Re: [Babel-users] failing over faster?

>> Babeld is already monitoring netlink messages (see funciton filter_netlink >> in kernel_netlink), but it looks like this mechanism is not working in >> your particular case. I'll try to reproduce the issue, but I don't own >> a Raspberry Pi. > Maybe the typical problem with IFF_UP and

Re: [homenet] Updating DNS

>>> Just 2136 isn't enfough, because there's no authentication scheme, >> I don't understand this argument. How is non-secured DDNS any less secure >> than mDNS? What am I missing? > This is an implementation issue, not a security issue--sorry for not making > that clear. In order to

Re: [Babel-users] failing over faster?

> http://stackoverflow.com/questions/7225888/how-can-i-monitor-the-nic-statusup-down-in-a-c-program-without-polling-the-ker Babeld is already monitoring netlink messages (see funciton filter_netlink in kernel_netlink), but it looks like this mechanism is not working in your particular case. I'll

Re: [Babel-users] failing over faster?

> # and we fail over in 32 seconds What happens if you apply the following patch? diff --git a/babeld.c b/babeld.c index 3127e72..0183b32 100644 --- a/babeld.c +++ b/babeld.c @@ -744,7 +744,7 @@ main(int argc, char **argv) if(timeval_compare(_interfaces_timeout, ) < 0) {

Re: [homenet] Site-local multicast [was: Updating DNS]

> Now you get me curious. How do you do efficient site-local multicast > when you have multiple wifi and ethernet links? Assuming you've got only transitive links, then any multicast routing protocol should work fine at the scale we envision for Homenet as long as it is able to avoid wireless

Re: [homenet] Updating DNS [was: How many people have installed the homenet code?]

> Juliusz, the problem is that existing home network devices that do > DNS-based service discovery do not support DNS update. They could, but > they don't, because we didn't define an easy way for them to do it. I'd be grateful if you could expand on that. Why can't we define a way for clients

[homenet] Updating DNS [was: How many people have installed the homenet code?]

>> Do you mean, (1) how is a DNS resolver advertised to clients, or >> (2) how clients are registered in DNS ? >> >> (1) is done by using the -N flag on the router advertising an external >> connection (-E). This flag can be repeated multiple times. > hnetd grabs this automatically from

Re: [homenet] How many people have installed the homenet code?

> I’m starting by running shncpd on a boundary router and tried a trivial > installation. Excellent, thanks. > I don’t see how dns gets updated. Are such updates out of scope of > shncpd? Do you mean, (1) how is a DNS resolver advertised to clients, or (2) how clients are registered in DNS ?

Re: [homenet] How many people have installed the homenet code?

> It’s the IPv6 stuff that I’m most interested in - IoT with ipv4 (NAT) is > too expensive to support at scale. At least for the retail market. My > sense was that the api to the openwrt tools was a likely stumbling > block. Tim, please try shncpd and let me know how it goes. I'm quite willing

Re: [Babel-users] route.c:1071: possible coding error ?

> So the for loop in i only does even numbered slots. I am not sure if this is > intentional or not. If it is, a comment might be a good idea. It's a bug. It was fixed on 14 March. -- Juliusz ___ Babel-users mailing list

Re: [Babel-users] ad-hoc networks behave like mesh using babel?

> 1) I waited for almost 10 minutes but Babel does not seem to recover > from the ! routes automatically. It does so only when I restart the > babel daemon That's not normal. If the state file is lost, babeld should recover after 200 seconds (3min 20s). If you manage to get it stuck again,

[homenet] HNCP question: multiple delegations for same prefix

Hi, Is it legal for multiple HNCP routers to announce the same delegated (IPv6) prefix? I assume it is legal, and that nodes should assign a single prefix per delegated prefix to attached links (the spec says "set of delegated prefixes", not "multiset"), but I'd like to be sure I'm not

Bug#821405: dhcpcd no longer starts wpa_supplicant

Package: dhcpcd5 Version: 6.10.1-1 Since upgrading to 6.10.1-1, wpa_supplicant is no longer automatically started when I run dhcpcd. Everything works fine if I start wpa_supplicant by hand. I'm starting dhcpcd by hand, not from an init script, and the version of wpa_supplicant is 2.3-2.3.

