michael's questions

I have an ubuntu server behind a router that I am configuring to have a GRE tunnel to another remote device across the interwebs.

in the config, I add the line:

pre-up iptunnel add tun1 mode gre local [local ip] remote [remote ip] ttl 255

Since the ubuntu server is behind a router, should the [local ip] be the wan ip of the router? or the actual local ip of the ethernet interface i wish the tunnel to connect through?

The router the server sits behind doesn't say anywhere anything about forwarding GRE 47 to lan devices, so I just DMZ'd the server (no ideal in product, but I am just testing using GRE to allow for multicast discovery from 1 subnet to another)

Here is the full script (with local ip set as lan ip of ubuntu server)

auto tun1
iface tun1 inet static
    address 10.0.33.1
    netmask 255.255.255.0
    pre-up iptunnel add tun1 mode gre local 192.168.33.72 remote x.x.x.x ttl 255
    up ifconfig tun1 multicast pointopoint 10.0.33.2
    post-down iptunnel del tun1

I have a small linux box with a dedicated task of reading sensor data and comparing values against a read only (ie i won't write stuff to it, but haven't done anything to make it actually readonly) local database.

The system doesn't need to write anything onto the file system,except maybe some temp files created when the local web service is accessed (web page shows the data, but doesn't log it)

The system autoboots when power is applied, and it is turned off by just cutting power.

Is there a good guide for setting up a system to be cool with abrupt power offs? I am currently using ubuntu (cause i am familiar with it and it runs well on the fit pc 2 that i am using)

Or given that I have no applications writing anything to disk (web server temp files are the exception), can I get away with not modifying the system? The only important point is there is no human to intervene if the boot process throws a question out the terminal and hangs waiting for a response.

I have four host computers running lighttpd webservers. they are sitting behind a hspa modem, which each occupying a http port between [81 - 84]. 80 is taken by the modem itself.

The port forwarding is setup correctly, however, only a portion of any webpage I request from any of the hosts comes through (they all fails after %20 of the page).

If I put the host on port 81 into the dmz, it serves pages fine. The others do not respond to the dmz treatment.

Is it possible the web content on the hosts somehow require ports aside from their respective http port? Or is it possible that even though the server.port in the lighttpd_ssl.conf file is set, the individual hosts are still expecting to serve on port 80?

I am not familiar with lighttpd, nor did i set them up. they are running on video encoders i purchased. I can grab any files from them required for further information on the problem.

I deploy cellular modems that currently have sims with static ip addresses. For various reasons, we may move to providing sims without static ips.

The cellular modem supports DDNS (bluetree modem).

Since we will have a very low query per second rate (once an hour max), but a high deployment rate (hundreds to thousands), my needs are somewhat different than what you would find with many dynamic dns service customers. I think.

Should I look into running my own dns service (i have not the foggiest idea in what that involves), or is there a service provider geared towards low QPS, high deployment. Or will something like DynDNS be fine?

All devices will be deployed in the same country (Canada), and for quite a while only in western canada.

I have a small embedded system that supports fat32. It does not have any encryption options for the files that it saves to a external usb connected disk. Modifying the source to support encryption is not an option. Are there any external drives, or somethings that can be used to encrypted the files once the land on the disk? Something that does not require the cooperation of the embedded system?

I realize that this is a bit far fetched, but you never know!

I am re-purposing some old Corinex powerline ethernet wall mounts. They came with a setup cd with a wizard application for setting up the hardware, but the application doesn't seem to work on 64bit windows 7 (not surprised really - the plugs are from 2004).

Anyone remember how to configure these things without the software tool? I have no idea what there current configuration on the hardware is.

I plugged them into separate wall sockets (the link status led indicated a link between them) and tried generating traffic and then looking at my arp table to see if i could see stuff, but nothing came up. My networking skills are limited at best.

We deploy windows machines with a 3rd party application that we modify slightly (config files and images) just after we install it on the soon to be deployed systems. This application does not have any auto update feature, but does have updates come out once every couple of months in the form of new install packages. Everytime you want to update the software, you have to download it, uninstall the old one, and install the new one.

Having the client manually update the software is a no go. RDP is not ideal either as there would be privacy issues.

Is there any good client/server update software that we can use to push updates to our clients computers from ours?

--edit I was just looking at the System Center Configuration Manager http://bit.ly/lxzoBr

At first glance it looks promising. Any thoughts?

I have a client who resells our systems enclosed for use in industrial environments. We have been looking at installing a small dc powered monitor, but have run into a few problems which has made me start thinking about using a netbook plus RDP instead (there is easy access to a free ethernet jack)

The issues we ran into with using a small dc powered lcd screen are:

Has to be at least 1024 x 768
Has to have DVI input (the computer is pure dvi, no vga)
Has to accept 12V DC

So far I haven't found anything that satisfies all of the criteria.

Plus they need to have a keyboard/mouse combo. Which leads to a bit of juggling.

The person going into the system is not an IT person. They will be going in to configure our software (and nothing else) which they are trained to use.

So I was thinking of providing a netbook instead, and have them access the system via RDP or the equivalent.

Which is the better way to go? I like having a monitor keyboard setup that satisfies the criteria over using a netbook+RDP because it seems like there is less potential for screwups. But so far I haven't found the appropriate solution.

Or is there something better that I haven't seen?