YouJiacheng

Routing tips for VPNs on OS X

When VPNs Just Work™, they're a fantastic way of allowing access to a private network from remote locations. When they don't work it can be an experience in frustration. I've had situations where I can connect to a VPN from my Mac, but various networking situations cause routing conflicts. Here are a couple of cases and how I've been able to get around them.

Specific cases

Case 1: conflicting additional routes.

In this example the VPN we are connecting to has a subnet that does not conflict with our local IP, but has additional routes that conflict in some way with our local network's routing. In my example the remote subnet is 10.0.x.0/24, my local subnet is 10.0.y.0/24, and the conflicting route is 10.0.0.0/8. Without the later route, I can't access all hosts on the VPN without manually adding the route after connecting to the VPN:

YouJiacheng

 INFO global: Vagrant version: 2.3.3
 INFO global: Ruby version: 2.7.6
 INFO global: RubyGems version: 3.1.6
 INFO global: VAGRANT_EXECUTABLE="C:\\HashiCorp\\Vagrant\\embedded\\gems\\2.3.3\\gems\\vagrant-2.3.3\\bin\\vagrant"
 INFO global: VAGRANT_INSTALLER_EMBEDDED_DIR="C:\\HashiCorp\\Vagrant\\embedded"
 INFO global: VAGRANT_INSTALLER_ENV="1"
 INFO global: VAGRANT_INSTALLER_VERSION="2"
 INFO global: VAGRANT_LOG="debug"
 WARN global: resolv replacement has not been enabled!
DEBUG global: Loading core plugin: C:/HashiCorp/Vagrant/embedded/gems/2.3.3/gems/vagrant-2.3.3/plugins/commands/autocomplete/plugi

YouJiacheng

from dataclasses import dataclass


class Default(dict[str, str]):
    def __missing__(self, key: str):
        return f"{{{key}}}"


@dataclass
class Pair:

YouJiacheng

import os
import base64


# dynamically generated at test time
A_identifier = base64.urlsafe_b64encode(os.urandom(6)).decode()
B_identifier = base64.urlsafe_b64encode(os.urandom(6)).decode()


meta_prompt = f"""

YouJiacheng

import os
from typing import cast

import torch
import torch._inductor.config as config
import torch.distributed as dist


def zeropower_via_newtonschulz5(G, steps=10, eps=1e-7) -> torch.Tensor: ...

YouJiacheng

import torch
from torch.distributed.device_mesh import init_device_mesh
from torch.distributed.tensor import DTensor, Shard

mesh_1d = init_device_mesh("cuda", (4,), mesh_dim_names=("shard",))

rank = mesh_1d.get_rank()

dtensors: list[DTensor] = []

YouJiacheng

import torch
import torch.utils.benchmark as benchmark


def benchmark_in_us(f, *args, **kwargs):
    t0 = benchmark.Timer(
        stmt="f(*args, **kwargs)", globals={"args": args, "kwargs": kwargs, "f": f}
    )
    return t0.blocked_autorange().mean * 1e6

YouJiacheng

import os
import sys

import torch._dynamo.compiled_autograd

with open(sys.argv[0]) as f:
    code = f.read()  # read the code of this file ASAP, for logging
import uuid
import glob
import time