Debian Bug report logs - #921557
start-stop-daemon: behavior change on "matching only on non-root pidfile /run/exim4/exim.pid is insecure" not fully documented

Toggle useless messages

Message #5 received at submit@bugs.debian.org (full text, mbox, reply):

From: Andreas Metzler <ametzler@bebt.de>

To: Debian Bug Tracking System <submit@bugs.debian.org>

Subject: start-stop-daemon: behavior change on "matching only on non-root pidfile /run/exim4/exim.pid is insecure" not fully documented

Date: Wed, 6 Feb 2019 19:58:52 +0100

Package: dpkg
Version: 1.19.3
Severity: important

With 1.19.3 the following command stopped working:
/sbin/start-stop-daemon --stop --retry 5 --quiet --pidfile /run/exim4/exim.pid
/sbin/start-stop-daemon: matching only on non-root pidfile /run/exim4/exim.pid is insecure

Afaict this broke exim #921326, amavisd-new #921016 and mldonkey-server
#920466.

dpkg's changelog.Debian says:
 * start-stop-daemon: Check whether standalone --pidfile use is secure.
    Prompted by Michael Orlitzky <michael@orlitzky.com>.

the regular changelog is more verbose:
------------------------
commit bc9736f6feae7625cc5ec063ea1b27d51a5f9317
Author: Guillem Jover <guillem@debian.org>
Date:   Sat Sep 22 12:12:05 2018 +0200

    s-s-d: Check whether standalone --pidfile use is secure

    If we are only matching on the pidfile, which is owned by a non-root
    user, and we are running as a root user then this is a security risk,
    and the contents cannot be trusted, because the daemon might have been
    compromised which would allow modifying the pid within.

    If we are then calling start-stop-daemon as a privileged user, that
    would allow acting on any PID in the system.

    Prompted-by: Michael Orlitzky <michael@orlitzky.com>
    Ref: https://redmine.kannel.org/issues/771
------------------------

However the manpage was not updated. Could you please describe which
restrictions were added, what behavior I can rely on to work?

For further entertainment exim does not use start-stop-daemon directly
but uses lsb, which seems to translate
killproc -p /run/exim4/exim.pid /usr/sbin/exim4
to
/sbin/start-stop-daemon --stop --retry 5 --quiet --pidfile /run/exim4/exim.pid

dropping the daemon name somewhere.

I would appreciate if you could agree to keep this dpkg update put of
testing a little bit to be able to solve this.

cu Andreas

Message #10 received at 921557@bugs.debian.org (full text, mbox, reply):

From: Guillem Jover <guillem@debian.org>

To: Andreas Metzler <ametzler@bebt.de>, 921557@bugs.debian.org

Subject: Re: Bug#921557: start-stop-daemon: behavior change on "matching only on non-root pidfile /run/exim4/exim.pid is insecure" not fully documented

Date: Thu, 7 Feb 2019 09:30:27 +0100

Hi!

On Wed, 2019-02-06 at 19:58:52 +0100, Andreas Metzler wrote:
> Package: dpkg
> Version: 1.19.3
> Severity: important

> With 1.19.3 the following command stopped working:
> /sbin/start-stop-daemon --stop --retry 5 --quiet --pidfile /run/exim4/exim.pid
> /sbin/start-stop-daemon: matching only on non-root pidfile /run/exim4/exim.pid is insecure
> 
> Afaict this broke exim #921326, amavisd-new #921016 and mldonkey-server
> #920466.

Ouch. :/

> dpkg's changelog.Debian says:
>  * start-stop-daemon: Check whether standalone --pidfile use is secure.
>     Prompted by Michael Orlitzky <michael@orlitzky.com>.
> 
> the regular changelog is more verbose:
> ------------------------
> commit bc9736f6feae7625cc5ec063ea1b27d51a5f9317
> Author: Guillem Jover <guillem@debian.org>
> Date:   Sat Sep 22 12:12:05 2018 +0200
> 
>     s-s-d: Check whether standalone --pidfile use is secure
> 
>     If we are only matching on the pidfile, which is owned by a non-root
>     user, and we are running as a root user then this is a security risk,
>     and the contents cannot be trusted, because the daemon might have been
>     compromised which would allow modifying the pid within.
> 
>     If we are then calling start-stop-daemon as a privileged user, that
>     would allow acting on any PID in the system.
> 
>     Prompted-by: Michael Orlitzky <michael@orlitzky.com>
>     Ref: https://redmine.kannel.org/issues/771
> ------------------------
> 
> However the manpage was not updated. Could you please describe which
> restrictions were added, what behavior I can rely on to work?

