Hi Tomas, Tomas Hodek a écrit : > [...] > issue with crond, [...] I was able to catch what it is doing with strace: > > rt_sigprocmask(SIG_BLOCK, [HUP USR1 USR2 PIPE ALRM CHLD TSTP URG VTALRM PROF > [...] > read(5, "root:x:0:0:root:/root:/bin/bash\n"..., 4096) = 4096 > [...] > But problem is within main crond proess: > > PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND > > 437055 root 20 0 67828 63792 2748 R 90.9 0.0 1000:40 cron
You are right, the problem is rather with cron than with its children
processes. Unfortunately, the output of strace reported previously does
not help to discover the bug in cron's source.
> On older hosting server (debian 7 even 8) we had same issue few years ago
> and installation of bcron seemed to fix this issue.
So, we only know that there is some issue which affects cron but does
not affect bcron in the same way. Still missing the clue.
---oOo---
I had just another try today: installing packages daemon and ucspi-unix,
then installing bcron, release = 0.11-18+verbose1, to hunt again the
mysterious exit with code 111.
To my surprise, everything seems to run smoothly; here is some output
from ps:
----------------------------8<---------------------------------
$ ps ax | grep cron
1425461 ? S 0:00 /usr/bin/daemon --noconfig --name bcron-update --
/usr/sbin/bcron-update /etc/crontab /etc/cron.d
1425462 ? S 0:00 bcron-update: /usr/sbin/bcron-update /etc/crontab
/etc/cron.d
1425557 ? S 0:00 /usr/bin/daemon --noconfig --name bcron-spool --
unixserver -u997 -g997 /var/run/bcron-spool bcron-spool
1425558 ? S 0:00 bcron-spool: unixserver -u997 -g997
/var/run/bcron-spool bcron-spool
----------------------------8<---------------------------------
and here, an excerpt of /var/log/syslog:
----------------------------8<---------------------------------
2022-12-12T16:37:35.972519+01:00 georges systemd[1]: Starting LSB: bcron system
scheduler...
2022-12-12T16:37:36.035203+01:00 georges bcron-update[1425451]: Starting update
system crontabs: bcron-update
2022-12-12T16:37:36.035826+01:00 georges systemd[1]: Started LSB: bcron system
scheduler.
2022-12-12T16:37:36.095382+01:00 georges systemd[1]: Reloading.
2022-12-12T16:37:38.193055+01:00 georges systemd[1]: Starting LSB: bcron system
scheduler...
2022-12-12T16:37:38.259372+01:00 georges bcron-sched[1425498]: Starting bcron
system scheduler: bcron-start
2022-12-12T16:37:38.260290+01:00 georges systemd[1]: Started LSB: bcron system
scheduler.
2022-12-12T16:37:38.266481+01:00 georges daemon: bcron-start: client (pid
1425506) exited with 111 status, exiting
2022-12-12T16:37:38.314002+01:00 georges systemd[1]: Reloading.
2022-12-12T16:37:40.454895+01:00 georges systemd[1]: Starting LSB: manage user
crontab submission...
2022-12-12T16:37:40.529314+01:00 georges bcron-spool[1425544]: Starting manage
user crontab submission: bcron-spool
2022-12-12T16:37:40.529961+01:00 georges systemd[1]: Started LSB: manage user
crontab submission.
----------------------------8<---------------------------------
So I am uploading the package bcron_0.11-18+verbose1.deb to Debian.
Please can you give a stress test to this package, when it appears in
debian/unstable? (at https://packages.debian.org/unstable/bcron)
Please upgrade also the package daemon to release 0.8-1.
If their behavior is satisfactory, I shall push a backport of bcron to
bullseye-backports, and backport also the package daemon: this package
has been upgraded recently (less than two weeks ago), while ucspi-unix
has remained unchanged since year 2018.
Let us hope that such backports can fix the bug report.
Best regards, Georges.
signature.asc
Description: PGP signature
