We're experiencing what we believe to be a bug in 10-printing.xzm.
A little backstory here. In early January, we began experiencing several events wherein our corporate Cisco ASA router would lockup for unknown reasons.
In one such event, we were lucky enough to have been watching the logs in real-time and saw the following line repeated literally thousands of times:
Code: Select all
4|Jan 18 2019|11:38:27|419002|10.0.0.123|33634|10.0.0.10|631|Duplicate TCP SYN from inside:10.0.0.123/33634 to insidesecondary:10.0.0.10/631 with different initial sequence number
In the above 10.0.0.123 is a instance of Porteus Kiosk and 10.0.0.10 is a printer physically connected to the insidesecondary on the LAN with kiosk_config configured to use it as their printer for all instances of Porteus Kiosk.
We're not aware of what is causing the above error in the logs. It appears to be an error sending a print job to the printer but we're not really concerned with why it is failing. What we are concerned with is that it appears that when it tries to send a print job, it is not failing elegantly seems to be attempting to resend it to the extent that it becomes a quasi-denial of service attack with the capability of taking down a corporate Cisco router.
For the record, we did test sending a print job from the kiosk on ip 10.0.0.123 and we could not replicate the problem.
However, in real-world experience, we have had he printing module disabled all instances of Porteus Kiosk for almost a month and have not had a recurrence of the router issues we were plagued with in early January since we disabled it.
We strongly suspect that their is a bug in 10-printing.xzm such that it potentially will not fail elegantly if there are issues sending data the printer.
Thanks in advance for looking into this.
Tim Levine