Discussion:
cups not starting on boot after last 2 kernel updates? -- where to put it in the DAEMONS line?
(too old to reply)
David C. Rankin
2011-08-08 16:31:19 UTC
Permalink
Guys,

Cups has been working like clockwork for the past year - up until the last 2
kernel updates. Now after boot, a manual '/etc/rc.d/cupsd restart' is required
to get it working again. Would moving it in the DAEMONS line make sense?
Currently I have:

DAEMONS=(hwclock syslog-ng network named @netfs @ntpd @sshd @dhcp4 @crond
@mysqld @postfix @dovecot @httpd @samba @cups @vsftpd @sensors @spamd)

When cups fails to start, I get the following warnings and errors:

W [08/Aug/2011:10:53:39 -0500] Duplicate listen address "[v1.::1]" ignored!
W [08/Aug/2011:10:53:39 -0500] Duplicate listen address "127.0.0.1" ignored!
W [08/Aug/2011:10:53:39 -0500] Duplicate listen address "[v1.::1]" ignored!
W [08/Aug/2011:10:53:39 -0500] Duplicate listen address "127.0.0.1" ignored!
W [08/Aug/2011:10:53:39 -0500] Duplicate listen address "192.168.7.15" ignored!
W [08/Aug/2011:10:53:39 -0500] Duplicate listen address "192.168.7.15" ignored!
W [08/Aug/2011:10:53:39 -0500] Duplicate listen address "192.168.7.15" ignored!
E [08/Aug/2011:10:53:39 -0500] Bad File number 1 on line 3225!
E [08/Aug/2011:10:53:39 -0500] Bad File number 1 on line 3235!
E [08/Aug/2011:10:53:39 -0500] Bad File number 1 on line 3245!
E [08/Aug/2011:10:53:39 -0500] Bad File number 1 on line 3255!
E [08/Aug/2011:10:53:39 -0500] Bad File number 1 on line 3265!
E [08/Aug/2011:10:53:39 -0500] Bad File number 1 on line 3275!
E [08/Aug/2011:10:53:39 -0500] Bad File number 1 on line 3285!
E [08/Aug/2011:10:53:39 -0500] Bad File number 1 on line 3295!
E [08/Aug/2011:10:53:39 -0500] Bad File number 1 on line 3305!
E [08/Aug/2011:10:53:39 -0500] Bad File number 1 on line 3315!
E [08/Aug/2011:10:53:39 -0500] Bad File number 1 on line 3325!
E [08/Aug/2011:10:53:39 -0500] Unable to bind socket for address [v1.::1]:631 -
Address already in use.
E [08/Aug/2011:10:53:39 -0500] Unable to bind socket for address 127.0.0.1:631 -
Address already in use.
E [08/Aug/2011:10:53:39 -0500] Unable to bind socket for address
192.168.7.15:631 - Address already in use.
E [08/Aug/2011:10:53:39 -0500] Unable to set ACLs on root certificate
"/var/run/cups/certs/0" - Operation not supported


I have no idea what the Bad File number errors are, but the warnings are just
from the multiple explicit listen designations in the cupsd.conf for client access:

# Only listen for connections from the local machine.
Listen localhost:631
Listen phoenix.rlfpllc.com:631
Listen phoenix:631
Listen www.rlfpllc.com:631
Listen 192.168.7.15:631

I think the real issue is the 'Address already in use' errors, but that
doesn't explain why it cups starts just fine with a manual '/etc/rc.d/cupsd
restart'. If the address/port was actually already taken, then I would expect
the restart to fail as well. The the ACL error is fine - cups isn't using acls
on my box and I don't want to set the acl mount option in fstab, so I can live
with that error.