The man page had been updated in a previous release (1.19.1):

,---
commit 7afd25e72f447a4a4b130f38bba28ff6661cbb6c
Author: Guillem Jover <guillem@debian.org>
Date:   Fri Sep 14 21:35:16 2018 +0200

    man: Add a warning in s-s-d(8) about using --pidfile alone with non-root files

    Prompted-by: Michael Orlitzky <michael@orlitzky.com>
    Ref: https://redmine.kannel.org/issues/771

diff --git a/debian/changelog b/debian/changelog
index ef297822c..2d2cf974c 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -204,6 +204,8 @@ dpkg (1.19.1) UNRELEASED; urgency=medium
     - Fix man page markup. Closes: #900033, #900035, #900040
       Thanks to Bjarni Ingi Gislason <bjarniig@rhi.hi.is>.
     - Fix Doxygen comment for libdpkg dpkg_arch_find() function.
+    - Document the dangers of using start-stop-daemon(8) only with --pidfile
+      as matching option with the pid file owned by a non-privileged user.
   * Code internals:
     - Do not use stringy eval to define different sub implementations,
       just assign an anonymous sub to the typeglob.
diff --git a/man/start-stop-daemon.man b/man/start-stop-daemon.man
index cb01cff19..2391d5e9f 100644
--- a/man/start-stop-daemon.man
+++ b/man/start-stop-daemon.man
@@ -116,9 +116,17 @@ Check for a process with the specified parent pid \fIppid\fP
 The \fIppid\fP must be a number greater than 0.
 .TP
 .BR \-p ", " \-\-pidfile " \fIpid-file\fP"
-Check whether a process has created the file \fIpid-file\fP. Note: using this
-matching option alone might cause unintended processes to be acted on, if the
-old process terminated without being able to remove the \fIpid-file\fP.
+Check whether a process has created the file \fIpid-file\fP.
+.IP
+Note: using this matching option alone might cause unintended processes to
+be acted on, if the old process terminated without being able to remove the
+\fIpid-file\fP.
+.IP
+\fBWarning:\fP Using this match option alone with a daemon that writes the
+pidfile as an unprivileged user is a security risk, because if the daemon
+gets compromised the contents of the pidfile cannot be trusted, and then
+a privileged runner (such as an init script executed as root) would end up
+acting on any system process.
 .TP
 .BR \-x ", " \-\-exec " \fIexecutable\fP"
 Check for processes that are instances of this \fIexecutable\fP. The
`---

Is that not enough?

> For further entertainment exim does not use start-stop-daemon directly
> but uses lsb, which seems to translate
> killproc -p /run/exim4/exim.pid /usr/sbin/exim4
> to
> /sbin/start-stop-daemon --stop --retry 5 --quiet --pidfile /run/exim4/exim.pid
> 
> dropping the daemon name somewhere.

Hrrm, given that lsb is pretty much abandoned, and the maintainer has
stated in the past little interest in touching the sysvinit script
parts, it might be quickest perhaps to switch to use s-s-d directly?

> I would appreciate if you could agree to keep this dpkg update put of
> testing a little bit to be able to solve this.

I was planning a new upload during this week (the version in sid, will
not migrate as is due to a regression in dgit's autopkgtests), but can
postpone it a few days until you've uploaded exim.

Thanks,
Guillem

Message #15 received at 921557@bugs.debian.org (full text, mbox, reply):

From: Andreas Metzler <ametzler@bebt.de>

To: 921557@bugs.debian.org

Subject: Re: Bug#921557: start-stop-daemon: behavior change on "matching only on non-root pidfile /run/exim4/exim.pid is insecure" not fully documented

Date: Thu, 7 Feb 2019 18:53:18 +0100

[Message part 1 (text/plain, inline)]

On 2019-02-07 Guillem Jover <guillem@debian.org> wrote:
> On Wed, 2019-02-06 at 19:58:52 +0100, Andreas Metzler wrote:
[...]
>> dpkg's changelog.Debian says:
>>  * start-stop-daemon: Check whether standalone --pidfile use is secure.
>>     Prompted by Michael Orlitzky <michael@orlitzky.com>.

