Dear BIRD Users, we're presenting you a new alpha version of the Multithreaded BIRD. Contrary to the previous version, you have to set your desired number of threads in config by the threads N; option. Version 3.0alpha1 fixes a lot of bugs found in previous versions. Our tests show that the overall performance may be about 3-8 times better than v2.0.12; yet the real life always brings questions and problems which the development can't anticipate fully. Please run your tests and we hope we get back to you soon with a stable version. Looking forward to your reports. I thank my colleagues and namely Maria for this release! Cheers Ondrej
Hello, On Fri 21 Apr 2023 09:20:29 GMT, Ondrej Filip wrote:
Dear BIRD Users, we're presenting you a new alpha version of the Multithreaded BIRD. Contrary to the previous version, you have to set your desired number of threads in config by the threads N; option.
Version 3.0alpha1 fixes a lot of bugs found in previous versions. Our tests show that the overall performance may be about 3-8 times better than v2.0.12; yet the real life always brings questions and problems which the development can't anticipate fully.
Please run your tests and we hope we get back to you soon with a stable version. Looking forward to your reports.
I thank my colleagues and namely Maria for this release!
Cheers Ondrej
Thanks a lot! If I try to push a fullview in FIB, it seems that the communication with netlink has issues: Apr 21 11:26:13 edge08-clouvider bird: Netlink: File exists Apr 21 11:26:13 edge08-clouvider bird: Netlink: File exists Apr 21 11:26:13 edge08-clouvider bird: Netlink: File exists Apr 21 11:26:13 edge08-clouvider bird: Netlink: File exists Apr 21 11:26:13 edge08-clouvider bird: Netlink: File exists Apr 21 11:26:13 edge08-clouvider bird: ... Apr 21 11:26:13 edge08-clouvider bird: I/O loop cycle took 6655.745 ms for 4 events Apr 21 11:26:13 edge08-clouvider bird: Kernel dropped some netlink messages, will resync on next scan. Apr 21 11:27:12 edge08-clouvider bird: Netlink: File exists Apr 21 11:27:12 edge08-clouvider bird: Netlink: File exists Apr 21 11:27:12 edge08-clouvider bird: Netlink: File exists Apr 21 11:27:12 edge08-clouvider bird: Netlink: File exists Apr 21 11:27:12 edge08-clouvider bird: Netlink: File exists Apr 21 11:27:13 edge08-clouvider bird: I/O loop cycle took 6705.487 ms for 4 events Apr 21 11:27:13 edge08-clouvider bird: Kernel dropped some netlink messages, will resync on next scan. Apr 21 11:27:36 edge08-clouvider bird: Kernel dropped some netlink messages, will resync on next scan. Apr 21 11:27:40 edge08-clouvider bird: Kernel dropped some netlink messages, will resync on next scan. Apr 21 11:27:55 edge08-clouvider bird: Kernel dropped some netlink messages, will resync on next scan. Apr 21 11:28:13 edge08-clouvider bird: Netlink: File exists Apr 21 11:28:13 edge08-clouvider bird: Netlink: File exists Apr 21 11:28:13 edge08-clouvider bird: Netlink: File exists Apr 21 11:28:13 edge08-clouvider bird: Netlink: File exists Apr 21 11:28:13 edge08-clouvider bird: Netlink: File exists Apr 21 11:28:13 edge08-clouvider bird: ... Apr 21 11:28:13 edge08-clouvider bird: I/O loop cycle took 6727.839 ms for 4 events Apr 21 11:28:13 edge08-clouvider bird: Kernel dropped some netlink messages, will resync on next scan. Apr 21 11:29:12 edge08-clouvider bird: Netlink: File exists Apr 21 11:29:12 edge08-clouvider bird: Netlink: File exists Apr 21 11:29:12 edge08-clouvider bird: Netlink: File exists Apr 21 11:29:12 edge08-clouvider bird: Netlink: File exists Apr 21 11:29:12 edge08-clouvider bird: Netlink: File exists Apr 21 11:29:12 edge08-clouvider bird: ... Apr 21 11:29:13 edge08-clouvider bird: I/O loop cycle took 6705.334 ms for 4 events Apr 21 11:29:13 edge08-clouvider bird: Kernel dropped some netlink messages, will resync on next scan. Regards, -- Alarig
Hello Alarig! On 4/21/23 11:32, Alarig Le Lay wrote:
Hello,
On Fri 21 Apr 2023 09:20:29 GMT, Ondrej Filip wrote:
Dear BIRD Users, we're presenting you a new alpha version of the Multithreaded BIRD. Contrary to the previous version, you have to set your desired number of threads in config by the threads N; option.
Version 3.0alpha1 fixes a lot of bugs found in previous versions. Our tests show that the overall performance may be about 3-8 times better than v2.0.12; yet the real life always brings questions and problems which the development can't anticipate fully.
