Open comicfans opened 3 months ago
Your changes look reasonable to me (other than the comment I made there). Can you please open a PR to run our test suite against it? If everything looks good, I'm happy to approve it.
(Nevermind the bit about the comment.)
Your changes look reasonable to me (other than the comment I made there). Can you please open a PR to run our test suite against it? If everything looks good, I'm happy to approve it.
Description of the bug:
when using linux-sandbox, it didn't remount /sys, leave /sys/class/net still being host network devices, not ones inside the network namespace. remount /sys (similar to /proc) resolve it. I have a branch https://github.com/comicfans/bazel/commit/891b41a2ef4106ae44caead07a95632925e67468 to address it, shall I create PR for this?
Which category does this issue belong to?
Local Execution
What's the simplest, easiest way to reproduce this bug? Please provide a minimal example if possible.
run command
ip link show will output correct devices in network namespace, but ls command will show host network devices
Which operating system are you running Bazel on?
linux
What is the output of
bazel info release
?I'm testing on master dacb794
If
bazel info release
returnsdevelopment version
or(@non-git)
, tell us how you built Bazel.add test (without requires-network tag, then linux-sandbox will create network namespace for it) and ls /sys/class/net in test
What's the output of
git remote get-url origin; git rev-parse HEAD
?If this is a regression, please try to identify the Bazel commit where the bug was introduced with bazelisk --bisect.
No response
Have you found anything relevant by searching the web?
https://stackoverflow.com/questions/15626088/tap-interfaces-and-dev-net-tun-device-using-ip-tuntap-command
Any other information, logs, or outputs that you want to share?
No response