cat /etc/cron.daily/yum.cron

OK, so there's a -e 0 option that looks like the one to use. But my
/etc/cron.daily/yum.cron already has it:

$ cat /etc/cron.daily/yum.cron
#!/bin/sh

if [ -f /var/lock/subsys/yum ]; then
        /usr/bin/yum -R 10 -e 0 -d 0 -y update yum
        /usr/bin/yum -R 120 -e 0 -d 0 -y update
fi

And I still get the verbose output.

Are you sure that cron job is being used? Did you setup a different
one, perchance?

I can't replicate this on any of my systems.

No; here's the headers of the email I received yesterday morning, from
a machine upgraded from FC2 to FC3 earlier this week. It appears to
point the finger at /etc/cron.daily/yum.cron

Received: (from root@localhost)
	by xy01m005.manchester.xyratex.com (8.13.1/8.13.1/Submit) id
iAB4265C019480;
	Thu, 11 Nov 2004 04:02:06 GMT
Date: Thu, 11 Nov 2004 04:02:06 GMT
Message-Id: <200411110402.iAB4265C019480.xyratex.com>
From: cron (Cron Daemon)
To: me
Subject: Cron <root@xy01m005> run-parts /etc/cron.daily
X-Cron-Env: <SHELL=/bin/bash>
X-Cron-Env: <PATH=/sbin:/bin:/usr/sbin:/usr/bin>
X-Cron-Env: <MAILTO=root>
X-Cron-Env: <HOME=/>
X-Cron-Env: <LOGNAME=root>
X-Cron-Env: <USER=root>

I do have a cron job to do "yum clean packages" but it's clearly not
that. There are no other mentions of yum in any file in
/etc/cron.daily other than the yum.conf from comment #2. The text
"/etc/cron.daily/yum.cron:" appears in the message body, as indicated
in the bug report.

I can't duplicate this at all. Can you post your yum.conf just to make
sure there's not something else going on maybe from there?

Is it still happening for you? can you make it happen again?

I get the identical behaviour on an FC3 system I installed just last
week.  I did a network install (via HTTP) from a local mirror
(www.muug.mb.ca), and every time cron.daily runs
/etc/cron.daily/yum.cron and yum updates some packages, I get this
verbose output.  My yum.cron script is just like the one Paul posted
in comment #2 above, and my /etc/yum.conf file has not been changed
since yum-2.1.11-3 was installed.  I just grabbed the
yum-2.1.11-3.noarch.rpm package from the mirror site and compared the
yum.conf in it to the installed one, and they're identical.  An "rpm
-V yum" also returns nothing.  The yum.conf contains "debuglevel=2",
and a "man yum.conf" says that this is the default debug level, but
one would expect the -d 0 in yum.cron would override this (it did in
yum 2.0.x in FC1 and FC2).

I now have three FC3 systems, all upgraded from FC2, and all of them
exhibit this problem.

yum.conf:

[main]
cachedir=/var/cache/yum
debuglevel=2
logfile=/var/log/yum.log
pkgpolicy=newest
distroverpkg=fedora-release
tolerant=1
exactarch=1
retries=20
obsoletes=1
gpgcheck=1

# PUT YOUR REPOS HERE OR IN separate files named file.repo
# in /etc/yum.repos.d


/etc/yum.repos.d/fedora-updates.repo:

[updates-released]
name=Fedora Core $releasever - $basearch - Released Updates
mirrorlist=http://fedora.redhat.com/download/mirrors/updates-released-fc$releasever
enabled=1
gpgcheck=1

I've also tried using a local repository, with baseurl= instead of the
mirrorlist; same result.

I can reproduce the problem at will by downgrading one of the packages
for which an update has been released and running yum manually with
the same parameters as in the cron job.

If it matters, yum is run using python-2.3.4-11 on FC3.  I just tried
downgrading jwhois to the original 3.2.2-6 package, and then I ran:

yum -e 0 -d 0 -y update 2>&1 | tee /tmp/y.out

which I thought would be a reasonable approximation of what run-parts
does (i.e. send both stdout and stderr through a pipe).  When I then
run "cat -v -t -e /tmp/y.out", I get the following output...

^MUpdating: jwhois 11 % done 1/2^MUpdating: jwhois 38 % done
1/2^MUpdating: jwhois 38 % done 1/2^MUpdating: jwhois 38 % done
1/2^MUpdating: jwhois 39 % done 1/2^MUpdating: jwhois 48 % done
1/2^MUpdating: jwhois 49 % done 1/2^MUpdating: jwhois 51 % done
1/2^MUpdating: jwhois 52 % done 1/2^MUpdating: jwhois 55 % done
1/2^MUpdating: jwhois 59 % done 1/2^MUpdating: jwhois 62 % done
1/2^MUpdating: jwhois 66 % done 1/2^MUpdating: jwhois 70 % done
1/2^MUpdating: jwhois 74 % done 1/2^MUpdating: jwhois 78 % done
1/2^MUpdating: jwhois 82 % done 1/2^MUpdating: jwhois 86 % done
1/2^MUpdating: jwhois 87 % done 1/2^MUpdating: jwhois 91 % done
1/2^MUpdating: jwhois 95 % done 1/2^MUpdating: jwhois 98 % done
1/2^MUpdating: jwhois 99 % done 1/2^MUpdating: jwhois 99 % done
1/2^MUpdating: jwhois 99 % done 1/2^MUpdating: jwhois 100 % done 1/2 $
Completing update for jwhois  - 2/2$

That's a lot of output for a fairly small package, despite the fact
the debuglevel is set below 2 and the output is not a tty, either of
which is supposed to cause the progress display to be suppressed.  I
also tried setting debuglevel=0 explicitly in /etc/yum.conf, and I get
the same results.

okay. I've seen this behavior now and I just applied a patch to cvs to
fix it.

It will be fixed in 2.1.12.

Thanks

Thanks, Seth.  Any chance you could share the patch with us here now?
 I think some of us are eager to get yum to shut up.

not at the momemnt.

The output your seeing is just going to stdout.

You should be able to safely do > /dev/null
and stderr messages will still get outputted.

Just a comment, I run the cron.daily/yum.cron set with -e 0 and -d 1.
 This used to give me a nice summary of updated packages:

/etc/cron.daily/yum.cron:

Updated:  samba-client 3.0.7-2.FC1.i386 samba-common 3.0.7-2.FC1.i386
samba 3.0.7-2.FC1.i386

But no output if nothing was done.  Now I get the verbose progress as
reported above.  I'd like any fix to be able to still allow me to get
the old output.  Simply sending output to /dev/null doesn't do it for me.

Thanks Seth for a great program!

Created attachment 106914 [details]
patch to yum to suppress vebose progress output

This patch is a backport of Seth's fix in CVS, with a few changes to simplify
it a tad.

I disagree that your changes in cli.py simplify it.

Created attachment 107037 [details]
patch to callback.py to make the output more sane.

That patch isn't entirely complete without another one to cli.py

In response to comment number 12, instead on counting on a cron report
for the update you should look in /var/log/yum.log and look for the
update information from there.

Any chance of an FC3 update for this, the yum service is pretty
unusable without it?

I have a yum RPM containing the patch from Comment #13 that fixes the problem
for me. It'll be available at http://www.city-fan.org/ftp/contrib/yum-repo/
until an official update is available.

Joe -- I'm planning to do an update of yum 2.1.12 when it's released as it has a
slew of minor but nice to have fixes in it.

*** Bug 142375 has been marked as a duplicate of this bug. ***