[cups.general] duplicate open sockets

[Matt LaPlante]

Cups 1.4.5

When a given printer is offline for awhile, I've noticed that the server may
build up multiple socket:// processes, even for the same print job.

lp        7139  0.0  0.0  38636  2052 ?        SN   Mar14   0:00
socket://hp:9100 83696 bob smbprn.00000001 http://url/ 1
job-originating-host-name=172.2.3.4
job-uuid=urn:uuid:05ba1e33-7260-3de7-79d9-ecb2f4e7dd9b
time-at-creation=1299881442 time-at-processing=1300134839 AP_D_InputSlot=
lp       10971  0.0  0.0  38636  1596 ?        SN   Mar11   0:01
socket://hp:9100 83696 bob smbprn.00000001 http://url/ 1
job-originating-host-name=172.2.3.4
job-uuid=urn:uuid:05ba1e33-7260-3de7-79d9-ecb2f4e7dd9b
time-at-creation=1299881442 time-at-processing=1299886084 AP_D_InputSlot=
/var/spool/cups/d83696-001
lp       17845  0.0  0.0  38636  1596 ?        SN   Mar14   0:00
socket://hp:9100 83696 bob smbprn.00000001 http://url/ 1
job-originating-host-name=172.2.3.4
job-uuid=urn:uuid:05ba1e33-7260-3de7-79d9-ecb2f4e7dd9b
time-at-creation=1299881442 time-at-processing=1300112358 AP_D_InputSlot=
lp       22299  0.0  0.0  38636  2040 ?        SN   08:05   0:00
socket://hp:9100 83696 bob smbprn.00000001 http://url/ 1
job-originating-host-name=172.2.3.4
job-uuid=urn:uuid:05ba1e33-7260-3de7-79d9-ecb2f4e7dd9b
time-at-creation=1299881442 time-at-processing=1300201547 AP_D_InputSlot=

Process 7139 attached - interrupt to quit
restart_syscall(<... resuming interrupted call ...>) = 0
socket(PF_INET, SOCK_STREAM, IPPROTO_IP) = 5
setsockopt(5, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
setsockopt(5, SOL_TCP, TCP_NODELAY, [1], 4) = 0
fcntl(5, F_SETFD, FD_CLOEXEC)           = 0
connect(5, {sa_family=AF_INET, sin_port=htons(9100),
sin_addr=inet_addr("172.2.3.5")}, 16) = -1 ETIMEDOUT (Connection timed out)
close(5)                                = 0
write(2, "DEBUG: Connection error: Connect"..., 46) = -1 EPIPE (Broken pipe)
--- SIGPIPE (Broken pipe) @ 0 (0) ---
write(2, "ERROR: recoverable: Unable to co"..., 78) = -1 EPIPE (Broken pipe)
--- SIGPIPE (Broken pipe) @ 0 (0) ---
rt_sigprocmask(SIG_BLOCK, [CHLD], [], 8) = 0
rt_sigaction(SIGCHLD, NULL, {SIG_DFL, [], 0}, 8) = 0
rt_sigprocmask(SIG_SETMASK, [], NULL, 8) = 0
nanosleep({30, 0}, 0x7fffc7116a70)      = 0
socket(PF_INET, SOCK_STREAM, IPPROTO_IP) = 5
setsockopt(5, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
setsockopt(5, SOL_TCP, TCP_NODELAY, [1], 4) = 0
fcntl(5, F_SETFD, FD_CLOEXEC)           = 0
connect(5, {sa_family=AF_INET, sin_port=htons(9100),
sin_addr=inet_addr("172.2.3.5")}, 16) = -1 ETIMEDOUT (Connection timed out)
close(5)                                = 0
write(2, "DEBUG: Connection error: Connect"..., 46) = -1 EPIPE (Broken pipe)
--- SIGPIPE (Broken pipe) @ 0 (0) ---
write(2, "ERROR: recoverable: Unable to co"..., 78) = -1 EPIPE (Broken pipe)
--- SIGPIPE (Broken pipe) @ 0 (0) ---
rt_sigprocmask(SIG_BLOCK, [CHLD], [], 8) = 0
rt_sigaction(SIGCHLD, NULL, {SIG_DFL, [], 0}, 8) = 0
rt_sigprocmask(SIG_SETMASK, [], NULL, 8) = 0
nanosleep({30, 0}, 0x7fffc7116a70)      = 0

This seems odd.  Shouldn't there only be one socket per job?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cups.org/pipermail/cups/attachments/20110315/acf5bfa7/attachment-0001.html>