Re: [Babel-users] babeld crashes

> if(channels_len == 0) { Not necessary, free(NULL) is a nop. ___ Babel-users mailing list Babel-users@lists.alioth.debian.org http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/babel-users

Re: [Babel-users] babeld crashes

> I think the attached patch solves this issue. Well spotted, thanks. Applied. (This is only relevant to the current head, people running 1.7.* are not affected.) -- Juliusz ___ Babel-users mailing list Babel-users@lists.alioth.debian.org

Re: [Babel-users] babeld crashes

Dave, could you please try to reproduce this under valgrind? ___ Babel-users mailing list Babel-users@lists.alioth.debian.org http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/babel-users

Re: [Babel-users] IPV6_SUBTREES vs policy routing

> Subject: [PATCH] Fix bug allowing the comparison of v4 and v6 prefixes. Applied, to both the head and the 1.7 branches. I cannot test right now, so I hope you know what you're doing. Do you think this is serious enough to justify releasing 1.7.2? Matthieu, could you please tell me when the

Re: [Babel-users] Question about source specific routing

>> 100:from 10.0.100.0/16 lookup 10 > Not the actual problem, but this should be 10.0.0.0/16 instead of > 10.0.100.0/16. Is your A configuration file correct? Hmm, looks like there's a call to mask_prefix missing somewhere. (Note to self -- rename mask_prefix to normalise_prefix.) --

Re: [Babel-users] [babel] Babel web pages back up

> Thanks for the information. I've filed a ticket with our network > administration people, but it's Saturday today. ...and they fixed it straight away. I'm under shock. ___ Babel-users mailing list Babel-users@lists.alioth.debian.org

Re: [Babel-users] IETF news

> attending remotely was not quite satisfactory. We'll need to think about how to make the babel WG as friendly to remote participation as possible. Obviously, having a competent Jabber scribe is important, one that ensures that Jabber questions get asked at the mike in timely manner. Anything

Re: [Babel-users] IETF news

> IETF is boring without Markus. And without Henning. And without Steven. And without Dave. (But I'm in touch with Dave more often than the other three.) ___ Babel-users mailing list Babel-users@lists.alioth.debian.org

[Babel-users] IETF news

Hi, IETF is over, it was a lot of fun. I think that the Babel BoF was a success, we had what looked like a hundred people or so, many thanks to Joel and Tony for chairing, and many, many thanks to Toke and Denis for agreeing to come and give talks about their work. One of the requirements for a

Re: [Babel-users] [babel] Babel web pages back up

> Try with IPv6: Thanks for the information. I've filed a ticket with our network administration people, but it's Saturday today. The new address is: https://www.irif.univ-paris-diderot.fr/~jch/software/babel/ but don't get used to it, since I don't count on our hierarchy to understand

Re: [Babel-users] [babel] Babel web pages back up

> http://www.pps.univ-paris-diderot.fr/~jch/software/babel/ is still a 404 > for me. Strange. For me, it redirects cleanly: $ curl -i http://www.pps.univ-paris-diderot.fr/~jch/software/babel/ | sed -n -e '1p' -e '/^Location:/p' HTTP/1.1 301 Moved Permanently Location:

[Babel-users] Babel web pages back up

Thanks to all who notified me about the Babel web pages being down -- our web server was moved. I think I have migrated everything to the new server, please let me know if you see anything amiss (broken links, incorrect MIME types, missing redirects, etc.). Sorry for the incompetence, --

[Babel-users] Fwd: [homenet] The HOMENET WG has placed draft-chroboczek-homenet-babel-profile in state "Call For Adoption By WG Issued"

--- Begin Message --- The HOMENET WG has placed draft-chroboczek-homenet-babel-profile in state Call For Adoption By WG Issued (entered by Ray Bellis) The document is available at https://datatracker.ietf.org/doc/draft-chroboczek-homenet-babel-profile/ Comment: As discussed during the WG

