1 |
On 04/09/2017 22:16, Grant wrote: |
2 |
>> I have a network of very nearly identical Dell XPS 13 laptops that I |
3 |
>> manage with a script. The master pushes the contents of its |
4 |
>> filesystem to the others so I only have to manage one system. It's |
5 |
>> worked really well over several years. I just got a new Dell XPS 13 |
6 |
>> to serve as the master and there have been some changes that were |
7 |
>> difficult to integrate with the network (high-res screen, /dev/sda |
8 |
>> replaced with /dev/nvme0n0) but those problems are fixed thanks to you |
9 |
>> guys. |
10 |
>> |
11 |
>> Now I'm running into "trap invalid opcode" errors on the older |
12 |
>> systems. Can I disable some of the newer CPU instruction sets on the |
13 |
>> master laptop when compiling to hopefully generate binaries that will |
14 |
>> work on the older systems? If so, could anyone point me in the right |
15 |
>> direction? I don't want to use distcc please. |
16 |
>> |
17 |
>> CHOST="x86_64-pc-linux-gnu" |
18 |
>> CFLAGS="-march=native -O2 -pipe -fomit-frame-pointer" |
19 |
> |
20 |
> |
21 |
> Switching to -mtune=native seems to work. Time for an emerge -e world. |
22 |
|
23 |
Also time for ansible. Why you managing a fleet of machines with a |
24 |
script that won't actually differentiate properly between machines? It |
25 |
will sorts mostly do it right, except when you forget something. |
26 |
|
27 |
This is exactly the use-case ansible was designed for: declarative, |
28 |
idempotent, predictable management of a fleet of machines that may or |
29 |
may not be around when you feel like updating something (so it catches |
30 |
up later), and needs only sshd and python to do it's magic :-) |
31 |
|
32 |
Never mind that ansible was written with servers in mind; in terms of |
33 |
management where you do $STUFF_THAT_NEEDS_MANAGING, there is no |
34 |
difference between servers and laptops. A computer is still just a computer. |
35 |
|
36 |
</me also laments the lack of poudriere on Gentoo. But that's for |
37 |
another time> |
38 |
|
39 |
-- |
40 |
Alan McKinnon |
41 |
alan.mckinnon@×××××.com |