I have the problem again:
# uptime 8:43pm up 5:44, 1 user, load average: 1.08, 1.02, 1.01
# ps ax 738 ? S 0:00 python bin/mailmanctl -n start 739 ? S 0:00 python bin/mailmanctl -n start 740 ? S 0:20 python bin/mailmanctl -n start 741 ? S 0:12 python bin/mailmanctl -n start 742 ? S 0:00 python bin/mailmanctl -n start 743 ? D 288:55 python bin/mailmanctl -n start 744 ? S 0:01 python bin/mailmanctl -n start 745 ? S 0:00 python bin/mailmanctl -n start
# ls locks/ uylug-directiva.lock.guru.743 (and others)
# strace kill 743
execve("/bin/kill", ["kill", "743"], [/* 41 vars */]) = 0 uname({sys="Linux", node="guru", ...}) = 0 brk(0) = 0x804aa20 open("/etc/ld.so.preload", O_RDONLY) = -1 ENOENT (No such file or directory) open("/etc/ld.so.cache", O_RDONLY) = 4 fstat64(4, {st_mode=S_IFREG|0644, st_size=17794, ...}) = 0 old_mmap(NULL, 17794, PROT_READ, MAP_PRIVATE, 4, 0) = 0x40017000 close(4) = 0 open("/lib/libc.so.6", O_RDONLY) = 4 read(4, "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0@\315\1"..., 1024) = 1024 fstat64(4, {st_mode=S_IFREG|0755, st_size=1343073, ...}) = 0 old_mmap(NULL, 1164516, PROT_READ|PROT_EXEC, MAP_PRIVATE, 4, 0) = 0x4001c000 mprotect(0x4012f000, 38116, PROT_NONE) = 0 old_mmap(0x4012f000, 24576, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED, 4, 0x112000) = 0x4012f000 old_mmap(0x40135000, 13540, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x40135000 close(4) = 0 munmap(0x40017000, 17794) = 0 getpid() = 4849 brk(0) = 0x804aa20 brk(0x804aa48) = 0x804aa48 brk(0x804b000) = 0x804b000 open("/usr/share/locale/locale.alias", O_RDONLY) = 4 fstat64(4, {st_mode=S_IFREG|0644, st_size=2567, ...}) = 0 old_mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x40017000 read(4, "# Locale name alias data base.\n#"..., 4096) = 2567 brk(0x804c000) = 0x804c000 read(4, "", 4096) = 0 close(4) = 0 munmap(0x40017000, 4096) = 0 open("/usr/lib/locale/en_US/LC_CTYPE", O_RDONLY) = 4 fstat64(4, {st_mode=S_IFREG|0644, st_size=110304, ...}) = 0 old_mmap(NULL, 110304, PROT_READ, MAP_PRIVATE, 4, 0) = 0x40139000 close(4) = 0 kill(743, SIGTERM) = 0 _exit(0) = ?
# tail logs/qrunner Oct 16 21:11:00 2001 (738) Master qrunner detected subprocess exit (pid: 743, sig: 15, sts: 0, class: ArchRunner, slice: 1/1)
# bin/mailmanctrl restart
.... and some time after restart mailmanctrl
# ps ax 5244 ? S 0:00 python bin/mailmanctl -n start 5245 ? S 0:00 python bin/mailmanctl -n start 5246 ? S 0:00 python bin/mailmanctl -n start 5247 ? S 0:00 python bin/mailmanctl -n start 5248 ? S 0:00 python bin/mailmanctl -n start 5249 ? R 28:34 python bin/mailmanctl -n start 5250 ? S 0:00 python bin/mailmanctl -n start 5251 ? S 0:00 python bin/mailmanctl -n start
Parece joda! (it seems like a joke!)
--
Rodolfo Pilas Quien los puso a estos tipos donde estan, rodolfo@linux.org.uy Quien los deja seguir en su lugar, http://rodolfo.pilas.net Quien los baja ahora de su altar, ICQ #17461636 Quien les paga para que hagan lo que haran http://xtralinux.org -=# Apocalipsis Now % Cuarteto de Nos #=-
Public GnuPG key: http://www.keyserver.net 1024D/57153363 2001-06-02 key fingerprint = DAAE 3246 3F7D A420 B7A0 48A5 D120 C773 5715 3363
"Rodolfo" == Rodolfo Pilas <rodolfo@linux.org.uy> writes:
Rodolfo> I have the problem again:
Rodolfo> load average: 1.08, 1.02, 1.01
Rodolfo> # tail logs/qrunner Oct 16 21:11:00 2001 (738) Master
Rodolfo> qrunner detected subprocess exit (pid: 743, sig: 15, sts:
Rodolfo> 0, class: ArchRunner, slice: 1/1)How big are your .mbox files in the mailman/archives/private/* directories? I had the exact same problem until I turned off internal archiving; once my .mbox files got too big, the ArchRunner took up 100% of the CPU and took hours to run.