[Babel-users] Reminder: Babel BoF this afternoon, online participation welcome

Dear all, This is just to remind you that the IETF Babel BoF will happen later today, at 15:50 Argentinian time (UTC-3) 18:50 UTC 20:50 Western European Time (UTC+2) in room Buen Ayre C. You can participate remotely, https://ietf.org/meeting/95/remote-participation.html I've had good

Re: [Polipo-users] "502 Couldn't parse server headers"

> I think that the one in violation of the specs is the server in this > case. Right, thinko. But that's not the issue I'm worried about. If the client says "Connection: Upgrade", we correctly discard the Upgrade. On the other hand, if the client says just "Upgrade: foo" with no suitable

Re: [Polipo-users] "502 Couldn't parse server headers"

>> (I am generally opposed to the death penalty, but whoever decided to split >> the spec into 6 RFCs deserves to be shot.) > How forcing them to use nothing but `ed` to do all their editing? That would constitute cruel and unusual punishment. -- Juliusz

Re: [Polipo-users] "502 Couldn't parse server headers"

commit 726e1390f1a0978f440ce8b216d7e54f37212da0 Author: Juliusz Chroboczek <j...@pps.univ-paris-diderot.fr> Date: Wed Mar 30 16:22:49 2016 +0200 Don't break on Upgrade headers. Thanks to voyageur and Black

Re: [Polipo-users] "502 Couldn't parse server headers"

>> h2c Upgrades are ignored. >> Any other Upgrade content leads to a failed parse like in the >> previous behaviour. > Shouldn't all unknown upgrade requests be "ignored" (tho logged at some > appropriate info/warning level)? Right. What should be done if there's no "Connection: upgrade"

Re: [Polipo-users] "502 Couldn't parse server headers"

> I have detected that my installed version of Polipo throws the message > "502 Couldn't parse server headers" when I try to browse the website > "http://richard-falken.com;. The website works ok when I test it with a > non-proxied web browser. > I have noticed the server response contains

Re: [Polipo-users] My polipo doesn't write anything to disk any more!

Stefan, I'm at a loss here. I've just checked that it works for me, for what it's worth. Are you running from git head? Could you perhaps put a breakpoint at diskcache.c line 173, to see what is the result of checkRoot?

Re: [Polipo-users] My polipo doesn't write anything to disk any more!

>> If you run polipo as root, and use it as a proxy, it doesn't fill the >> cache directory, correct? > Correct. Could you please try strace -o foo ./polipo Polipo doesn't write out until it runs out of memory, so you should see a flurry of writes when you kill the Polipo process. --

Re: [Polipo-users] My polipo doesn't write anything to disk any more!

> Ping? Hello? Anybody there? I'm genuinely sorry, Stefan. Polipo has the lowest priority among my projects, and I'm really busy with more important stuff right now. Your message has been noted and marked important, but I don't intend to look at it in more detail in the coming weeks. (And

Re: [Babel-users] Babel on itty bitty boxes

> "Autodetection" is just based on the kernel version (via uname). So > it's normal it doesn't change anything on c1 (3.10 < 3.11). And it's > normal too that you need to use it on c2 (3.14 > 3.11, but compiled > without IPv6_SUBTREES). Matthieu, Why don't you write to the netdev list to ask

Re: [Babel-users] Ready for 1.8.0?

> A) I take it that "diversity routing" now matches the draft but does > not use the right stuff to get the channel(s)? You mean for master mode? Yes, it's still broken. ("We don't break userspace", they said.) I agree, it's annoying. > /me hides. I know, patches gladly accepted, (but ooh,

[Babel-users] Routes with source ::/128 [was: Investigating kernel's cache]

> From 3fdbb1f797ee9fe9260af92f5d7ea760684cd271 Mon Sep 17 00:00:00 2001 > From: Steven Barth > Date: Tue, 18 Feb 2014 13:18:32 +0100 > Subject: [PATCH] Allow routes with source ::/128 for SAS on Linux I don't understand this patch. Shouldn't we be avoiding announcing such