>> the regular changelog is more verbose:
[...] 
>> However the manpage was not updated. Could you please describe which
>> restrictions were added, what behavior I can rely on to work?

> The man page had been updated in a previous release (1.19.1):

> ,---
> commit 7afd25e72f447a4a4b130f38bba28ff6661cbb6c
> Author: Guillem Jover <guillem@debian.org>
> Date:   Fri Sep 14 21:35:16 2018 +0200

>     man: Add a warning in s-s-d(8) about using --pidfile alone with non-root files
[...]
> Is that not enough?

Hello,

Imho no. It warns that this is dangerous, while it actually *was*
dangerous up to 1.19.1 and *is* non-functional now, since s-s-d will
refuse to act and will not kill the process but exit with an error
instead.

There should be something like "If -pidfile is not used in combination
with x, y, z and the pidfile is not owned by root than ssd (version
1.19.2 and later) will exit with an error instead of kiling the process."

I diffed 1.19.1 and 1.19.3 manpages and searched in vain for the
description of a behavior change. :-(

>> For further entertainment exim does not use start-stop-daemon directly
>> but uses lsb, which seems to translate
>> killproc -p /run/exim4/exim.pid /usr/sbin/exim4
>> to
>> /sbin/start-stop-daemon --stop --retry 5 --quiet --pidfile /run/exim4/exim.pid

>> dropping the daemon name somewhere.

> Hrrm, given that lsb is pretty much abandoned, and the maintainer has
> stated in the past little interest in touching the sysvinit script
> parts, it might be quickest perhaps to switch to use s-s-d directly?

I have done that.

>> I would appreciate if you could agree to keep this dpkg update put of
>> testing a little bit to be able to solve this.

> I was planning a new upload during this week (the version in sid, will
> not migrate as is due to a regression in dgit's autopkgtests), but can
> postpone it a few days until you've uploaded exim.

Thanks for the offer, I have just uploaded a fix for exim, so it will
not be necessary.

Thanks for maintaining dpkg, BTW!

cu Andreas
-- 
`What a good friend you are to him, Dr. Maturin. His other friends are
so grateful to you.'
`I sew his ears on from time to time, sure'

[signature.asc (application/pgp-signature, inline)]

Message #20 received at 921557@bugs.debian.org (full text, mbox, reply):

From: Brian May <bam@debian.org>

To: Andreas Metzler <ametzler@bebt.de>, 921557@bugs.debian.org

Subject: Re: Bug#921557: start-stop-daemon: behavior change on "matching only on non-root pidfile /run/exim4/exim.pid is insecure" not fully documented

Date: Tue, 12 Feb 2019 17:33:36 +1100

On Thu, Feb 07, 2019 at 06:53:18PM +0100, Andreas Metzler wrote:
> > I was planning a new upload during this week (the version in sid, will
> > not migrate as is due to a regression in dgit's autopkgtests), but can
> > postpone it a few days until you've uploaded exim.
> 
> Thanks for the offer, I have just uploaded a fix for exim, so it will
> not be necessary.

How did you fix the problem for exim?

I have a suspicion that amavisd-new drops privileges immediately (isn't
this considered good behaviour?) so cannot write the pid file as root.

As a result, I don't know how to fix #921016.
-- 
Brian May <bam@debian.org>

Message #25 received at 921557@bugs.debian.org (full text, mbox, reply):

From: Andreas Metzler <ametzler@bebt.de>

To: 921557@bugs.debian.org

Subject: Re: Bug#921557: start-stop-daemon: behavior change on "matching only on non-root pidfile /run/exim4/exim.pid is insecure" not fully documented

Date: Tue, 12 Feb 2019 07:59:48 +0100

On 2019-02-12 Brian May <bam@debian.org> wrote:
> On Thu, Feb 07, 2019 at 06:53:18PM +0100, Andreas Metzler wrote:
>>> I was planning a new upload during this week (the version in sid, will
>>> not migrate as is due to a regression in dgit's autopkgtests), but can
>>> postpone it a few days until you've uploaded exim.

>> Thanks for the offer, I have just uploaded a fix for exim, so it will
>> not be necessary.

> How did you fix the problem for exim?