I've been through the forums, google and bug tracker task/22895, but still no
solution. Should I move cupsd earlier in the DAEMONS line? Anybody else seeing
cups fail to start on boot, but starting fine on a manual restart? Thanks for
any thoughts or help you can give.
--
David C. Rankin, J.D.,P.E.
Peter Bui
2011-08-08 16:41:39 UTC
Permalink
Post by David C. Rankin
Guys,
Cups has been working like clockwork for the past year - up until
the last 2 kernel updates. Now after boot, a manual '/etc/rc.d/cupsd
restart' is required to get it working again. Would moving it in the
@crond @mysqld @postfix @dovecot @httpd @samba @cups @vsftpd
@sensors @spamd)
You have @cups in DAEMONS ... probably need @cupsd?
--
Peter Bui
J. W. Birdsong
2011-08-08 23:18:59 UTC
Permalink
Post by David C. Rankin
Guys,
Cups has been working like clockwork for the past year - up until
the last 2 kernel updates. Now after boot, a manual '/etc/rc.d/cupsd
restart' is required to get it working again. Would moving it in the
@crond @mysqld @postfix @dovecot @httpd @samba @cups @vsftpd
@sensors @spamd)
W [08/Aug/2011:10:53:39 -0500] Duplicate listen address "[v1.::1]" ignored!
W [08/Aug/2011:10:53:39 -0500] Duplicate listen address "127.0.0.1" ignored!
W [08/Aug/2011:10:53:39 -0500] Duplicate listen address "[v1.::1]" ignored!
W [08/Aug/2011:10:53:39 -0500] Duplicate listen address "127.0.0.1" ignored!
W [08/Aug/2011:10:53:39 -0500] Duplicate listen address "192.168.7.15" ignored!
W [08/Aug/2011:10:53:39 -0500] Duplicate listen address "192.168.7.15" ignored!
W [08/Aug/2011:10:53:39 -0500] Duplicate listen address "192.168.7.15" ignored!
E [08/Aug/2011:10:53:39 -0500] Bad File number 1 on line 3225!
E [08/Aug/2011:10:53:39 -0500] Bad File number 1 on line 3235!
E [08/Aug/2011:10:53:39 -0500] Bad File number 1 on line 3245!
E [08/Aug/2011:10:53:39 -0500] Bad File number 1 on line 3255!
E [08/Aug/2011:10:53:39 -0500] Bad File number 1 on line 3265!
E [08/Aug/2011:10:53:39 -0500] Bad File number 1 on line 3275!
E [08/Aug/2011:10:53:39 -0500] Bad File number 1 on line 3285!
E [08/Aug/2011:10:53:39 -0500] Bad File number 1 on line 3295!
E [08/Aug/2011:10:53:39 -0500] Bad File number 1 on line 3305!
E [08/Aug/2011:10:53:39 -0500] Bad File number 1 on line 3315!
E [08/Aug/2011:10:53:39 -0500] Bad File number 1 on line 3325!
E [08/Aug/2011:10:53:39 -0500] Unable to bind socket for address
[v1.::1]:631 - Address already in use.
E [08/Aug/2011:10:53:39 -0500] Unable to bind socket for address
127.0.0.1:631 - Address already in use.
E [08/Aug/2011:10:53:39 -0500] Unable to bind socket for address
192.168.7.15:631 - Address already in use.
E [08/Aug/2011:10:53:39 -0500] Unable to set ACLs on root
certificate "/var/run/cups/certs/0" - Operation not supported
I have no idea what the Bad File number errors are, but the
warnings are just from the multiple explicit listen designations in
# Only listen for connections from the local machine.
Listen localhost:631
Listen phoenix.rlfpllc.com:631
Listen phoenix:631
Listen www.rlfpllc.com:631
Listen 192.168.7.15:631
I think the real issue is the 'Address already in use' errors, but
that doesn't explain why it cups starts just fine with a manual
'/etc/rc.d/cupsd restart'. If the address/port was actually already
taken, then I would expect the restart to fail as well. The the ACL
error is fine - cups isn't using acls on my box and I don't want to
set the acl mount option in fstab, so I can live with that error.
I've been through the forums, google and bug tracker task/22895,
but still no solution. Should I move cupsd earlier in the DAEMONS
line? Anybody else seeing cups fail to start on boot, but starting
fine on a manual restart? Thanks for any thoughts or help you can
give.
--
David C. Rankin, J.D.,P.E.
You need to start cupsd in rc.conf daemons... not cups
David C. Rankin
2011-08-08 18:28:19 UTC
Permalink
Post by J. W. Birdsong
You need to start cupsd in rc.conf daemons... not cups
Peter, J.W. -- thank you. (slaps self for stupidity) That is a
'forest-for-the-trees' issue it would have taken days to see!
--
David C. Rankin, J.D.,P.E.
Joe(theWordy)Philbrook
2011-08-09 15:25:17 UTC
Permalink
Post by J. W. Birdsong
You need to start cupsd in rc.conf daemons... not cups
Peter, J.W. -- thank you. (slaps self for stupidity) That is a
'forest-for-the-trees' issue it would have taken days to see!
Hmmmnnn. Now I'm confused... According to my console login prompt, I'm
currently using 2.6.39-ARCH And after a recent pacman -Syu cups stopped
loading for me until I changed my Daemons line to say cups rather than cupsd.

