Security Gateway Hangs, No Internet, No DHCP

[Theo]

Appliance: Checkpoint 2200

OS Version: R80.30

Type: Standalone

Satellite Gateway

One of our office had this practice that whenever there is a long days of no work, or long holidays, he pulled out the power of CheckPoint appliance to turn it off (even I advised him to not practiced it).

Recently after lunar new year vacation, they returned to office, turn on the device and it started to happen the following:

• Security gateway stops/ hangs in no permanent time (sometimes after an hour, or just few minutes after it was restarted)
• The DHCP stopped, clients obviously didn't get correct local IP addresses
• Internet connection becomes broken even local services like printers, access to local server
• Not able to respond even I connect the console cable (unless force reboot).
• To sum up, the security gateway looks hanging and stopped totally.

I would like to ask if there are similar incidents you guys successfully fixed and encounter? At the moment, no other way but to force OFF and ON the device to come up again. I would appreciate some leads and advices on how can this be fixed.

Is this a possible hardware problem? And if yes, can I run something to identify the possible reason?

Thanks in advance.

[Maarten_Sjouw]

Yep, a 2200 is very bad with disc recovery when you just pull the plug, write operations fail and can corrupt the disc. Your only way forward is to reinstall the box. We had a lot of these units in the field, also in countries where power is not so stable, and we always advised the customers to install a UPS to prevent problems like this.

[PhoneBoy]

To troubleshoot this, you need to use the serial console and see what messages come across.
This could easily be a hardware failure but without seeing console messages, that's only a guess.

[Christopher__C2]

If I understand correctly, everything stops working, you connect a console cable and get no output from it.

Reboot it, and it starts working again, console connection works?

It is most likely hardware. Something forced the kernel to probably just stop working correctly.

You can check /var/log/messages and find what, if anything, was logged while the system stopped responding.  And check /var/log/crash and /var/crash for core files that might be left.  But if the disk is failing (usually the least reliable part), it might have stopped writing anything to disk.

Another option would be to leave a device connected to the console logging output from the 2200. Take care to make sure this device is adequately secured, especially if left logged into the console.  Possibly run "dmesg -w" from the console (in expert mode), to catch any error output from the kernel that might not be saved any other way when it stops responding.

These messages are usually not that easy to understand, but usually errors from hardware failures, especially the disk, show up quickly in google searches and can help validate that the device is failing.