Episode #125 of the Stack Overflow podcast is here. We talk Tilde Club and mechanical keyboards. Listen now
5

No, a basic Bourne/POSIX shell (/bin/sh) cannot be expected to include any built-in facilities for TCP connections. See comparision of command shells in Wikipedia. The bash shell would have TCP and UDP client capabilities with a special handling of certain filenames: for example, using /dev/tcp/<hostname>/<port> in input/output redirection on a ...


5

While I am not sure if this will work on I350 NIC I think the answer is in the ip-link man page: vf NUM specify a Virtual Function device to be configured. The associated PF device must be specified using the dev parameter. [--cut--] state auto|enable|disable - set the virtual link state as seen by the specified VF. Setting to auto means a ...


2

Your title and question do not really match. Some services block the list of well-known-ports, some don't. If your service tries to bind to a WKP and it is unoccupied, it will work (but needs superuser-rights on unix like systems). So you can use them in a non-standard way, but really shouldn't. But if you create your own HTTP service, for example, you are ...


2

Here, our Docker container is reachable through docker0 (network: 172.17.x.x) From your browser, you're trying to connect to some 10.x.x.x address. Using a Browser on the Docker host Assuming you just want to connect from your browser to a locally-hosted container, then the easier would just be to reach your container address (172.17.x.x). In some cases, ...


1

This sounds like an MTU issue. Because you are using PPP over Ethernet the maximum packet size is reduced - this can cause issues with large packets not being forwarded. Packets sent directly from the router are smaller because they use the smaller MTU of the PPP interface. One way of solving this problem for TCP traffic is MTU clamping - try add ...


1

It is very unlikely that the problem needs to be solved at the system level instead of the application level. If something is lost in TCP it gets resent by the peer - this is how TCP is designed. It is more likely that there are wrong assumption about how TCP works in the Java application. For example it is common error to treat TCP as a message based ...


1

In general, what you depicted is exactly an ECMP route. Although dynamic routing protocols can (and will) install such routes, if you don't expect your network topology to change, running an dynamic routing protocol adds complexity and I feel that is an overkill. OSPF is not a piece of cake, trust me. Sure, you can install such route by hand: ip route add ...


1

If your purpose in having multiple routes is robustness (redundancy in case of failure of a mid node), run OSPF on all five nodes. One option is Quagga, which provides OSPF and other routing protocols for for various Linux flavors including CentOS. Running OSPF on the nodes makes each one of them a router, and communications between master and slave will ...


1

Absolutely. You can do this using OpenVPN among other VPN or tunneling software. Ideally 1. Ensure That each office is on a different, non-overlapping subnet. 2. Use routers that support acting as openvpn clients (dd-wrt can be flashed onto many Ethernet routers and does this)


1

To fully understand packet flow through Linux iptables, you will need this diagram or something similar for reference. In the diagram, ping lives in the "local process" box in the application layer. With your current routing table, the system wil "think" only traffic addressed to 10.8.0.0/24 will be routed to the VPN tunnel interface; everything else must ...


1

The Debian wiki mentions what may be the same problem, for Stretch and Buster. If your system has the file /etc/network/interfaces.d/setup (mine does not) you could try removing it. Otherwise, I like your idea of adding the bridge on startup. You could try in /etc/network/interfaces: auto br0 iface br0 inet dhcp pre-up brctl addbr br0 && brctl ...


1

I had exactly the same problem as you (I even referenced your serverfault question in this posting to the Debian User Forums) My problem turned out to be another instance of "Boottime Entropy Starvation" (I'm guessing that wpa-supplicant needed some random data for key exchange, and that the kernel blocked while waiting for the random number generator to ...


1

Check the proxy settings for subscription-manager in this below file: /etc/rhsm/rhsm.conf Set the correct value for proxy_hostname and proxy_port. If needed fill out the proxy_user and proxy_password lines as well.


Only top voted, non community-wiki answers of a minimum length are eligible