[eside-ghost] Problemas con sendmail

[OASPIAZU en terra.es]

 Aupa gente!

Resulta que estamos trasteando con un redhat por aquí en la que estoy 
investigando un poco un montón de correos encolados con sendmail. Me he 
quedado bloqueado ya que el log no me dice gran cosa y no tenemos ni 
idea de strace

Os comento para el que quiera leerse un poco la chapa:

#ls /var/spool/clientmqueue | wc -l
980

(son unos cuantos archivos ;-)

#tail -f /var/log/maillog
Apr 1 09:32:21 darkstar sendmail[32728]: k317WL0a032728: from=, 
size=873, class=0, nrcpts=1, msgid=<200604010732.k317WL7s032726@," 
target="_BLANK">www.oinatz.com>, bodytype=8BITMIME, proto=ESMTP, 
daemon=MTA, relay=darkstar.euskaltel.es [127.0.0.1]
Apr 1 09:32:21 darkstar sendmail[32726]: k317WL7s032726: to=yo en oinatz.
com, ctladdr=yo en oinatz.com (48/48), delay=00:00:00, xdelay=00:00:00, 
mailer=relay, pri=30597, relay=[127.0.0.1] [127.0.0.1], dsn=2.0.0, 
stat=Sent (k317WL0a032728 Message accepted for delivery)
Apr 1 09:32:21 darkstar sendmail[32731]: dangerous permissions=40777 
on queue directory /var/spool/clientmqueue/
Apr 1 09:32:21 darkstar sendmail[32731]: k317WLcE032731: 
Authentication-Warning: www.oinatz.com: apache set sender to yo en oinatz.
com using -f
Apr 1 09:32:21 darkstar sendmail[32725]: k317WL0a032723: to=, delay=00:
00:00, xdelay=00:00:00, mailer=relay, pri=58410, relay=correo.euskaltel.
es. [IP], dsn=2.0.0, stat=Sent (Ok.)
Apr 1 09:32:21 darkstar sendmail[32731]: k317WLcE032731: 
from=yo en oinatz.com, size=594, class=0, nrcpts=1, msgid=<200604010732.
k317WLcE032731@," target="_BLANK">www.oinatz.com>, bodytype=8BITMIME, 
relay=apache en localhost
Apr 1 09:32:21 darkstar sendmail[32733]: k317WL0a032733: from=, 
size=870, class=0, nrcpts=1, msgid=<200604010732.k317WLcE032731@," 
target="_BLANK">www.oinatz.com>, bodytype=8BITMIME, proto=ESMTP, 
daemon=MTA, relay=darkstar.euskaltel.es [127.0.0.1]
Apr 1 09:32:21 darkstar sendmail[32731]: k317WLcE032731: 
to=pepito en terra.es, ctladdr=yo en oinatz.com (48/48), delay=00:00:00, 
xdelay=00:00:00, mailer=relay, pri=30594, relay=[127.0.0.1] 
[127.0.0.1], dsn=2.0.0, stat=Sent (k317WL0a032733 Message accepted for 
delivery)
Apr 1 09:32:21 darkstar sendmail[32730]: k317WL0a032728: to=, delay=00:
00:00, xdelay=00:00:00, mailer=relay, pri=30873, relay=correo.euskaltel.
es. [IP], dsn=2.0.0, stat=Sent (Ok.)
Apr 1 09:32:22 darkstar sendmail[32735]: k317WL0a032733: to=, delay=00:
00:01, xdelay=00:00:01, mailer=relay, pri=30870, relay=correo.euskaltel.
es. [IP], dsn=2.0.0, stat=Sent (Ok.)
Apr 1 09:46:48 darkstar sendmail[4641]: k317JPOK011264: bogus queue 
file, uid=48, gid=51, mode=100660
Apr 1 09:46:48 darkstar sendmail[4641]: k317JPOK011264: Losing .
/qfk317JPOK011264: bogus file uid/gid in mqueue
Apr 1 09:46:48 darkstar sendmail[4641]: k317SXnK024378: bogus queue 
file, uid=48, gid=51, mode=100660
Apr 1 09:46:48 darkstar sendmail[4641]: k317SXnK024378: Losing .
/qfk317SXnK024378: bogus file uid/gid in mqueue
Apr 1 09:46:48 darkstar sendmail[4641]: k317NmCQ017476: bogus queue 
file, uid=48, gid=51, mode=100660
Apr 1 09:46:48 darkstar sendmail[4641]: k317NmCQ017476: Losing .
/qfk317NmCQ017476: bogus file uid/gid in mqueue
Apr 1 09:46:48 darkstar sendmail[4641]: k317MAH9015172: bogus queue 
file, uid=48, gid=51, mode=100660
Apr 1 09:46:48 darkstar sendmail[4641]: k317MAH9015172: Losing .
/qfk317MAH9015172: bogus file uid/gid in mqueue
....