[Babel-users] Babel BoF in Buenos Aires

Hi, For those of you who haven't been following the babel@ietf mailing list: - there will be a Babel BoF at the IETF meeting in Buenos Aires: http://tools.ietf.org/bof/trac/#Routing - tentatively scheduled for Monday 4 April at 15:50 GMT-3, but please check again; - remote

Re: [Bloat] review: Deployment of RITE mechanisms, in use-case trial testbeds report part 1

>> Oh, and of course HAS is in itself a hack to work around badly managed >> queues in the network. In a nicely fairness queued world, we could do >> away with HAS entirely and just, y'know, stream things at the desired >> rate... Perhaps I'm missing something -- how do you to pick the right rate

Re: [Babel-users] IPv4: Default Route stability

>> Hysteresis is applied to route selection. The amount of hysteresis is >> controlled with the parameter smoothing-half-life (-M), 4s by default. >> Increase this parameter to have higher stability at the cost of slower >> convergence. > We're going to experiment with that. Cool, please report

Re: [Babel-users] Routs: Neither import table, nor export table, but exported?

> IMHO it might be helpful to clarify the manpage: I'm making a lot of efforts to make the manual page as useful as possible, but it's always difficult to guess what's going to be unclear for the users. I'll be very grateful if you can submit a patch to the manual page. -- Juliusz

Re: [Babel-users] Routs: Neither import table, nor export table, but exported?

> The manpage does not does not say - and I don't know, yet: > - What's the default? The default is RT_TABLE_MAIN, for both export- and import-table. > - If set, is it exclusive? Will routes in id=254 (linux, main) be > redistributed, too? If you set either import-table or export-table, the

Re: [Babel-users] IPv4: Default Route stability

> How stable is default route selection here? Since NAT-Sessions break > when using a different default route, a certain stability would be great. Hysteresis is applied to route selection. The amount of hysteresis is controlled with the parameter smoothing-half-life (-M), 4s by default. Increase

[Babel-users] Contributing to Babel IETF effort

Dear all, There's a discussion going on on babel@ietf about the upcoming Babel BoF, the goal of which is to form an IETF working group to standardise Babel. https://www.ietf.org/mailman/listinfo/babel If you think you could help -- either writing drafts or simply reviewing drafts (reading

Re: [Babel-users] babeld and systemd [was: babeld wishlist]

>> Awesome! Any chance of also including a 'babeld@.service' file to easily >> run babeld on a single interface without editing the configuration file? :) > ExecStart=/usr/bin/babeld -D -I /run/babeld-%I.pid -S /var/lib/babel-state-%I > -c /etc/babeld-%I.conf %I Perhaps I'm confused, but are

[Babel-users] ANNOUNCE: babeld-1.7.1

Dear all, Babeld-1.7.1 is available from http://www.pps.univ-paris-diderot.fr/~jch/software/files/babeld-1.7.1.tar.gz http://www.pps.univ-paris-diderot.fr/~jch/software/files/babeld-1.7.1.tar.gz.asc For more information about babeld and the Babel routing protocol, please see

Re: [Babel-users] ANNOUNCE: babeld-1.7.0

Er, no, I've just checked POSIX, and the following makes more sense. (Why can't I just say "#include " and be done with it?) diff --git a/rule.c b/rule.c index 4845802..ba70baf 100644 --- a/rule.c +++ b/rule.c @@ -24,6 +24,7 @@ THE SOFTWARE. #include #include #include +#include

Re: [Babel-users] [PATCH] Add support for unix-domain local socket

> I'm not sure it makes sense to accept local-path*/local-port* changes > from the local interface itself. Doing that means I'll unlink the wrong > path on exit. The same is already true for pid-file, though. Right. We also need to call reopen_logfile when the logfile changes. Interestingly,

Re: [Babel-users] ANNOUNCE: babeld-1.7.0

Thanks, Gabriel. I've just cut a stable branch babeld-1.7-branch with this fix. Let me know when you've checked everything works, and I'll release 1.7.1. > However, is there any reason why you don't want to put the #include in > babeld.h instead? That would pull it automatically in every .c

