't recall which of the two about net it is tho. Thing is, it seems > > to work with eudev but not udev. > > With udev the filenames you want are: > 80-net-name-slot.rules > 80-net-setup-link.rules > > Or at least, that is what I am using with the systemd-bundled udev and > my physical interface is eth0. Disregard that. I'm also using net.ifnames=0 - I'm guessing the filename changed at some point. You probably can dig around in the package-supplied udev rules to figure out which one needs to be overridden now. -- Rich
Re: [gentoo-user] eudev/udev changeover: a warning to Linode customers
On Wed, Dec 1, 2021 at 7:15 PM Dale wrote: > > > root@fireball / # ls -al /etc/udev/rules.d/ > -rw-r--r-- 1 root root 1903 Apr 4 2012 70-persistent-cd.rules > -rw-r--r-- 1 root root 814 Jan 1 2008 70-persistent-net.rules > -rw-r--r-- 1 root root0 Mar 22 2015 80-net-name-slot.rules > > I can't recall which of the two about net it is tho. Thing is, it seems > to work with eudev but not udev. With udev the filenames you want are: 80-net-name-slot.rules 80-net-setup-link.rules Or at least, that is what I am using with the systemd-bundled udev and my physical interface is eth0. -- Rich
Re: [gentoo-user] eudev/udev changeover: a warning to Linode customers
On Wed, 01 Dec 2021 19:15:36 -0500, Dale wrote: > > Neil Bothwick wrote: > > On Wed, 1 Dec 2021 17:27:32 -0600, Dale wrote: > > > >> What made this affect me, I think the method is different to disable it > >> in udev than it is in eudev. > > net.ifnames=0 works on both udev and eudev, I've had it in my GRUB config > > for years and it needed no changes when switching from eudev to udev. > > > > > > > I'm pretty sure mine is done with a udev rules file. I never had mine > on the kernel line. This is the list of rules files I have: > > > root@fireball / # ls -al /etc/udev/rules.d/ > total 20 > drwxr-xr-x 2 root root 4096 Nov 28 13:29 . > drwxr-xr-x 4 root root 4096 Nov 28 13:29 .. > -rw-r--r-- 1 root root 2064 Apr 27 2021 69-libmtp.rules > -rw-r--r-- 1 root root 1903 Apr 4 2012 70-persistent-cd.rules > -rw-r--r-- 1 root root 814 Jan 1 2008 70-persistent-net.rules > -rw-r--r-- 1 root root 0 Mar 22 2015 80-net-name-slot.rules > -rw-r--r-- 1 root root 0 Nov 27 17:53 .keep_sys-fs_udev-0 > root@fireball / # > > > I can't recall which of the two about net it is tho. Thing is, it seems > to work with eudev but not udev. I'd think it would but based on > experience, it doesn't. I guess if someone is switching a remote > machine, or any machine, and they want to be sure, add the option to the > kernel line to be safe. That may be a more dependable method. > > Either way, at least maybe these threads will help someone else avoid > the problem. I had this happen to me when I got this box, the 80.rules hack stopped working, and so I gave up and used the "predictable" names. I must have forgotten about the kernel command line parameter at the time. -- Your life is like a penny. You're going to lose it. The question is: How do you spend it? John Covici wb2una cov...@ccs.covici.com
Re: [gentoo-user] eudev/udev changeover: a warning to Linode customers
Neil Bothwick wrote: > On Wed, 1 Dec 2021 17:27:32 -0600, Dale wrote: > >> What made this affect me, I think the method is different to disable it >> in udev than it is in eudev. > net.ifnames=0 works on both udev and eudev, I've had it in my GRUB config > for years and it needed no changes when switching from eudev to udev. > > I'm pretty sure mine is done with a udev rules file. I never had mine on the kernel line. This is the list of rules files I have: root@fireball / # ls -al /etc/udev/rules.d/ total 20 drwxr-xr-x 2 root root 4096 Nov 28 13:29 . drwxr-xr-x 4 root root 4096 Nov 28 13:29 .. -rw-r--r-- 1 root root 2064 Apr 27 2021 69-libmtp.rules -rw-r--r-- 1 root root 1903 Apr 4 2012 70-persistent-cd.rules -rw-r--r-- 1 root root 814 Jan 1 2008 70-persistent-net.rules -rw-r--r-- 1 root root 0 Mar 22 2015 80-net-name-slot.rules -rw-r--r-- 1 root root 0 Nov 27 17:53 .keep_sys-fs_udev-0 root@fireball / # I can't recall which of the two about net it is tho. Thing is, it seems to work with eudev but not udev. I'd think it would but based on experience, it doesn't. I guess if someone is switching a remote machine, or any machine, and they want to be sure, add the option to the kernel line to be safe. That may be a more dependable method. Either way, at least maybe these threads will help someone else avoid the problem. Dale :-) :-)