Veo lo de dangerous permissions, no sé si está relacionado:
#ls -l /var/spool
drwxrwxrwx 2 smmsp smmsp 180224 abr 1 09:46 clientmqueue

Parece que parece que están todos (yo no he sido xD). En principio no 
sé si esto puede ser un problema para que se encolen. Diría que no pero 
por si acaso...
Lo de bogus file uid/gid igual está relacionado con ello. Los archivos 
dentro de ese directorio están a:
#ls -l /var/spool/clientmqueue/ls -l *fk317JPOK011264
-rw-rw---- 1 apache smmsp 27888 abr 1 09:20 dfk317JPOK011264
-rw-rw---- 1 apache smmsp 798 abr 1 09:20 Qfk317JPOK011264

No aparece qfk317JPOK011264, los que dice Losing...en el log. Los 
permisos de los archivos encolados vienen dados en
/etc/mail/submit.cf en QueueFileMode 660, equivalente al -rw-rw----

Procesos:
#ps -fea

root 2195 1 0 Mar31 ? 00:00:06 sendmail: accepting connections
smmsp 2204 1 0 Mar31 ? 00:00:00 sendmail: Queue runner en 01:00:00 for 
/var/spool/clientmqueue
...
apache 12355 32055 0 10:13 ? 00:00:01 /opt/apache/bin/httpd -k start -
DSSL
apache 18366 32055 0 10:31 ? 00:00:00 /opt/apache/bin/httpd -k start -
DSSL
apache 18867 32055 0 10:33 ? 00:00:04 [httpd ]
apache 19995 32055 0 10:36 ? 00:00:03 /opt/apache/bin/httpd -k start -
DSSL
apache 21929 32055 0 10:44 ? 00:00:00 /opt/apache/bin/httpd -k start -
DSSL
...

El zombie de apache parece que no está ahora. ¬_¬

Sobre strace, ¿cómo puedo lanzarlo? He hecho:
#strace -p 2204
Process 2204 attached - interrupt to quit
--- SIGSTOP (Stopped (signal)) @ 0 (0) ---
--- SIGSTOP (Stopped (signal)) @ 0 (0) ---
--- SIGALRM (Alarm clock) @ 0 (0) ---
alarm(0) = 0
time(NULL) = 1143881514
getpid() = 2204
time(NULL) = 1143881514
getpid() = 2204
alarm(0) = 0
time(NULL) = 1143881514
sigreturn() = ? (mask now [])
rt_sigaction(SIGTERM, {0x80993e0, [], SA_RESTORER|SA_RESTART, 
0xb7238c08}, {0x80993e0, [], SA_RESTORER|SA_RESTART, 0xb7238c08}, 8) = 
0
rt_sigaction(SIGHUP, {0x80994a0, [], SA_RESTORER|SA_RESTART, 
0xb7238c08}, {0x80994a0, [], SA_RESTORER|SA_RESTART, 0xb7238c08}, 8) = 
0
time([1143881514]) = 1143881514
open("/proc/loadavg", O_RDONLY) = 4
fstat64(4, {st_mode=S_IFREG|0444, st_size=0, ...}) = 0
read(4, "0.38 0.49 0.63 2/201 23948\n", 4096) = 27
close(4) = 0
rt_sigprocmask(SIG_BLOCK, [CHLD], [], 8) = 0
rt_sigaction(SIGCHLD, {0x805f480, [], SA_RESTORER|SA_RESTART, 
0xb7238c08}, {0x805f480, [], SA_RESTORER|SA_RESTART, 0xb7238c08}, 8) = 
0
clone(child_stack=0, 
flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD, 
child_tidptr=0xb714e0c8) = 23949
rt_sigprocmask(SIG_BLOCK, [ALRM], [CHLD], 8) = 0
rt_sigprocmask(SIG_UNBLOCK, [ALRM], [ALRM CHLD], 8) = 0
rt_sigprocmask(SIG_UNBLOCK, [CHLD], [CHLD], 8) = 0
time([1143881514]) = 1143881514
rt_sigprocmask(SIG_BLOCK, [ALRM], [], 8) = 0
time(NULL) = 1143881514
getpid() = 2204
rt_sigaction(SIGALRM, {0x80d7bc0, [], SA_RESTORER|SA_RESTART, 
0xb7238c08}, {0x80d7bc0, [], SA_RESTORER|SA_RESTART, 0xb7238c08}, 8) = 
0
alarm(3600) = 0
rt_sigprocmask(SIG_UNBLOCK, [ALRM], [ALRM], 8) = 0
pause() = ? ERESTARTNOHAND (To be restarted)
--- SIGCHLD (Child exited) @ 0 (0) ---
waitpid(-1, [WIFEXITED(s) && WEXITSTATUS(s) == 0], WNOHANG) = 23949
waitpid(-1, 0xbfff9c94, WNOHANG) = -1 ECHILD (No child processes)
sigreturn() = ? (mask now [])
pause(

No sé si se puede sacar algo de esto

Por último, puedo mandarme un correo desde esa máquina usando sendmail 
y parece que llega al momento.

Eskerrik asko
Oinatz




Prueba el Nuevo Correo Terra; Seguro, Rápido, Fiable.