Re: [Babel-users] [PATCH v2] Add support for unix-domain local socket

Looks good. Applied, not tested yet. Thanks, Julien. -- Juliusz ___ Babel-users mailing list Babel-users@lists.alioth.debian.org http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/babel-users

Re: [Babel-users] babeld and systemd [was: babeld wishlist]

> Is it possible to tell babeld *not* to write a PID file at all? babeld -I '' Hmm, looks like it's not documented. ___ Babel-users mailing list Babel-users@lists.alioth.debian.org http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/babel-users

Re: [Babel-users] [PATCH] Add support for unix-domain local socket

Thanks, Julien. > I guess you'll tell me you don't put spaces between "if" and "(", and I > should use consistent style. I can easily edit that myself. > case 'g': This case is very similar to 'G' below. Please refactor to share the two cases (with an if inside). > +

Re: [Babel-users] Adding and removing interfaces dynamically in babeld

>> So if the parameter to -g/-G is not numeric, I make a Unix domain socket? >> > Do you prefer pull requests on github or patches on the list? Slight preference for patches to the list, but either is fine. Depends on where you want the discussion to happen. -- Juliusz

Re: [Babel-users] Adding and removing interfaces dynamically in babeld

> I noticed that "add interface" is new. Is there any other new token that > can now appear after "add"? I need a complete list to make babelweb > happy. Yeah, I've noticed. There's nothing more right now, but I reserve the right to add new stuff without bumping the version number (in

Re: [Babel-users] Adding and removing interfaces dynamically in babeld

>> Uh-huh. Shall we make -g read-only, and require -G for changes? > > -G seems to follow the principle of least surprise for updating existing > deployments. commit 3208c75b56e653cc8fac4fddfc9c306a879cffde Author: Juliusz Chroboczek <j...@pps.univ-paris-diderot.fr> Date:

Re: [Babel-users] Adding and removing interfaces dynamically in babeld

> -g for read-only, -G for read-write, Uh-huh. > and support for unix domain sockets seems like a good combination... So if the parameter to -g/-G is not numeric, I make a Unix domain socket? -- Juliusz ___ Babel-users mailing list

Bug#814380: Please add port 6696/udp (Babel routing protocol) to /etc/services

Package: netbase Version: 5.3 Severity: wishlist Port 6696/udp has been assigned by IANA to the Babel routing protocol. This port is used by Debian's "babeld" package. I'll be grateful if you can add it to Debian's /etc/services.

Re: [Babel-users] Adding and removing interfaces dynamically in babeld

> 1) Is there a recommended method for tunnelling to this? > > dair-1120:~ d$ ssh -W ip6-localhost:33123 admin@wherever That's what we do. > 2) It is worrisome that now random users on the box can make changes > without root perms. Uh-huh. Shall we make -g read-only, and require -G for

Re: [Babel-users] Adding and removing interfaces dynamically in babeld

> 1) Is there a recommended method for tunnelling to this? > > dair-1120:~ d$ ssh -W ip6-localhost:33123 admin@wherever That's what we do. > 2) It is worrisome that now random users on the box can make changes > without root perms. Uh-huh. Shall we make -g read-only, and require -G for

Re: [RFC v4] mac80211: add A-MSDU tx support

> One iwl user has reported that AMPDUs are disabled by default on iwl, That was me. > and had to do this: > > options iwlwifi 11n_disable=8 > > For a 4fold reduction in latency under load in his tests, iwl to ath9k. No, it was ath9k to iwl. I don't understand it either: enabling AMPDUs on the

Bug#813595: Upstream version 6.10.1 available

Package: dhcpcd5 Version: 6.9.3-1 Severity: minor Hi, Right now, the latest upstream is 6.10.1. This version fixes an unpleasant IPv6 related bug (duplicate addresses) that is triggered by recent OpenWRT routers. Please upgrade. -- Juliusz

