search for: gsliepen

Hello everyone, I am running into the following error messages everytime I try to use a docker overlay network on top of tinc-vpn: Error getting read result from Windows tap device {F30C422F-4524-435F-A15B-71A7E08C260D}: (995) The I/O operation has been aborted because of either a thread exit or an application request. Received packet of 106 bytes from ... (... port 655) Writing packet of 106

...r me to understand what is probably causing our issue (the syscall chain done on every UDP packet). Very interesting that you're able to see around 90% of a Gig line on bare metal. Were you ever able to make any further progress on adjusting Tinc based on the investigation in https://github.com/gsliepen/tinc/issues/110 ? Martin - Yeah, I was looking into Wireguard too. However, we're pretty embeded with Tinc right now (as we've been running it for a few years now) so, I'd love to be able to adjust our existing setup not get such a perf hit. - Jared On Wed, May 17, 2017, at 10:26 A...

Hi Etienne, Am 20.05.2018 um 10:58 schrieb Etienne Dechamps: > Hi Marc, > > A number of bugs have been found and fixed in the code that deals with > Windows devices in tinc 1.1: > > https://github.com/gsliepen/tinc/pull/169 > https://github.com/gsliepen/tinc/pull/173 > https://github.com/gsliepen/tinc/pull/174 > https://github.com/gsliepen/tinc/pull/181 > https://github.com/gsliepen/tinc/pull/182 Thanks for the listing. To me it looks like #174 and #181 might solve my issue. > Unfortunat...

Hi Marc, A number of bugs have been found and fixed in the code that deals with Windows devices in tinc 1.1: https://github.com/gsliepen/tinc/pull/169 https://github.com/gsliepen/tinc/pull/173 https://github.com/gsliepen/tinc/pull/174 https://github.com/gsliepen/tinc/pull/181 https://github.com/gsliepen/tinc/pull/182 Unfortunately, these fixes have not made it in a tinc 1.1 release yet, but when they do, I suspect you will observe...

...ute add 1.2.3.4 dev tun On May 17, 2017 3:10 PM, "Niklas Hambüchen" <mail at nh2.me> wrote: > On 17/05/17 21:50, Jared Ledvina wrote: > > Were you ever able to make any further > > progress on adjusting Tinc based on the investigation in > > https://github.com/gsliepen/tinc/issues/110 ? > > Hi Jared, > > No, not yet. > > I list a few ways for potential improvements in the ticket, but the one > that I suspect would do most on the type of virtualisation that > DigitalOcean does is to add a feature to the Linux kernel to sending the > data...

...ng the default cipher and digest settings (so, > > blowfish and sha1). I've been testing using aes-256-cbc for the cipher > > and seeing ~5% increases across the board. Each Tinc node does have > > AES-NI. > > > > I've also read through/found https://github.com/gsliepen/tinc/issues/110 > > which is very interesting. > > > > The TInc nodes are all on Centos6 AWS EC2 instances as c3.large's w/ > > EIP's. I've been testing with iperf3 and am able to get around 510Mb/s > > on the raw network. Over the tun interface/Tinc networ...

...ine, so I did not run tinc with debug messages enabled, since the issue that had caused my tinc crashes in 1.0.19 had probably been fixed. Are you aware where I could find a deb src for 1.0.26, which I could build myself, or how I could build a .deb package from the sources here https://github.com/gsliepen/tinc/tree/release-1.0.26 ? BR Max. > On Fri, Jan 22, 2016 at 12:50:39PM +0100, Maximilian H. wrote: > >> I am attempting to run tinc on a raspberry pi (version 1 B HW, 512mb >> ram). Tinc starts, but then the tinc daemon crashes. > > What do you mean by "crashe...

...cOS the memory appears to be cleared much of the time. This results in an attempt to connect to an AF_UNSPEC socket with no address, which of course fails. A potential fix, modeled on the "recently seen addresses not in our cache" part of get_recent_address() is: https://github.com/gsliepen/tinc/pull/177 Another approach would be to pass the sockaddr_t to get_recent_address() and fill it in there. I can write a diff for that if you'd like. I also noticed a problem with the recently seen but uncached code which was hidden by a cast as well as a buffer overflow (read, not write)....

...ages (as documented here https://www.tinc-vpn.org/documentation/The-meta_002dprotocol.html#The-meta_002dprotocol) seem to imply direct data packets are possible. 3. What is the routing behavior when multiple nodes advertise the same subnet at the same weight? I think this is it: https://github.com/gsliepen/tinc/blob/master/src/subnet.c#L87-L97, and it means that it's a lexical comparison between names, which will always be the same. We would prefer they be load balanced, due to... 4. Is it correct that tinc 1.0 is still single-threaded? I've seen some references to it becoming multithreaded,...

