summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorSimon Rettberg2024-09-06 12:50:55 +0200
committerSimon Rettberg2024-09-06 12:50:55 +0200
commita6f58c215d5ef4ec65f1da0eef0eddaf74662175 (patch)
tree1803aee22560f5e48630dc0ecdb2bdf9b075734c
parent[dhcpc-busybox] Wait for udev to rename eth* devices (diff)
downloadmltk-a6f58c215d5ef4ec65f1da0eef0eddaf74662175.tar.gz
mltk-a6f58c215d5ef4ec65f1da0eef0eddaf74662175.tar.xz
mltk-a6f58c215d5ef4ec65f1da0eef0eddaf74662175.zip
[dhcpc-busybox] bridging: Add another sleep for good measure
-rwxr-xr-xcore/modules/dhcpc-busybox/data/opt/openslx/scripts/systemd-bridge_additional_nics3
1 files changed, 3 insertions, 0 deletions
diff --git a/core/modules/dhcpc-busybox/data/opt/openslx/scripts/systemd-bridge_additional_nics b/core/modules/dhcpc-busybox/data/opt/openslx/scripts/systemd-bridge_additional_nics
index 605e2e9d..f33a8445 100755
--- a/core/modules/dhcpc-busybox/data/opt/openslx/scripts/systemd-bridge_additional_nics
+++ b/core/modules/dhcpc-busybox/data/opt/openslx/scripts/systemd-bridge_additional_nics
@@ -28,6 +28,9 @@ is_slave() {
# end up either taking a device up as e.g. eth0 and prevent systemd
# from renaming it, or end up missing a device as it will get renamed
# just after the glob expanded.
+# We already wait for udev trigger and settle, but that doesn't seem
+# enough...
+sleep .5
for nic in /sys/class/net/*; do
ifname="${nic##*/}"
# do not handle the primary interface