NgTech-LTD/gns3-server
2
0
Fork 0
mirror of https://github.com/GNS3/gns3-server.git synced 2026-08-27 12:30:13 +03:00
Code Issues Packages Projects Releases Wiki Activity

vendor: skip /etc/network in volume bridge and permission passes

Browse Source 
GNS3_SKIP_INIT containers never run init.sh, so /etc/network (GNS3's own
network config consumed by init.sh's ifup) has no consumer — the NOS
manages its own interfaces. VendorDockerVM._persistent_volumes() filters it
out for both _setup_skip_init_volumes and _fix_permissions, saving one
docker exec per pass. The shared _mount_binds is untouched, and without
GNS3_SKIP_INIT the full volume list is returned so behaviour matches the
base class.
This commit is contained in:
YueGuobin 2026-08-12 23:32:08 +08:00
parent 486c5cc1cb
commit 46e863b975
No known key found for this signature in database
2 changed files with 22 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
docs/features/docker-exec-console.md
View File

@ -241,6 +241,13 @@ host ──Docker bind mount──▶ /gns3volumes/etc/opt/srlinux (always m
> host-side root-owned, and an unprivileged GNS3 process cannot chown them
> from the host. Container-side root (with GNS3's `UsernsMode: host`) can.
With `GNS3_SKIP_INIT`, `/etc/network` is skipped by both vendor passes
(`VendorDockerVM._persistent_volumes()`): it holds GNS3's own network config
for init.sh's `ifup`, which never runs for SKIP_INIT containers — the NOS
manages its own interfaces. The Docker mount itself is left alone (shared
`_mount_binds`, hardcoded at `docker_vm.py` `_mount_binds()`); only the
vendor-side bridge/fix passes skip it.
### Lifecycle summary
| Phase | Normal Docker node | `VendorDockerVM` + `GNS3_SKIP_INIT` |

17
gns3server/compute/docker/vendor_docker_vm.py
View File

@ -95,6 +95,19 @@ class VendorDockerVM(DockerVM):
last_ifname = f"eth{self.adapters - 1}"
params["Env"].append(f"GNS3_MAX_ETHERNET={last_ifname}")
def _persistent_volumes(self):
"""
Volumes relevant to vendor persistence. With GNS3_SKIP_INIT, drop
/etc/network — GNS3's own network config for init.sh's ifup, unused
when init.sh is skipped (the NOS manages its own interfaces). The
Docker mount itself is left alone (shared _mount_binds); only the
vendor-side bridge/fix passes skip it. Without SKIP_INIT the full
list is returned so behaviour matches the base class.
"""
if self._gns3_init:
return self._volumes
return [v for v in self._volumes if v != "/etc/network"]
def _get_container_ifname(self, adapter_number):
"""
Override: honour GNS3_INTERFACE_NAMES (e.g. mgmt0, e1-1) in adapter
@ -157,7 +170,7 @@ class VendorDockerVM(DockerVM):
return
uid, gid = os.getuid(), os.getgid()
for volume in self._volumes:
for volume in self._persistent_volumes():
target = f"/gns3volumes{volume}"
log.debug("Docker container '%s' fix ownership on %s", self._name, target)
try:
@ -199,7 +212,7 @@ class VendorDockerVM(DockerVM):
Permission-changes recorded by _fix_permissions at the previous
stop are restored (best-effort).
"""
for volume in self._volumes:
for volume in self._persistent_volumes():
vol_target = f"/gns3volumes{volume}"
# fmt: off
script = (