According to course material :

"Processes are assigned pids in sequential order. Eventually the system will read the maximum pid value which is defined in /proc/sys/kernel/pid_max. Once the maximum pid value is reached, the system will roll over and assign future pids from the bottom of the range."

This is a little confusing. I understand the need to rollover obviously. But what exactly does "from the bottom of the range" ?

So right now in a VM I access 32768 is defined in pid_max. When the system reaches this number ... does it roll back to 1 (well actually 2) or, in example, if 32600 is free, will that number be assigned to the next process ?

I wanted to edit this to clarify the question a little bit..

Systems talking to us
"Ive reached the max pid of 32768."
"Start from the bottom and see if 32767 is open"
"No check 32766"
"Yes assign 32767 to new process"

or

Systems talking to us
"Ive reached the max pid of 32768."
"Start from the top and see if 2 is open"
"No check 3"
"Yes assign 2 to new process"


Thanks alot for the help !

It will return to 1 and use that, but only if 1 is free. The earliest process numbers many times end up being assigned to services which never terminate. If you perform "ps -e" you should see all processes and their PIDs which are currently in use. Therefore it will choose the next available PID value which isn't being used by another process.

Like on my system, 1, 2, and 3 are still taken. 4 is unused, so if/when my PIDs wrap, it will eventually use 4, and next 5. But since 6, 7, and 8 are taken, it will skip those and use 9.

1 members found this post helpful.

Ok thanks thats what I thought .. but the text wording is a little confusing.

Thanks man !!