[Babel-users] ANNOUNCE: babeld-1.7.0

Dear all, Babeld-1.7.0 is available from http://www.pps.univ-paris-diderot.fr/~jch/software/files/babeld-1.7.0.tar.gz http://www.pps.univ-paris-diderot.fr/~jch/software/files/babeld-1.7.0.tar.gz.asc For more information about babeld and the Babel routing protocol, please see

[Bloat] STARTTLS [was: nearly 5 years of bufferbloat.net]

> http://the-edge.taht.net/post/starttls_considered_helpful/ Did you bounce mail when the first MX contacted didn't do STARTTLS, or did you bounce when none of the MXes for a domain supported it? In other words, did you treat lack of STARTTLS as a transient or permanent error? -- Juliusz

Re: [OpenWrt-Devel] BGP-LS support in OPENWRT

> It's been great to see how some new protocols like HomeNet[1] get > supported in latest Trunk of OpenWRT and I was wondering if there was any > plan or possibility to do the same for BGP-LS and PCEP? Iben, The routing package feed is part of the default configuration of OpenWRT. Both Bird and

Re: [Polipo-users] MinGW build with IPv6

This patch is corrupt (a hunk header is missing). I could fix it up manually, but I'm wondering if there's anything else that's missing. -- Juliusz -- Site24x7 APM Insight: Get Deep Visibility into Application

Re: [PATCH v2 2/2] Add the Babel protocol.

Nice. > + tx length ; Please don't make this configurable, it only breaks interoperability. Section 4 of RFC 6126 is very clear about the maximum size of a Babel packet: It MUST NOT send packets larger than the attached interface's MTU (adjusted for lower-layer headers) or 512

Bug#748901: dma: does not recognize addresses in MIME enocded-word syntax

retitle 748901 dma -t fails to parse folded header lines thanks Thanks for the debugging, Olaf. The following snippet reproduces the bug: -- echo | dma -t -i < To: L. User Subject: this breaks dma This breaks dma. EOF

Re: [homenet] [Babel-users] Detecting bridges

> Is there room in the protocol for a router to announce what link type it > is on? This could be carried by a sub-TLV of Hello (or a sub-TLV of IHU if you want to make it per-host). > I.e., a router on wifi announces wifi and when a router that is on wired > receives an announcement from a

Re: [Babel-users] [homenet] Detecting bridges

> Is there room in the protocol for a router to announce what link type it > is on? This could be carried by a sub-TLV of Hello (or a sub-TLV of IHU if you want to make it per-host). > I.e., a router on wifi announces wifi and when a router that is on wired > receives an announcement from a

Re: [Babel-users] Some thoughts about Babel (feature requests)

> We forked babeld and have included an ipc process which allows us to add > and remove interfaces while babeld is running: > github.com/sudomesh/babeld Nice. My plan is a little different: I want a text interface over the existing TCP local interface, so that it's easily scriptable -- so that

Re: [homenet] HNCP: interaction with routing protocol?

> If you're announcing an external connection into the HNCP domain, shncpd > will install a proto 43 source-specific default route. See route_externals > in prefix.c. In case that wasn't clear -- shncpd doesn't act as a DHCPv6-PD or DHCPv4 client, it expects you (the human operator or the

Re: [homenet] HNCP: interaction with routing protocol?

>> Shncpd has closer binding to the routing protocol, it marks its routes >> as "proto 43" and expects the routing protocol to redistribute just >> that; shncpd also occasionally inserts dummy "proto 43" routes into the >> kernel, just so that they get redistributed into the routing protocol. >

Re: [homenet] HNCP: interaction with routing protocol?

