[Bug 1155879] New: mandb.service: "find: ‘standard output’: Broken pipe", "find: write error", "status=0/SUCCESS"
http://bugzilla.opensuse.org/show_bug.cgi?id=1155879 Bug ID: 1155879 Summary: mandb.service: "find: ‘standard output’: Broken pipe", "find: write error", "status=0/SUCCESS" Classification: openSUSE Product: openSUSE Distribution Version: Leap 15.1 Hardware: Other OS: Other Status: NEW Severity: Normal Priority: P5 - None Component: Basesystem Assignee: bnc-team-screening@forge.provo.novell.com Reporter: Ulrich.Windl@rz.uni-regensburg.de QA Contact: qa-bugs@suse.de Found By: --- Blocker: --- I noticed this problem in the journal: mandb.service seems to fail for unknown reasons, but "success" is reported: # systemctl status mandb.service ● mandb.service - Do daily mandb update Loaded: loaded (/usr/lib/systemd/system/mandb.service; static; vendor preset: disabled) Active: inactive (dead) since Tue 2019-11-05 09:52:59 CET; 17min ago Docs: man:mandb(8) man:catman(8) Process: 2568 ExecStart=/usr/lib/man-db/do_mandb (code=exited, status=0/SUCCESS) Main PID: 2568 (code=exited, status=0/SUCCESS) Nov 05 09:51:30 pc systemd[1]: Starting Do daily mandb update... Nov 05 09:52:59 pc do_mandb[2568]: find: ‘standard output’: Broken pipe Nov 05 09:52:59 pc do_mandb[2568]: find: write error Nov 05 09:52:59 pc systemd[1]: Started Do daily mandb update. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1155879 http://bugzilla.opensuse.org/show_bug.cgi?id=1155879#c1 --- Comment #1 from Ulrich Windl <Ulrich.Windl@rz.uni-regensburg.de> --- Running /usr/lib/man-db/do_mandb interactively showed no such problem. Grepping for do_mandb in the journal, it shows that the problem started on "Jul 04 08:08:00", while it was last OK on "Jul 02 15:30:06". Seems I upgraded to Leap 15.1 on "Wed 03 Jul"... -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1155879 http://bugzilla.opensuse.org/show_bug.cgi?id=1155879#c3 --- Comment #3 from Ulrich Windl <Ulrich.Windl@rz.uni-regensburg.de> --- (In reply to Dr. Werner Fink from comment #2)
What shell are you using as /bin/sh ... see output of
update-alternatives --config sh
# update-alternatives --config sh update-alternatives: error: no alternatives for sh -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1155879 http://bugzilla.opensuse.org/show_bug.cgi?id=1155879#c5 --- Comment #5 from Ulrich Windl <Ulrich.Windl@rz.uni-regensburg.de> --- (In reply to Dr. Werner Fink from comment #4)
But /bin/sh is linked to /bin/bash or /usr/bin/bash ?
# ll /bin/sh lrwxrwxrwx 1 root root 4 Sep 20 14:14 /bin/sh -> bash -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1155879 http://bugzilla.opensuse.org/show_bug.cgi?id=1155879#c8 --- Comment #8 from Ulrich Windl <Ulrich.Windl@rz.uni-regensburg.de> --- (In reply to Dr. Werner Fink from comment #6)
The line which may cause this is:
find $mp -follow -newer $db | read -t 1 line
that is somehow the read builtin of the bash dies before find is finished here.
IMHO read does not DIE, but it may run into the specified TIMEOUT of one second when the load is a bit high. Effectively the code was not testing whether find returned something, but it timed out find via broken pipe, causing a write error. (In reply to Ulrich Windl from comment #0)
Nov 05 09:52:59 pc do_mandb[2568]: find: ‘standard output’: Broken pipe Nov 05 09:52:59 pc do_mandb[2568]: find: write error
(In reply to Dr. Werner Fink from comment #6)
Nevertheless as this does not hurt I'll decrease severity (I wonder why this is priority "enhancement") Of course not updating the mandb is not serious, but seen from the perspective of the code it was just wrong (and thus is a bug).
-- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1155879 http://bugzilla.opensuse.org/show_bug.cgi?id=1155879#c9 Dr. Werner Fink <werner@suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution|--- |FIXED --- Comment #9 from Dr. Werner Fink <werner@suse.com> --- SR#745706 -- You are receiving this mail because: You are on the CC list for the bug.
participants (1)
-
bugzilla_noreply＠novell.com