Skip to content

Commit 1383b9c

Browse files
authored
[Docs]: Mention SSH fleet networking requirements (#2643)
1 parent 4b15d69 commit 1383b9c

File tree

1 file changed

+14
-10
lines changed

1 file changed

+14
-10
lines changed

docs/docs/concepts/fleets.md

Lines changed: 14 additions & 10 deletions
Original file line numberDiff line numberDiff line change
@@ -263,26 +263,30 @@ Define a fleet configuration as a YAML file in your project directory. The file
263263
</div>
264264

265265
??? info "Requirements"
266-
1.&nbsp;Hosts should be pre-installed with Docker.
266+
1.&nbsp;Hosts must be pre-installed with Docker.
267267

268268
=== "NVIDIA"
269-
2.&nbsp;Hosts with NVIDIA GPUs should also be pre-installed with CUDA 12.1 and
269+
2.&nbsp;Hosts with NVIDIA GPUs must also be pre-installed with CUDA 12.1 and
270270
[NVIDIA Container Toolkit :material-arrow-top-right-thin:{ .external }](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html).
271271

272272
=== "AMD"
273-
2.&nbsp;Hosts with AMD GPUs should also be pre-installed with AMDGPU-DKMS kernel driver (e.g. via
273+
2.&nbsp;Hosts with AMD GPUs must also be pre-installed with AMDGPU-DKMS kernel driver (e.g. via
274274
[native package manager :material-arrow-top-right-thin:{ .external }](https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/native-install/index.html)
275275
or [AMDGPU installer :material-arrow-top-right-thin:{ .external }](https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/amdgpu-install.html).)
276276

277277
=== "Intel Gaudi"
278-
2.&nbsp;Hosts with Intel Gaudi accelerators should be pre-installed with [Gaudi software and drivers](https://docs.habana.ai/en/latest/Installation_Guide/Driver_Installation.html#driver-installation).
279-
This should include the drivers, `hl-smi`, and Habana Container Runtime.
278+
2.&nbsp;Hosts with Intel Gaudi accelerators must be pre-installed with [Gaudi software and drivers](https://docs.habana.ai/en/latest/Installation_Guide/Driver_Installation.html#driver-installation).
279+
This must include the drivers, `hl-smi`, and Habana Container Runtime.
280280

281281
=== "Tenstorrent"
282-
2.&nbsp;Hosts with Tenstorrent accelerators should be pre-installed with [Tenstorrent software](https://docs.tenstorrent.com/getting-started/README.html#software-installation).
283-
This should include the drivers, `tt-smi`, and HugePages.
282+
2.&nbsp;Hosts with Tenstorrent accelerators must be pre-installed with [Tenstorrent software](https://docs.tenstorrent.com/getting-started/README.html#software-installation).
283+
This must include the drivers, `tt-smi`, and HugePages.
284284

285-
3.&nbsp;The user specified should have passwordless `sudo` access.
285+
3.&nbsp;The user specified must have passwordless `sudo` access.
286+
287+
4.&nbsp;The SSH server must be running and configured with `AllowTcpForwarding yes` in `/etc/ssh/sshd_config`.
288+
289+
5.&nbsp;The firewall must allow SSH and should forbid any other connections from external networks. For `placement: cluster` fleets, it should also allow any communication between fleet nodes.
286290

287291
To create or update the fleet, pass the fleet configuration to [`dstack apply`](../reference/cli/dstack/apply.md):
288292

@@ -341,10 +345,10 @@ divided into, allowing multiple jobs to use these blocks concurrently.
341345
hosts:
342346
- hostname: 3.255.177.51
343347
blocks: 4
344-
- hostaname: 3.255.177.52
348+
- hostname: 3.255.177.52
345349
# As many as possible, according to numbers of GPUs and CPUs
346350
blocks: auto
347-
- hostaname: 3.255.177.53
351+
- hostname: 3.255.177.53
348352
# Do not sclice. This is the default value, may be omitted
349353
blocks: 1
350354
```

0 commit comments

Comments
 (0)