...ke sure that a solid initialization mechanism is included in precompiled .rpm and .deb packages (CentOS, Debian, Fedora, etc.) https://bugzilla.redhat.com/show_bug.cgi?id=1261717https://bugzilla.redhat.com/show_bug.cgi?id=1373704https://bugzilla.redhat.com/show_bug.cgi?id=1394512https://github.com/gsliepen/tinc/issues/133 Thanks a lot for the excellent work, Răzvan -------------- partea următoare -------------- Un ataşament HTML a fost eliminat URL: <http://www.tinc-vpn.org/pipermail/tinc/attachments/20161117/6f51c524/attachment-0001.html>

On 17/05/17 21:50, Jared Ledvina wrote: > Were you ever able to make any further > progress on adjusting Tinc based on the investigation in > https://github.com/gsliepen/tinc/issues/110 ? Hi Jared, No, not yet. I list a few ways for potential improvements in the ticket, but the one that I suspect would do most on the type of virtualisation that DigitalOcean does is to add a feature to the Linux kernel to sending the data for multiple UDP packets in one syscall,...

Hi Todd, I did some tests with NDIS 6 TAP-Windows about three years ago. The outcome of that was https://github.com/gsliepen/tinc/pull/73/commits/ 0c010ff9fe50b4046b5c7977bafac3e74037f075 It was a long time ago and I don't remember the details, but I remember I was quite confused because I couldn't explain why OpenVPN did not seem to exhibit any performance issues on the same driver, even though the code paths a...

...around 4,000 packets/sec. Right now, we are using the default cipher and digest settings (so, blowfish and sha1). I've been testing using aes-256-cbc for the cipher and seeing ~5% increases across the board. Each Tinc node does have AES-NI. I've also read through/found https://github.com/gsliepen/tinc/issues/110 which is very interesting. The TInc nodes are all on Centos6 AWS EC2 instances as c3.large's w/ EIP's. I've been testing with iperf3 and am able to get around 510Mb/s on the raw network. Over the tun interface/Tinc network, I'm only able to max it out to around 120...

While I'm aware that tinc is known to have problems with the latest Windows TAP driver (NDIS 6), we want to run tinc on Windows server 2016 which doesn't support NDIS 5 drivers. Since using the old version of the TAP driver is not an option, I'd like to better understand the issues with tinc and the NDIS 6 TAP driver. Can anyone point me to a discussion of the known issues? From

On Windows, freeaddrinfo(NULL) will result in a segv. In get_recent_address(), there is the following block of code: if(cache->aip) { sockaddr_t *sa = (sockaddr_t *)cache->aip->ai_addr; cache->aip = cache->aip->ai_next; if(!cache->aip) { freeaddrinfo(cache->aip);

A heads up: very soon the code base of tinc will be reformatted using Artistic Style. I've already pushed the .astylerc to the master and 1.1 branches, including a Makefile rule so you can run "make astyle" from the top directory and have all source files automatically formatted. The .astylerc reflects the current code style of tinc, but with the following important changes: -

...c .../hosts/... files, and mentions this in the man page. Of those, (3) seems ugly and requires a directory scan. I prefer (1), as administrators are still very likely to get it accidentally wrong, especially when setting up tinc in automated deployments. Thank you! Niklas [1] https://github.com/gsliepen/tinc/blob/5c344f297682cf11793407fca4547968aee22d95/src/net_setup.c#L341 [2] https://github.com/gsliepen/tinc/blob/1.1/src/net_setup.c#L389

You might want to try with https://github.com/gsliepen/tinc/pull/120 - that said, this bug probably doesn't explain everything because tinc is supposed to log a message from setup_vpn_in_socket() anyway, but there's no such message in your log. In addition, I really don't see any way the "Received UDP packet from unknown source" m...

...ing such a solution for a special purpose on my own tinc network right now. The simplest solution for the tunnel is to use IP/IP, which has minimal overhead and is easy to understand and troubleshoot. I contributed some code to tinc that provides better support for that use case: https://github.com/gsliepen/tinc/pull/166 > Regards > > Hans de Groot > > > configs at hosta > tinc.conf > Name = hosta > Device =/dev/net/tun > Hostnames = No > connectto = hostb > Mode = Router > KeyExpire = 3600 > PingInterval = 10 > PingTimeout = 15 > PrivateKeyFile = /etc...

...from standby on Windows, despite the latest fixes that I wrote (which are included in 1.1pre15). I think my fixes covered some of the causes of tinc misbehaving on standby, but not all of them. Are you able to reproduce the issue reliably? If so, can you please file an issue at https://github.com/gsliepen/tinc/issues (be sure to mention @dechamps) along with a verbose tinc log (-d5)? It would also help if you included the thread stacks that Process Explorer is showing you for tincd.exe while it's stuck. Oh, by the way, make sure you're using the TAP-Win32 driver that's bundled with tinc...