Replies: 7 comments 1 reply
-
I created a loxilb in hostmode like below docker run -u root --cap-add SYS_ADMIN --restart unless-stopped --privileged -dit -v /var/run/docker.sock:/var/run/docker.sock -v /dev/log:/dev/log --name loxilb-host-mode --network host 1a8df5eae3fa98df3e3a45e4c1417c3cbd36ad2d0ad711d8137f0930d50f187f even though I have created a lb with 2020 port and in host mode, I dont see 2020 port in listening mode .. cps@pn-distributor-3: |
Beta Was this translation helpful? Give feedback.
-
hi any clues or pointers ? |
Beta Was this translation helpful? Give feedback.
-
hi team , any pointers ? |
Beta Was this translation helpful? Give feedback.
-
Hi @megarajan, Loxilb does not listen on 2020 port because things are implemented with eBPF. Not sure why your multi-homing setup is not working. Our sctp-mh cicd works fine. I will take a look and get back. |
Beta Was this translation helpful? Give feedback.
-
thanks for the reponse, have sent a request to join the community you pointed . waiting for your detailed response.. |
Beta Was this translation helpful? Give feedback.
-
to give more background on our test setup, we have 3 linux ubuntu vms client vm ( running sctp traffic sending app) -> loxilb vm -> endpoint vm ( running app to process to sctp traffic ) |
Beta Was this translation helpful? Give feedback.
-
Can you check if ipv6 is enabled. If so please disable
…On Wed, Jan 8, 2025, 2:27 PM megarajan ***@***.***> wrote:
to give more background on our test setup, we have 3 linux ubuntu vms
client vm ( running sctp traffic sending app) -> loxilb vm -> endpoint vm
( running app to process to sctp traffic )
—
Reply to this email directly, view it on GitHub
<#922 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/BMXPDCB4ZBAYZWNBRVPK6V32JSZNZAVCNFSM6AAAAABUVQ2GM6VHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTCNZWHA4DKNA>
.
You are receiving this because you are subscribed to this thread.Message
ID: ***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
I have created a sctp multi homing loxilb instance in fullnat mode with this config.
Ip : 10.77.87.205
sec ips : 10.77.87.201, 10.77.87.202,10.77.87.203
endpoint ip : 10.77.87.181
endpoint port : 3869
root@pn-distributor-3:/# loxicmd get lb -o wide
when we send messages to the 10.77.87.205 ip , the connections are getting aborted. a sample tcpdump is below.
12:30:20.468526 IP 10.77.87.87.33682 > 10.77.87.205.2020: sctp (1) [INIT] [init tag: 196020939] [rwnd: 8388608] [OS: 10] [MIS: 65535] [init TSN: 840558473]
12:30:20.468928 IP 10.77.87.205.2020 > 10.77.87.87.33682: sctp (1) [ABORT]
12:30:25.467253 IP 10.77.87.87.33682 > 10.77.87.205.2020: sctp (1) [INIT] [init tag: 3870108015] [rwnd: 8388608] [OS: 10] [MIS: 65535] [init TSN: 473674365]
12:30:25.467836 IP 10.77.87.205.2020 > 10.77.87.87.33682: sctp (1) [ABORT]
12:30:30.465660 IP 10.77.87.87.33682 > 10.77.87.205.2020: sctp (1) [INIT] [init tag: 1360199288] [rwnd: 8388608] [OS: 10] [MIS: 65535] [init TSN: 2930406781]
12:30:30.466264 IP 10.77.87.205.2020 > 10.77.87.87.33682: sctp (1) [ABORT]
Could you please share any pointers on why this could be happening ?
I see that the counters are getting incremented in the get lb command output.
thanks
Beta Was this translation helpful? Give feedback.
All reactions