software-engineering/spark-vllm-docker
2
0
Fork 0
Code Issues Actions Packages Projects Releases Wiki Activity

Don't start the cluster if only --setup/--discover is specified

Browse Source
This commit is contained in:
Eugene Rakhmatulin
2026-03-31 13:29:56 -07:00
parent bb177383ff
commit 9370b2bb34
2 changed files with 18 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
docs/NETWORKING.md
View File

@@ -106,8 +106,9 @@ network:
enP2p1s0f1np1:
dhcp4: no
dhcp6: no
link-local: [ ipv4 ]
link-local: []
mtu: 9000
addresses: [192.168.178.11/24]
```
Create `/etc/netplan/40-cx7.yaml` on `spark2`:
@@ -124,16 +125,12 @@ network:
enP2p1s0f1np1:
dhcp4: no
dhcp6: no
link-local: [ ipv4 ]
link-local: []
mtu: 9000
addresses: [192.168.178.12/24]
```
Please note, that only one interface of the "twin" pair needs an IP address, but MTU needs to be set on both.
You can also assign a separate address to another "twin" if you want to utilize the second interface independently, but make sure you assign an IP address from a different subnet.
For instance, for the example above, if you want to assign an IP to `enP2p1s0f1np1`, you need to use `192.168.177.12` on `spark`. **DO NOT use the same subnet on both "twins"** - it will confuse autodiscovery and mess up routing.
This will not affect vLLM performance as it will use RDMA over RoCE using both "twins", even if the IP is only set on one.
**DO NOT use the same subnet on both "twins"** - it will confuse autodiscovery and mess up routing.
Then run on each node: