On 08/07/2019 21:13, Greg Wooledge wrote: > On Mon, Jul 08, 2019 at 03:09:45PM -0500, David Wright wrote: >> On Mon 08 Jul 2019 at 14:55:45 (-0400), Greg Wooledge wrote: >>> On Mon, Jul 08, 2019 at 02:49:40PM -0400, Stephen P. Molnar wrote: >>>> It came from the Biovia web site. >>> Like I know what a "Biovia" is. >> It looks like a company to me. >> >>> I'm guessing it has something to do with "Molecular-Modeling", and >>> therefore I lump it into the broad category of "academia", which means >>> it's poorly written but "works" as long as you build your entire machine >>> to conform to whatever it requires. >> Like Gnu, Linux, TeX, Python, …? > Like scripts that are written for bash but with a #!/bin/sh shebang > because they only tested on one CentOS box. > > Wouldn't surprise me if they parse the output of /sbin/ifconfig eth0 to > get information to create/validate a license key. And oh boy will THAT > be fun, if that's the case, now that eth0 is not an officially valid > interface name, and /sbin/ifconfig is not installed by default, and even > if you DO install it, its output format has changed.
Point of order. There' s nothing wrong with "eth0" as an interface name. It's perfectly valid. It's perfectly legal. It's just not the default. If you want to rename interfaces, that's a well-respected path. It may be that "eth0" (zeroth ethernet device) is a more meaningful name to you than enp0s1 (ethernet card in the first PCI slot) or "dmz" (the ethernet device connected to the DMZ). If so, go for it. > > But maybe I'll be pleasantly surprised and the /bin/sh symlink will be > the only backflip required. >
