Bugzilla – Bug 8180
Squeezecenter prevents user logout after restart
Last modified: 2012-05-11 10:28:42 UTC
If a user restarts squeezecenter, they cannot log out. If (using SecureCRT 3.2.1, Windows XP Home Edition Version 2002 Service Pack 2) I ssh into my CentOS 4.6 box and restart squeezecenter, I cannot log out because squeezecenter (and mysqld and mDNSResponderPosix) still have my pty open on stdin. Redirecting stdin to /dev/null in /etc/init.d/squeezecenter, line 59 fixes the problem.
Ross: Can you investigate this one
Strange this worked for me, I was able to logout without issue. Have you tried SqueezeCenter 7.0.1?
I installed 7.0.1 -- no joy. -fs::sedwards:~$ sudo lsof | grep squeeze | grep pts squeezece 15431 squeezecenter 0u CHR 136,0 2 /dev/pts/0 mysqld 15434 squeezecenter 0u CHR 136,0 2 /dev/pts/0 mDNSRespo 15468 squeezecenter 0u CHR 136,0 2 /dev/pts/0 I cannot log out from my CentOS 4.6 server when I restart (/etc/init.d/squeezecenter restart) Squeezecenter. If I redirect stdin to /dev/null, then I have no problem.
Michael Herger says he's seen this as well. Fletch, the 'fix' in comment 1 seems like a hack. Is there a better fix that we should do? Thanks!
Hmm. I never noticed this before, but it is easily reproducible on my CentOS 5 system. Interestingly, on Fedora 9, the pty is still open on stdin, but it doesn't prevent me from logging out. I'm not really sure what the "right" fix is. Maybe Andy or one of the other perl guys know of a way to prevent slimserver.pl from opening stdin in the first place.
Matt, is this something you could help with?
If no one has a better idea, I'll make the </dev/null change for 7.1. We can always change it later if/when something more elegant comes along...
BTW, has anyone tested this on Debian/Ubuntu?
(In reply to comment #8) > BTW, has anyone tested this on Debian/Ubuntu? I have and I cannot reproduce, but I wasn't able to reproduce with CentOS either, and I just tried again on both and still unable. From a remote PC all we're doing is /etc/init.d/squeezecenter -stop as root then starting, then exit/logout, right?
I just tested the suggested fix from comment 1 and it doesn't work for me on CentOS 5. Not sure why...
Since we don't have a good fix for this yet, I've added a note to the SC RPM wiki page.
Until we can come up with a good fix, the workaround will have to suffice.
This is standard for many applications on Unix and Unix-like systems. Whether it should be is a different question. The standard workaround is to restart the service using the nohup command. For example: nohup /etc/init.d/squeezecenter restart &
Per Alan's comments this should be considered in the future.
it's a bug not an enhancement ? My system: Version: 7.6.0 - r31915 @ Fri Feb 18 08:39:50 MST 2011 Hostname: hal.home.lan IP: 192.168.1.5 HTTP Port: 9000 OS: Red Hat - EN - utf8 Platform: i686-linux Perl Version: 5.8.8 - i686-linux-thread-multi Database Version: DBD::SQLite 1.32_01 (sqlite 3.7.4) Total Players Recognized: 3
And restart command does not work due to another bug...
Will this issue be covered by bug 17729?
Fixed as a side effect of cleanup done in 17729. r33893
..and this works for me too as I upgraded my OS never linux and never perl must have done some good . FYI you may still have this problem if you run CC4.2 OS ,but as I was the only tes tcase for that and have scrapped that server so I can't test anymore.