Re: [gentoo-user] eudev/udev changeover: a warning to Linode customers
On Wed, 1 Dec 2021 17:27:32 -0600, Dale wrote: > What made this affect me, I think the method is different to disable it > in udev than it is in eudev. net.ifnames=0 works on both udev and eudev, I've had it in my GRUB config for years and it needed no changes when switching from eudev to udev. -- Neil Bothwick There are some micro-organisms that exhibit characteristics of both plants and animals. When exposed to light they undergo photosynthesis; and when the lights go out, they turn into animals. But then again, don't we all? pgpgFOgPtl8OS.pgp Description: OpenPGP digital signature
Re: [gentoo-user] eudev/udev changeover: a warning to Linode customers
Marco Rebhan wrote: > On Wednesday, 1 December 2021 20:44:48 CET Arve Barsnes wrote: >> I have been running eudev for as long as it has existed, and have also >> been using the predictable interface names more or less since they >> were introduced. The eudev ebuild also shows a message about this >> every single time you emerge it (with ewarn messages in pkg_pretend). >> This was apparently available in eudev within a month of the change >> in systemd. No one should be surprised by this. > Yeah, I was wondering why people were hitting this problem, the > predictable interface names have been in eudev as well for a > considerable while. I had them disabled with the same net.ifnames=0 that > others are mentioning now to get the old names (mainly since they're > just easier to remember). I was maybe thinking that there could be a > configuration option for it that didn't get changed on existing installs > when this was initially introduced in eudev, which would have explained > it since my installs aren't that old. But if it automatically used the > new names for you then I have no idea either. Nothing should have > changed in this regard with this update as far as I can tell... > > -Marco What made this affect me, I think the method is different to disable it in udev than it is in eudev. I had something set, not on the kernel line tho, to disable it on mine when using eudev. I think it is a udev rules file. Thing is, when I rebooted with udev installed, it ignored the method eudev uses and used the newer naming method. If the same option worked for both, then I would likely have seen no difference at all. I might add, that is what I was expecting and was surprised to find it not to be the case. When I switched from udev to eudev ages ago, I did nothing but remove udev and install eudev. That's it. I don't recall changing anything else but that was ages ago. I'm hoping others doing this switch will notice my thread and this thread to prevent them from switching and not realizing it can break things until it is configured correctly. Bad thing is, it breaks one thing that is needed to get help, the connection to the internet. If it is a remote machine, that is really bad. Let's hope this alerts others to double and maybe even triple check things before rebooting. Dale :-) :-)
Re: [gentoo-user] eudev/udev changeover: a warning to Linode customers
On Wed, 1 Dec 2021 at 20:28, Rich Freeman wrote: > I suspect they would have had similar issues with other distros, but > that would have been years ago when udev made the change and eudev > decided not to merge it. That dates to around the time eudev started. I have been running eudev for as long as it has existed, and have also been using the predictable interface names more or less since they were introduced. The eudev ebuild also shows a message about this every single time you emerge it (with ewarn messages in pkg_pretend). This was apparently available in eudev within a month of the change in systemd. No one should be surprised by this. Cheers, Arve
Re: [gentoo-user] eudev/udev changeover: a warning to Linode customers
On Wed, 2021-12-01 at 14:28 -0500, Rich Freeman wrote: > indeed being able to change this stuff is part of the appeal of > Gentoo. Besides, Gentoo users probably would want to be aware of it > anyway... Amen. I knew what I was signing up for. Just hoping to save someone else an unexpected trip through the recovery console. Learning things the hard way so that others won't have to!
