Printer causing segmentation fault

[morlanius]

I'm using a git version as of today, when putting paper in the printer it causes the server to segmentation fault with nothing in the debug log, it just halts.

[bell07]

Which Minetest version? Windows or Linux? Which build?
In case of Linux: Any messages in console if you run the Minetest from command line?

[morlanius]

build 0.4.16 running on linux, nothing in console except "segmentation fault" and nothing in the log.

looks like its something to do with the itemstack for the paper it behaves strangely, duplicates the paper then crashes.

[bell07]

Please try 'minetest --verbose' or 'minetest --trace' if you see more informations about what happens.
Please try also to upgrade to current stable 0.4.17.1

[morlanius]

I can't upgrade to 4.17 just yet the build is having problems and not letting clients connect.
I enabled trace and nothing, you can just see the data packets being send and then a seg. fault.

It seems to be something with the input methods to the printer. This is what happens exactly.

• switch on printer
• right click for gui
• paper is selected by default so I put a stack of paper, the stack vanishes and the paper value does not increase.
• I select ink, and put a stack of black ink in there, the stack vanishes and the ink value does not increase.
• reselect paper and put another stack in, the number goes to 99 and the stack remains in the slot
• remove paper
• select ink and put a stack of black ink
• a full 5 seconds later
• segmentation fault with nothing in the log.

I have tried variations of this and they all lead to the seg. fault. If i put the ink in first, paper in, leave the stack in the slot and switch between paper and ink buttons does nothing unless they have been inserted twice already.

If this was an issue in a C or java program I would suggest that the contents of the item slot is being dereferenced somewhere, I don't know about this environment though.

[morlanius]

Just logged in and tested something else, with the paper value at 99 (having had paper in there, took out and put back in again) then the server rebooted. So, I log in and the paper value is already at 99, I put ink in there and nothing happens, I take it out and put it back in and seg fault.

so something screwy is up with the way its counting up paper, and theres something in the equivalent routine for the ink thats causing he crash.