Please run your tests and we hope we get back to you soon with a stable version. Looking forward to your reports.
I thank my colleagues and namely Maria for this release!
If I try to push a fullview in FIB, it seems that the communication with netlink has issues: Apr 21 11:26:13 edge08-clouvider bird: Netlink: File exists Apr 21 11:26:13 edge08-clouvider bird: Netlink: File exists Apr 21 11:26:13 edge08-clouvider bird: Netlink: File exists Apr 21 11:26:13 edge08-clouvider bird: Netlink: File exists Apr 21 11:26:13 edge08-clouvider bird: Netlink: File exists Apr 21 11:26:13 edge08-clouvider bird: ... Apr 21 11:26:13 edge08-clouvider bird: I/O loop cycle took 6655.745 ms for 4 events Apr 21 11:26:13 edge08-clouvider bird: Kernel dropped some netlink messages, will resync on next scan.
There should be no significant difference between v2 and v3 in these parts of BIRD. Does it happen only on 3.0-alpha1 or also on 2.0.12? Could you also please share with me your `show route all` output and configuration? (Possibly off-list.) I'll try to reproduce the issue here. Thank you in advance. Maria
I was looking for a live stream of @Maria Matejka <maria.matejka@nic.cz> presentation about Bird 3 on 38º Euro-IX. But I was not able to find anyone. This event was transmitted? https://www.euro-ix.net/en/events/fora/38th-euro-ix-forum/#:~:text=BIRD%20in... Em sex., 21 de abr. de 2023 às 04:24, Ondrej Filip <feela@network.cz> escreveu:
Dear BIRD Users, we're presenting you a new alpha version of the Multithreaded BIRD. Contrary to the previous version, you have to set your desired number of threads in config by the threads N; option.
Version 3.0alpha1 fixes a lot of bugs found in previous versions. Our tests show that the overall performance may be about 3-8 times better than v2.0.12; yet the real life always brings questions and problems which the development can't anticipate fully.
Please run your tests and we hope we get back to you soon with a stable version. Looking forward to your reports.
I thank my colleagues and namely Maria for this release!
Cheers Ondrej
-- Douglas Fernando Fischer Engº de Controle e Automação
Hello! This talk is gonna happen tomorrow, therefore there is not any recording yet, if I haven't confused the dates. Maria On 4/24/23 15:40, Douglas Fischer wrote:
I was looking for a live stream of @Maria Matejka <mailto:maria.matejka@nic.cz> presentation about Bird 3 on 38º Euro-IX. But I was not able to find anyone. This event was transmitted?
https://www.euro-ix.net/en/events/fora/38th-euro-ix-forum/#:~:text=BIRD%20in... <https://www.euro-ix.net/en/events/fora/38th-euro-ix-forum/#:~:text=BIRD%20in%20muiltiple%20threads>
Em sex., 21 de abr. de 2023 às 04:24, Ondrej Filip <feela@network.cz <mailto:feela@network.cz>> escreveu:
Dear BIRD Users, we're presenting you a new alpha version of the Multithreaded BIRD. Contrary to the previous version, you have to set your desired number of threads in config by the threads N; option.
Version 3.0alpha1 fixes a lot of bugs found in previous versions. Our tests show that the overall performance may be about 3-8 times better than v2.0.12; yet the real life always brings questions and problems which the development can't anticipate fully.
Please run your tests and we hope we get back to you soon with a stable version. Looking forward to your reports.
I thank my colleagues and namely Maria for this release!
Cheers Ondrej
-- Douglas Fernando Fischer Engº de Controle e Automação
Hi Douglas, Come to CSNOG 2023. Maria should give a presentation on Bird 3 there too! :) On Mon, Apr 24, 2023 at 3:48 PM Douglas Fischer <fischerdouglas@gmail.com> wrote:
I was looking for a live stream of @Maria Matejka <maria.matejka@nic.cz> presentation about Bird 3 on 38º Euro-IX. But I was not able to find anyone. This event was transmitted?
https://www.euro-ix.net/en/events/fora/38th-euro-ix-forum/#:~:text=BIRD%20in...
Em sex., 21 de abr. de 2023 às 04:24, Ondrej Filip <feela@network.cz> escreveu:
Dear BIRD Users, we're presenting you a new alpha version of the Multithreaded BIRD. Contrary to the previous version, you have to set your desired number of threads in config by the threads N; option.
Version 3.0alpha1 fixes a lot of bugs found in previous versions. Our tests show that the overall performance may be about 3-8 times better than v2.0.12; yet the real life always brings questions and problems which the development can't anticipate fully.
Please run your tests and we hope we get back to you soon with a stable version. Looking forward to your reports.
I thank my colleagues and namely Maria for this release!