Re: [gentoo-user] eudev/udev changeover: a warning to Linode customers
On Wed, Dec 1, 2021 at 2:17 PM Jigme Datse wrote: > > I honestly was afraid of this with running some updates lately. > Like... Not this specifically, but because someone was commenting > about eudev->udev causing problems... And then the problems I was > having a hard time with some updates on a relatively new instance, and > I noticed that was something that needed to be done for Gentoo > purposes... They do seem to be doing *somewhat* better for having > Gentoo workable than when I started looking at using them in part > because of that (yeah I know it's not recommended, and I don't really > recommend it for anyone else, but it's what I like). > I suspect they would have had similar issues with other distros, but that would have been years ago when udev made the change and eudev decided not to merge it. That dates to around the time eudev started. They may very well have done more hand-holding or mitigation for the other distros simply due to their popularity, and also uniformity. The distros themselves probably also did some mitigation around this change so that the average Ubuntu user who doesn't know what an "eth0" is wouldn't have to be aware of the change. With Gentoo a certain amount of this stuff just has to be dropped on the user simply because we can't be sure users didn't change things, and indeed being able to change this stuff is part of the appeal of Gentoo. Besides, Gentoo users probably would want to be aware of it anyway... -- Rich
Re: [gentoo-user] eudev/udev changeover: a warning to Linode customers
On Wed, 01 Dec 2021 12:49:59 -0500 "Matt Connell (Gmail)" wrote: > If you have a Gentoo machine running on Linode, take care to note that > the eudev => udev changeover requires some manual intervention before > the next reboot. > > You will need to DISABLE the network autoconfiguration option for the > VM, and edit /etc/conf.d/net to specify the new, 'predictable' network > interface name. This is because Linode's network autoconfigurator is > hard-coded to generate a configuration using the traditional style of > network interface names, eg. eth0 > > Alternatively, you can make whatever changes are required to have the > system enumerate the network interfaces with the old style names. > > If you do neither of those things, you will need to use the rescue > console in order to log into the machine and fix your network > configuration, as the system will not be able to bring up the network > on its next boot. Ask me how I learned this. > > I honestly was afraid of this with running some updates lately. Like... Not this specifically, but because someone was commenting about eudev->udev causing problems... And then the problems I was having a hard time with some updates on a relatively new instance, and I noticed that was something that needed to be done for Gentoo purposes... They do seem to be doing *somewhat* better for having Gentoo workable than when I started looking at using them in part because of that (yeah I know it's not recommended, and I don't really recommend it for anyone else, but it's what I like).
[gentoo-user] eudev/udev changeover: a warning to Linode customers
If you have a Gentoo machine running on Linode, take care to note that the eudev => udev changeover requires some manual intervention before the next reboot. You will need to DISABLE the network autoconfiguration option for the VM, and edit /etc/conf.d/net to specify the new, 'predictable' network interface name. This is because Linode's network autoconfigurator is hard-coded to generate a configuration using the traditional style of network interface names, eg. eth0 Alternatively, you can make whatever changes are required to have the system enumerate the network interfaces with the old style names. If you do neither of those things, you will need to use the rescue console in order to log into the machine and fix your network configuration, as the system will not be able to bring up the network on its next boot. Ask me how I learned this.
Re: [gentoo-user] Re: Switching from eudev to udev, disaster.
On 12/1/21 10:02 AM, Grant Edwards wrote: IIRC, there are situations where using udev rules to rename them "ethN" based on MAC addresses will fail because that can conflict with the low-level kernel names. Or something like that. I don't think I ever ran into a problem re-using the original kernel eth# names /as/ /long/ /as/ the target name wasn't currently in use. Sometimes I needed to vacate the target name before I could re-use it. -- Grant. . . . unix || die
[gentoo-user] Re: Switching from eudev to udev, disaster.
On 2021-11-30, Grant Taylor wrote: > I guess I never really gave the renaming much thought because I > almost always complied drivers into the kernel, which meant that > they had a consistent ~> predictable enumeration and naming order. I think that's generally true on most motherboards for PCI devices. Possibly not so much for USB. -- Grant
[gentoo-user] Re: Switching from eudev to udev, disaster.
On 2021-11-30, Grant Taylor wrote: > Besides, it's a LOT easier to /just/ `tcpdump -nni eth0` when logging > into a machine than it is to have to figure out the interface name first. Yep. I always add udev rules to rename the boards net0, net1, etc. based don the MAC addresses. > That being said, I was okay with what CentOS 6.x did, where the new name > was matched against the MAC address. I had eth0 based on MAC for > outside and eth1 based on MAC for inside on a number of systems. IIRC, there are situations where using udev rules to rename them "ethN" based on MAC addresses will fail because that can conflict with the low-level kernel names. Or something like that. -- Grant (the other one)