Try turning on external archiving and switching to Hypermail. It's a better archiver anyway.
Ben
-- Brought to you by the letters K and Z and the number 7. "Elate means having wings." Debian GNU/Linux maintainer of Gimp and GTK+ -- http://www.debian.org/
En Tue, 2001-10-16 a 21:42, Ben Gertzfield escribio:
"Rodolfo" == Rodolfo Pilas <rodolfo@linux.org.uy> writes:
Rodolfo> I have the problem again: Rodolfo> load average: 1.08, 1.02, 1.01 Rodolfo> # tail logs/qrunner Oct 16 21:11:00 2001 (738) Master Rodolfo> qrunner detected subprocess exit (pid: 743, sig: 15, sts: Rodolfo> 0, class: ArchRunner, slice: 1/1)How big are your .mbox files in the mailman/archives/private/* directories? I had the exact same problem until I turned off internal archiving; once my .mbox files got too big, the ArchRunner took up 100% of the CPU and took hours to run.
It is not big:
# du -h /var/spool/mailman/archives/private/ 9.1M /var/spool/mailman/archives/private
but I am thinking that the problem is perhaps in the pipermail system.
--
Rodolfo Pilas Quien los puso a estos tipos donde estan, rodolfo@linux.org.uy Quien los deja seguir en su lugar, http://rodolfo.pilas.net Quien los baja ahora de su altar, ICQ #17461636 Quien les paga para que hagan lo que haran http://xtralinux.org -=# Apocalipsis Now % Cuarteto de Nos #=-
Public GnuPG key: http://www.keyserver.net 1024D/57153363 2001-06-02 key fingerprint = DAAE 3246 3F7D A420 B7A0 48A5 D120 C773 5715 3363
"Rodolfo" == Rodolfo Pilas <rodolfo@linux.org.uy> writes:
Rodolfo> It is not big:
Rodolfo> # du -h /var/spool/mailman/archives/private/ 9.1M
Rodolfo> /var/spool/mailman/archives/private
Rodolfo> but I am thinking that the problem is perhaps in the
Rodolfo> pipermail system.Try turning off internal archiving (only archive to the mbox file) and see if your problem goes away.
Ben
-- Brought to you by the letters U and J and the number 8. "I don't want the world.. I just want your half." Debian GNU/Linux maintainer of Gimp and GTK+ -- http://www.debian.org/
On Thu, Oct 18, 2001 at 10:42:23AM +0900, Ben Gertzfield wrote:
"Rodolfo" == Rodolfo Pilas <rodolfo@linux.org.uy> writes:
Rodolfo> It is not big: Rodolfo> # du -h /var/spool/mailman/archives/private/ 9.1M Rodolfo> /var/spool/mailman/archives/private Rodolfo> but I am thinking that the problem is perhaps in the Rodolfo> pipermail system.Try turning off internal archiving (only archive to the mbox file) and see if your problem goes away.
I don't know pipermail very well, but I know that it can add one message to an archive, or rebuild a whole archive from an mbox file.
Can it however update an existing archive with whatever messages have been added to a mailbox in the last X hours (i.e. disable HTML archiving in mailman and run arch from cron) If I have to, I could extract a portion of the mbox file if pipermail can't figure out which messages are new in the mbox and not yet in the HTML archive.
Marc
Microsoft is to operating systems & security .... .... what McDonalds is to gourmet cooking
Home page: http://marc.merlins.org/ | Finger marc_f@merlins.org for PGP key
"BG" == Ben Gertzfield <che@debian.org> writes:
BG> How big are your .mbox files in the mailman/archives/private/*
BG> directories? I had the exact same problem until I turned off
BG> internal archiving; once my .mbox files got too big, the
BG> ArchRunner took up 100% of the CPU and took hours to run.
BG> Try turning on external archiving and switching to Hypermail.
BG> It's a better archiver anyway.I can't dispute that there are better archivers than Pipermail, but I'd still like to know why Pipermail/ArchRunner takes up so much cpu, and that only occasionally.
Did you see my timing tests on a list with a 280MB .mbox file? It wasn't nearly that bad.
-Barry
participants (4)
-
barry＠zope.com -
Ben Gertzfield -
Marc MERLIN -
Rodolfo Pilas