Take the 2-minute tour ×
Server Fault is a question and answer site for system and network administrators. It's 100% free, no registration required.

We recently upgraded from Centos 4.3 to 5.9 and have come across a perplexing condition with regard to a very few direct-character printers we use. Since the beginning of Unix we have placed a command nohup sleep 1024000

Now, with Centos 5.9 Linux 2.6.18-348.6.1.el5xen #1 SMP Tue May 21 16:10:52 EDT 2013 x86_64 x86_64 x86_64 (on a Proliant server) we find the sleep is in place but the ending characters get truncated as if the port is in fact closed.

Interestingly on 4.3 the command /sbin/fuser /dev/ttyUSB0 would produce the PID of the sleep, but on 5.9 it does not. As an experiment, I rebooted the system and ran the following command on one port cat - </dev/ttyUSB0 and on another I typed /sbin/fuser /dev/ttyUSB0 and again the process id of my running port is not reported.

Yes, there are 100 other ways to process the print output, but anything the USED to work and suddenly STOPS working requires an explanation

Did I miss something in a README somwhere?

share|improve this question
Can you give more details about how you're using this? I assume this is in some script somewhere? Are you ssh'ing to the server, launching the script and backgrounding it, then disconnecting or something? I'm not sure why a sleep command should keep a TTY open in the first place and perhaps more details would help. –  Isaac Freeman Jun 9 '13 at 17:30

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.