Cheers Ondrej
-- Douglas Fernando Fischer Engº de Controle e Automação
W dniu 21.04.2023 o 09:20, Ondrej Filip pisze:
Dear BIRD Users, we're presenting you a new alpha version of the Multithreaded BIRD. Contrary to the previous version, you have to set your desired number of threads in config by the threads N; option.
Version 3.0alpha1 fixes a lot of bugs found in previous versions. Our tests show that the overall performance may be about 3-8 times better than v2.0.12; yet the real life always brings questions and problems which the development can't anticipate fully.
Please run your tests and we hope we get back to you soon with a stable version. Looking forward to your reports.
I thank my colleagues and namely Maria for this release!
Cheers Ondrej
Dear BIRD Developers, I have tested the Netlink flavor of FreeBSD port. At a glance there are at least 2 problems: 1. Memory (resident) usage at the init is about 125% of BIRD 2.13, moreover it's constantly growing and after 30 min it was about 300% of shown by BIRD 2.13. 2. Some routes are not learned correctly, for example: Apr 27 20:20:23 rtr bird[40469]: bgp_rmsk_v4: Invalid route x.y.140.0/22 withdrawn Apr 27 20:20:23 rtr bird[40469]: bgp_rmsk_v4: Invalid route x.y.160.0/22 withdrawn Apr 27 20:20:23 rtr bird[40469]: bgp_rmsk_v4: Invalid route x.y.184.0/22 withdrawn Apr 27 20:20:23 rtr bird[40469]: bgp_rmsk_v4: Invalid route y.y.184.0/24 withdrawn Apr 27 20:20:23 rtr bird[40469]: bgp_rmsk_v4: Malformed bgp_unknown_0x15 attribute - conflicting flags (e0, expected 80) Apr 27 20:20:23 rtr bird[40469]: bgp_rmsk_v4: Malformed bgp_unknown_0x15 attribute - conflicting flags (e0, expected 80) Apr 27 20:20:23 rtr bird[40469]: bgp_rmsk_v4: Malformed bgp_unknown_0xff attribute - conflicting flags (e0, expected 80) Apr 27 20:20:23 rtr bird[40469]: bgp_rmsk_v4: Malformed bgp_unknown_0xff attribute - conflicting flags (e0, expected 80) It came out, that in BIRD 2.13 at least some of these problematic routes have special attributes, for example: BGP.ff [t]: 00 00 07 db 00 00 00 01 00 01 00 0a ff 08 00 00 00 00 ea 44 42 79 Thanks for giving the opportunity to test ! Cheers -- Marek Zarychta
Hello Marek, this looks fishy. We haven't yet had an opportunity to properly test the Netlink at FreeBSD so thank you for reporting, this is indeed a bug. As soon as we get to spin up the newest FreeBSD, we'll look into it and try to debug. Thank you for your report! Maria
Dear BIRD Developers,
I have tested the Netlink flavor of FreeBSD port. At a glance there are at least 2 problems:
1. Memory (resident) usage at the init is about 125% of BIRD 2.13, moreover it's constantly growing and after 30 min it was about 300% of shown by BIRD 2.13.
2. Some routes are not learned correctly, for example:
Apr 27 20:20:23 rtr bird[40469]: bgp_rmsk_v4: Invalid route x.y.140.0/22 withdrawn Apr 27 20:20:23 rtr bird[40469]: bgp_rmsk_v4: Invalid route x.y.160.0/22 withdrawn Apr 27 20:20:23 rtr bird[40469]: bgp_rmsk_v4: Invalid route x.y.184.0/22 withdrawn Apr 27 20:20:23 rtr bird[40469]: bgp_rmsk_v4: Invalid route y.y.184.0/24 withdrawn Apr 27 20:20:23 rtr bird[40469]: bgp_rmsk_v4: Malformed bgp_unknown_0x15 attribute - conflicting flags (e0, expected 80) Apr 27 20:20:23 rtr bird[40469]: bgp_rmsk_v4: Malformed bgp_unknown_0x15 attribute - conflicting flags (e0, expected 80) Apr 27 20:20:23 rtr bird[40469]: bgp_rmsk_v4: Malformed bgp_unknown_0xff attribute - conflicting flags (e0, expected 80) Apr 27 20:20:23 rtr bird[40469]: bgp_rmsk_v4: Malformed bgp_unknown_0xff attribute - conflicting flags (e0, expected 80)
It came out, that in BIRD 2.13 at least some of these problematic routes have special attributes, for example:
BGP.ff [t]: 00 00 07 db 00 00 00 01 00 01 00 0a ff 08 00 00 00 00 ea 44 42 79
Thanks for giving the opportunity to test !
Cheers
participants (6)
-
Alarig Le Lay -
Alexander Zubkov -
Douglas Fischer -
Marek Zarychta -
Maria Matejka -
Ondrej Filip