[Jonathan Brewer]

Is there some reason no one from Multitech wants to comment here?

The requirements below are wrong. They contravene US govt standards published in NIST 800-63-3: https://pages.nist.gov/800-63-3/sp800-63-3.html

`The user password must meet the complexity requirements and be at least 8 characters and contain three or more different types of characters:

uppercase alphabetical characters (A through Z)
lowercase alphabetical characters (a through z)
numerals (0 through 9)
special characters

The password must not contain any common dictionary word.

[Jonathan Brewer]

Need a solution here Multitech.

[Jonathan Brewer]

Hi Jeff. It was an upgrade from a previous version – probably an early 2016 firmware.

[Jonathan Brewer]

This was with a brand-new unit that had no configuration. In the “out of the box” state of 1.1.2, Eth0 is DHCP. When the 1.2.2 upgrade is applied, the unit sets itself to static.

I verified again with a second unit. Four more to configure in this batch!

[Jonathan Brewer]

For all: advice from support was to set up a reboot every 24 hours.

[Jonathan Brewer]

Logging into one of these Conduit units, I see its PPP interface is not up, and doesn’t have an IP address, even though it’s got full cellular signal. It was stuck in this state for 2 days.

Nothing in /var/log/messages indicated it was trying to reestablish a cellular connection.

In /var/log/ppp_trace

7:38:2: Connecting for 0 time
7:38:5: Serial connection established.
7:38:9: IPCP: LAYER IS UP
7:38:9: PPP Established
07:38:22 PPP Rx Monitor:Started Monitoring
06:50:10: PPP Rx Monitor: Redialing…
6:50:10: IPCP: Layer is down

and in /var/log_ppp_trace0

20:3:46: Connecting for 0 time
20:3:47: Physical link failed
20:3:48: Current timing 60
20:3:48: Sleeping 58 seconds before next attempt
20:4:46: Connecting for 1 time
20:4:49: Serial connection established.
20:4:56: IPCP: LAYER IS UP
20:4:56: PPP Established
20:05:15 PPP Rx Monitor:Started Monitoring
7:37:29: IPCP: Layer is down
7:38:2: Current timing 41

It seems like there is a software bug where if the cellular connection is up but something goes wrong with the IP layer, the Multitech just sits there and does nothing. 🙁

[Jonathan Brewer]

We’ve got two Conduit units with this behaviour. Full signal, but they lose and don’t re-establish their cellular connection. Very troubling.

[Jonathan Brewer]

Hi Multitech, can you please help? In particular I would like to see evidence that the 868 card meets EN 300 220-1 V2.4.1.

[Jonathan Brewer]

Making those changes persistent should require:

# /usr/sbin/iptables-save

But in fact this script doesn’t make the rule change persist. 🙁

• This reply was modified 8 years, 5 months ago by Jonathan Brewer.

[Jonathan Brewer]

Hi Jeff,

I’m working with a MTCDT-H5-210A Firmware 1.0.33. I have looked at the web interface. The firewall configuration menu only has options for TCP/UDP and ANY.

From the command line I’ve enabled ICMP echo response with:

# iptables -A INPUT -p icmp --icmp-type 8 -s 0/0 -d $router_ip -m state --state NEW,ESTABLISHED,RELATED -j ACCEPT

It would be excellent if in future firmware you always enable ICMP – or at least make it easy to enable in the web UI. Some support for IPv6 would also be appreciated.

Thanks,

Jon

• This reply was modified 8 years, 5 months ago by Jonathan Brewer. Reason: highlighted the code
• This reply was modified 8 years, 5 months ago by Jonathan Brewer.

[Author]

Posts