> hnetd does address configuration on interfaces, the routing protocol picks > this up because that's how it's configured...? Hnetd doesn't communicate > directly with the routing protocol at all, right? It just sets up the > landscape so the routing protocol can come and survey it and communicate

Re: [homenet] [Babel-users] Detecting bridges

Added Mikael and the Homenet list to CC. Homenet, the issue we're dealing with is that babeld performs badly when there is a transparent wireless bridge connected to a wired interface: the interface is treated as a lossless wired interface, and if it suffers packet loss, there is repeated link

Re: [homenet] New Version Notification for draft-barth-homenet-wifi-roaming-00.txt

> here is some attempt to formalize a simple WiFi roaming approach > using host routes and a stateless proxy for DAD NDP messages. http://www.cisco.com/en/US/products/ps9390/products_white_paper09186a00800a3ca5.shtml (See also RFC 1925 Section 2.11.) -- Juliusz

Re: [Babel-users] latency in WLAN-SI

> In pouring through the astonishing *wealth* of data available via > nodewatcher, Yes, it's pretty amazing. > so are you really seeing real-world peaks in the 7 second range or is > that an artifact of something else? Looking from 4 hops away, I'm seeing a reasonable amount of route

Re: [Babel-users] babeld wishlist

> ** Version output in the monitoring interface Done. ___ Babel-users mailing list Babel-users@lists.alioth.debian.org http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/babel-users

Re: [Babel-users] Babel instability in WLAN-SI

> Doesn't a lost packet contribute to RTT with (an appropriate value of) > infinity? Not in Babel. RTT and packet loss are accounted for separately, and RTT is just a smoothed, exponential average of the packets that weren't lost. -- Juliusz ___

[Babel-users] Wlan-SI: please disable split-horizon on the hubs

Mitar, Jernej, It looks like you're running a point-to-multipoint topology on your VPN servers; this breaks the assumption made by babeld that wired links are transitive. Please disable split-horizon: interface tun42 split-horizon false You only need to do that on the hub, not on the spokes.

Re: [Babel-users] Babel instability in WLAN-SI

And while I have your ear, could you please display the reachability field in hex? It's a bitmap, so it's difficult to read in decimal. ___ Babel-users mailing list Babel-users@lists.alioth.debian.org

Re: [Babel-users] Babel instability in WLAN-SI

> If I look at the nodewatcher database, I see that there's a number of > wireless links that were incorrectly detected as wired. For example, if > I look at node urban-1: > > https://nodes.wlan-si.net/node/abe997c1-5f28-4d5f-8335-b3af182b0de4/ > > I see that most of the links have a TX-cost of

Re: [Babel-users] Babel instability in WLAN-SI

> I remember a case in Vienna where an Olsr daemon was its own neighbor, > because (over a bridged connection) one of its 2.4 GHz neighbors > bridged everything to a 5 GHz interface the local node could hear. That's not the issue here -- a Babel node is allowed to be its own neighbour, and this

[Babel-users] Multiple interfaces [was: Bucket full, dropping packet]

> I have VPS running babel and tunneldigger which is part of Wlan Slovenia > network. If it helps I can give you [...] root access to this server Thanks a lot, Valent. It's interesting. When the network is stable, the traffic is normal, with the periodic full update taking 5 full-size packets.

Re: [Babel-users] Bucket full, dropping packet

> The attached patch should solve the problem. Nice, applied. > +if(!check_specific_first()) > +fprintf(stderr, "Invariant failed: specific routes first in RIB.\n"); #ifdef DEBUG, or if(debug_level >= 2) ? -- Juliusz ___ Babel-users

Re: [Babel-users] Babel instability in WLAN-SI

> So luckily we do know which links are wireless and which are VPN, so > we could configure that in the Babel configuration. Cool. > About VPN, I think ideally, we would want a metric for VPN links which: > - take RTT into the account Already implemented, as you know. > - take packet loss into

Re: [Babel-users] Babel instability in WLAN-SI

> BTW, how should VPN links be handled in this case? They are currently > marked as wired, but they can also experience packet loss. Does this > mean that bad VPN links can also cause huge amounts of control traffic? It depends. How likely are they to lose two Hellos in a row? Babeld marks a

Re: [Babel-users] Bucket full, dropping packet

>> Ok, I'll see on Monday if I can get an extra VM before Christmas. > > Which VM system are you using? We might be able to generate you a > ready-made image. Please don't -- I'll let our system administrators clone their usual VMWare image, it's better for everyone if I use what they're familiar

Re: [homenet] HNCP: interaction with routing protocol?