There had, if I remember right, been a {is it 'rc.confpacnew' or
'rc.conf.pacnew'?} that didn't have cups listed in the daemons line.
Anyway I remember putting "cupsd" the new rc.conf first, and when that didn't
work I edited it to just say "cups". I didn't think much about it at the time.
All I cared was that my printer worked again.

But why would one Arch Linux installation need "cups" in the daemons line
while another needs "cupsd"???
--
| ~^~ ~^~
| <?> <?> Joe (theWordy) Philbrook
| ^ J(tWdy)P
| \___/ <<***@ttlc.net>>
Karol Blazewicz
2011-08-09 15:44:10 UTC
Permalink
Post by Joe(theWordy)Philbrook
You need to start  cupsd in rc.conf daemons... not cups
Peter, J.W. -- thank you. (slaps self for stupidity) That is a
'forest-for-the-trees' issue it would have taken days to see!
Hmmmnnn. Now I'm confused... According to my console login prompt, I'm
currently using 2.6.39-ARCH And after a recent pacman -Syu cups stopped
loading for me until I changed my Daemons line to say cups rather than cupsd.
There had, if I remember right, been a {is it 'rc.confpacnew' or
'rc.conf.pacnew'?} that didn't have cups listed in the daemons line.
Anyway I remember putting "cupsd" the new rc.conf first, and when that didn't
work I edited it to just say "cups". I didn't think much about it at the time.
All I cared was that my printer worked again.
But why would one Arch Linux installation need "cups" in the daemons line
while another needs "cupsd"???
--
|  ~^~   ~^~
|  <?>   <?>       Joe (theWordy) Philbrook
|      ^                J(tWdy)P
Which version of cups needs 'cups' in the DAEMONS array?
Norbert Zeh
2011-08-09 20:25:39 UTC
Permalink
Post by Karol Blazewicz
Post by Joe(theWordy)Philbrook
You need to start  cupsd in rc.conf daemons... not cups
Peter, J.W. -- thank you. (slaps self for stupidity) That is a
'forest-for-the-trees' issue it would have taken days to see!
Hmmmnnn. Now I'm confused... According to my console login prompt, I'm
currently using 2.6.39-ARCH And after a recent pacman -Syu cups stopped
loading for me until I changed my Daemons line to say cups rather than cupsd.
There had, if I remember right, been a {is it 'rc.confpacnew' or
'rc.conf.pacnew'?} that didn't have cups listed in the daemons line.
Anyway I remember putting "cupsd" the new rc.conf first, and when that didn't
work I edited it to just say "cups". I didn't think much about it at the time.
All I cared was that my printer worked again.
But why would one Arch Linux installation need "cups" in the daemons line
while another needs "cupsd"???
--
|  ~^~   ~^~
|  <?>   <?>       Joe (theWordy) Philbrook
|      ^                J(tWdy)P
Which version of cups needs 'cups' in the DAEMONS array?
It seems that older versions of the cups package installed /etc/rc.d/cups and
not /etc/rc.d/cupsd. At least I had cups in the DAEMONS array, and it always
worked until I did some pacman -Suy. It didn't take long to figure out,
though, that the name of the script was now changed to cupsd.

Cheers,
Norbert
Karol Blazewicz
2011-08-09 20:31:10 UTC
Permalink
Post by Norbert Zeh
It seems that older versions of the cups package installed /etc/rc.d/cups and
not /etc/rc.d/cupsd.  At least I had cups in the DAEMONS array, and it always
worked until I did some pacman -Suy.  It didn't take long to figure out,
though, that the name of the script was now changed to cupsd.
Cheers,
Norbert
If I'm reading this [1] right, daemon name change occurred on
2011-07-14 on cups version update to 1.4.7-2.

[1] http://projects.archlinux.org/svntogit/packages.git/commit/cups/trunk/PKGBUILD?id=6fabc2134a9106574606adab5382e282095a0421
Continue reading on narkive:
Loading...