Skip to content

Commit b783d57

Browse files
committed
Update up to 'Pulling manually'
1 parent fca5055 commit b783d57

File tree

1 file changed

+15
-52
lines changed

1 file changed

+15
-52
lines changed

docs/software/container-engine.md

Lines changed: 15 additions & 52 deletions
Original file line numberDiff line numberDiff line change
@@ -181,69 +181,32 @@ To choose an alternative image store path (e.g., to use a directory owned by a g
181181

182182
To work with images stored from the NGC Catalog, please refer also to the next section "Using images from third party registries and private repositories".
183183

184-
To bypass any caching behavior, users can manually pull an image and directly plug it into their EDF. To do so, users may execute `enroot import docker://[REGISTRY#]IMAGE[:TAG]` to pull container images from OCI registries to the current directory.
184+
To bypass any caching behavior, users can manually pull an image and directly plug it into their EDF.
185+
To do so, users may execute `enroot import docker://[REGISTRY#]IMAGE[:TAG]` to pull container images from OCI registries to the current directory.
185186

186-
For example, the command below pulls an `nvidia/cuda:11.8.0-cudnn8-devel-ubuntu22.04` image.
187+
After the import is complete, images are available in Squashfs format in the current directory and can be used in EDFs:
187188

188-
```bash
189-
$ enroot import docker://nvidia/cuda:11.8.0-cudnn8-devel-ubuntu22.04
190-
```
191-
192-
??? example "Image import w/ full output"
193-
```bash
194-
> srun enroot import docker://nvidia/cuda:11.8.0-cudnn8-devel-ubuntu22.04
189+
!!! example "Manually pulling an `nvidia/cuda:11.8.0-cudnn8-devel-ubuntu22.04` image"
190+
```console
191+
$ cd ${SCRATCH}
192+
$ enroot import docker://nvidia/cuda:11.8.0-cudnn8-devel-ubuntu22.04
195193
[INFO] Querying registry for permission grant
196194
[INFO] Authenticating with user: <anonymous>
197-
[INFO] Authentication succeeded
198-
[INFO] Fetching image manifest list
199-
[INFO] Fetching image manifest
200-
[INFO] Downloading 13 missing layers...
201-
[INFO] Extracting image layers...
202-
[INFO] Converting whiteouts...
203-
[INFO] Creating squashfs filesystem...
204-
Parallel mksquashfs: Using 64 processors
205-
Creating 4.0 filesystem on /scratch/aistor/<username>/nvidia+cuda+11.8.0-cudnn8-devel-ubuntu22.04.sqsh, block size 131072.
206-
207-
Exportable Squashfs 4.0 filesystem, zstd compressed, data block size 131072
208-
uncompressed data, compressed metadata, compressed fragments,
209-
compressed xattrs, compressed ids
210-
duplicates are removed
211-
Filesystem size 9492185.87 Kbytes (9269.71 Mbytes)
212-
98.93% of uncompressed filesystem size (9594893.12 Kbytes)
213-
Inode table size 128688 bytes (125.67 Kbytes)
214-
17.47% of uncompressed inode table size (736832 bytes)
215-
Directory table size 132328 bytes (129.23 Kbytes)
216-
46.42% of uncompressed directory table size (285091 bytes)
217-
Number of duplicate files found 1069
218-
Number of inodes 13010
219-
Number of files 10610
220-
Number of fragments 896
221-
Number of symbolic links 846
222-
Number of device nodes 0
223-
Number of fifo nodes 0
224-
Number of socket nodes 0
225-
Number of directories 1554
226-
Number of ids (unique uids + gids) 1
227-
Number of uids 1
228-
root (0)
195+
...
229196
Number of gids 1
230197
root (0)
231-
```
232-
233-
After the import is complete, images are available in Squashfs format in the current directory and can be used in EDFs, for example:
234198

235-
```bash
236-
> ls -l *.sqsh
237-
-rw-r--r-- 1 <username> csstaff 9720037376 Sep 11 14:46 nvidia+cuda+11.8.0-cudnn8-devel-ubuntu22.04.sqsh
199+
$ ls *.sqsh
200+
nvidia+cuda+11.8.0-cudnn8-devel-ubuntu22.04.sqsh
238201

239-
> realpath nvidia+cuda+11.8.0-cudnn8-devel-ubuntu22.04.sqsh  /capstor/scratch/cscs/<username>/nvidia+cuda+11.8.0-cudnn8-devel-ubuntu22.04.sqsh
202+
$ cat ${HOME}/.edf/example.toml (1)
203+
image = "/capstor/scratch/cscs/${USER}/nvidia+cuda+11.8.0-cudnn8-devel-ubuntu22.04.sqsh"
204+
```
240205

241-
> cat $HOME/.edf/cudnn8.toml
242-
image = "/capstor/scratch/cscs/<username>/nvidia+cuda+11.8.0-cudnn8-devel-ubuntu22.04.sqsh"
243-
```
206+
1. Assuming `example.toml` was already written at `${HOME}/.edf`.
244207

245208
!!! note
246-
It is recommended to save images in `/capstor/scratch/cscs/<username>` or its subdirectories before using them with the CE.
209+
It is recommended to save images in `/capstor/scratch/cscs/${USER}` or its subdirectories before using them with the CE.
247210

248211
[](){#ref-ce-third-party-private-registries}
249212
### Third-party and private registries

0 commit comments

Comments
 (0)