>> I'm probably missing something -- but where in the HNCP document does it >> say that applied prefixes must be announced over the routing protocol? >> I don't see it in Section 6.3.3. > I'd say the role of HNCP and the routing protocol and what does what, is > not really specified anywhere.

Re: [Babel-users] Bucket full, dropping packet

>> Oh, sadly not 75 max. The number of neighbours of one node is >> potentially very large, because all nodes with Internet uplink >> connects to the VPN server. So, all those nodes are then neighbours of >> that VPN server node. Currently this is for example 140 nodes on one > So yes, 75 is the

Re: [Babel-users] Bucket full, dropping packet

> Hm, I thought that Babel was tested on large networks and that it was > tested on simulated large networks? Babel has been tested in one large network (1500 nodes). It took 20 minutes to converge, and worked fine after convergence time. I fixed the initial convergence issues (too many

Re: [Babel-users] Bucket full, dropping packet

> Attaching some more dumps. Thanks, Kosko. These are all IPv4 retractions, and they look just fine to me. I'd need to see the full packet to be sure, but it probably means that there's some ambiguity in the code about handling retractions with no suitable next hop. I'll have a look when I

Re: [Babel-users] babeld wishlist

Dear Dave, Here, at Babel Towers, we are busy working on Babel, working on babeld, lecturing undergrads, correcting exams, writing papers, and protesting against the stupidity of the current state of emergency. We have all but given up on having a life outside of work. You will therefore

Re: [Babel-users] babeld and systemd [was: babeld wishlist]

> Systemd is your friend! It's more like family. You can choose your friends. -- Juliusz ___ Babel-users mailing list Babel-users@lists.alioth.debian.org http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/babel-users

Re: [homenet] Protocol Action: 'Home Networking Control Protocol' to Proposed Standard (draft-ietf-homenet-hncp-10.txt)

> The IESG has approved the following document: > - 'Home Networking Control Protocol' > (draft-ietf-homenet-hncp-10.txt) as Proposed Standard Yay! (Congratulations to Markus, Stephen and Pierre, and also to Mark and everyone else who sweated water and bl -- Juliusz

Bug#807430: Firewalld depends on systemd

Package: firewalld Version: 0.3.14.2-2 Severity: wishlist Firewalld depends on systemd, which makes it unusable for those of us who run SV init on their systems. Would it be a lot of work to make firewalld not depend on a specific init system?

[Pkg-utopia-maintainers] Bug#807430: Firewalld depends on systemd

Package: firewalld Version: 0.3.14.2-2 Severity: wishlist Firewalld depends on systemd, which makes it unusable for those of us who run SV init on their systems. Would it be a lot of work to make firewalld not depend on a specific init system? ___

[Pkg-utopia-maintainers] Bug#807430: closed by Michael Biebl <bi...@debian.org> (Re: Bug#807430: Firewalld depends on systemd)

reopen 807430 thanks $ sudo apt-get install --no-install-recommends firewalld [...] The following NEW packages will be installed: cgmanager firewalld libapparmor1:i386 libcap2:i386 libcgmanager0 libcryptsetup4:i386 libdevmapper1.02.1:i386 libgcrypt20:i386 libgpg-error0:i386 libkmod2:i386

Bug#807430: closed by Michael Biebl <bi...@debian.org> (Re: [Pkg-utopia-maintainers] Bug#807430: Firewalld depends on systemd)

reopen 807430 thanks $ sudo apt-get install --no-install-recommends firewalld [...] The following NEW packages will be installed: cgmanager firewalld libapparmor1:i386 libcap2:i386 libcgmanager0 libcryptsetup4:i386 libdevmapper1.02.1:i386 libgcrypt20:i386 libgpg-error0:i386 libkmod2:i386

Re: [Babel-users] Bucket full, dropping packet

> It is around 550 routes and there is one node, which currently has 75 > neighbours. Excellent. > Is this be enough to trigger the limits? Yes, that's expected. Please increase the limits, be bold, multiply them by 20. -- Juliusz ___ Babel-users

<    5   6   7   8   9   10   11   12   13   14   >