Differences

This shows you the differences between two versions of the page.

--- 0x13:reports:d1t3t03-frrouting-workshop [2019/03/27 23:29] – ehalep
+++ 0x13:reports:d1t3t03-frrouting-workshop [2019/09/28 17:04] (current) – external edit 127.0.0.1
@@ Line 4: / Line 4: @@
 Report by: Donald Sharp
+The first presentation was from Donald Sharp and David Lamparter - cumulus
+New stuff introduced was nexthop groups (David Ahern was doing this code) and learned later on in zebra session how this is just the start.
-FRR workshop
+A BGP memory rework was discussed where the memory footprint was reduced from 500 MByte to 450 MByte for full internet BGP feed nice. In regards to VRF vs. name-spaces, it would be preferred to get rid of complexity of using name-spaces as VRFs.
-Presentation 1 from Donald Sharp, David Lamparter - cumulus
-New stuff:
-  nexthop groups (David Ahern was doing this code)
-  learned later on in zebra session how this is just the start.
-BGP memory rework:
-MByte -> 450 MByte for full internet BGP feed
-  nice.
-VRF vs. name-spaces: Woud like to get rid of complexity of
-  using name-spaces as VRFs.
-  commented:
-   name-spaces likely only used earlier on instead of VRF,
-   because VRF was only implemented much later in Linux.
-   Logically now to use VRF as VRFs and accordingly treat in Linux.
-   Can always run different instanes of FRR in different name-spaces.
-Presentation 2: session about data-plane / zebra afterwards:
-voltanet.io presenter, not the coder himself though.
-- ZAPI - API between zebra and routing daemons.
-  Now one thread per client in ZEBRA.
-- Aynchronuous communication with FIB. E.g.: remote switch (voltanet).
-  two threads now: zebra-main, zebra-dplane connected by two queues.
-  zbra dplane talks with "kernel".
-- ACK path (installed in HW): only used by sharpd (installs 1 million
-  routes,
-  testing daemon), but needed for BGP, RIP according to donald.
-  cumulus linux: currently no feedback from linux-kernel when route is
-  installed onto ASIC (eg: broadcom).
-- kiran: could expand dplane daemon to directly use chip API to install
-  routes, not relying on linux-kernel driver doing it. A: Yes.
-- toerless: change nexthop-group-members ? Donald: No, not there yet.
-Site: https://www.netdevconf.org/0x13/session.html?workshop-FRRouting
-Slides:
-Videos:
+A comment received was that name-spaces likely only used earlier on instead of VRF, because VRF wa only implemented much later in Linux. Logically now to use VRF as VRFs and accordingly treat in Linux. Can always run different instances of FRR in different name-spaces.
+Site: https://www.netdevconf.info/0x13/session.html?workshop-FRRouting