> I have a suspicion that amavisd-new drops privileges immediately (isn't
> this considered good behaviour?) so cannot write the pid file as root.

> As a result, I don't know how to fix #921016.

Hello,
You will need to limit the processes that s-s-d in considering for
killing, otherwise the amasvis user could kill arbitrary processes by
listing them in the pid file. For exim I am now passing --exec
/path/to/daemon as argument to s-s-d

https://salsa.debian.org/exim-team/exim4/commit/7d69adc64e5006ce1d033acbd51681dc8aa3640d

Alternatively (or additionally) setting --user should also work.

hth, cu Andreas
-- 
`What a good friend you are to him, Dr. Maturin. His other friends are
so grateful to you.'
`I sew his ears on from time to time, sure'

Message #30 received at 921557@bugs.debian.org (full text, mbox, reply):

From: Brian May <bam@debian.org>

To: Andreas Metzler <ametzler@bebt.de>, 921557@bugs.debian.org

Subject: Re: Bug#921557: start-stop-daemon: behavior change on "matching only on non-root pidfile /run/exim4/exim.pid is insecure" not fully documented

Date: Tue, 12 Feb 2019 21:03:01 +1100

Andreas Metzler <ametzler@bebt.de> writes:

> You will need to limit the processes that s-s-d in considering for
> killing, otherwise the amasvis user could kill arbitrary processes by
> listing them in the pid file. For exim I am now passing --exec
> /path/to/daemon as argument to s-s-d
>
> https://salsa.debian.org/exim-team/exim4/commit/7d69adc64e5006ce1d033acbd51681dc8aa3640d
>
> Alternatively (or additionally) setting --user should also work.

OK, thanks for this. I hadn't realized you could resolve this by passing
--exec and/or --user.

Hmmm. Presumably --exec may not be so useful for a Perl program - I am
guessing I would need to point to the Perl interpreter, so probably
should use --user too.
-- 
Brian May <bam@debian.org>

Message #35 received at 921557@bugs.debian.org (full text, mbox, reply):

From: Andreas Metzler <ametzler@bebt.de>

To: 921557@bugs.debian.org

Subject: Re: Bug#921557: start-stop-daemon: behavior change on "matching only on non-root pidfile /run/exim4/exim.pid is insecure" not fully documented

Date: Tue, 12 Feb 2019 17:51:00 +0100

On 2019-02-12 Brian May <bam@debian.org> wrote:
> Andreas Metzler <ametzler@bebt.de> writes:
[...]
>> For exim I am now passing --exec
>> /path/to/daemon as argument to s-s-d
[...]
>> Alternatively (or additionally) setting --user should also work.

> OK, thanks for this. I hadn't realized you could resolve this by passing
> --exec and/or --user.

> Hmmm. Presumably --exec may not be so useful for a Perl program - I am
> guessing I would need to point to the Perl interpreter, so probably
> should use --user too.

Ah yes. --name should work with a perl program, too.

cu Andreas
-- 
`What a good friend you are to him, Dr. Maturin. His other friends are
so grateful to you.'
`I sew his ears on from time to time, sure'

Message #38 received at 921557-submitter@bugs.debian.org (full text, mbox, reply):

From: Guillem Jover <guillem@debian.org>

To: 921557-submitter@bugs.debian.org

Subject: Bug#921557 in package dpkg marked as pending

Date: Sat, 23 Feb 2019 18:34:21 +0100

Control: tag 921557 pending

Hi!

Bug #921557 in package dpkg reported by you has been fixed in
the dpkg/dpkg.git Git repository. You can see the changelog below, and
you can check the diff of the fix at:

    https://git.dpkg.org/cgit/dpkg/dpkg.git/diff/?id=754511aae

---
commit 754511aaef556ebad03560e689eeb2044a7ee1fd
Author: Guillem Jover <guillem@debian.org>
Date:   Fri Feb 15 05:09:11 2019 +0100

    s-s-d(1): Document behavior of --pidfile security checks
    
    Describe what to expect from these new checks.
    
    Closes: #921557

diff --git a/debian/changelog b/debian/changelog
index f9de3194c..36d296541 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -8,6 +8,9 @@ dpkg (1.19.5) UNRELEASED; urgency=medium
     use errno for error reporting or not.
   * start-stop-daemon: Always refuse to parse a world-writable pidfile,
     except when that is /dev/null.
+  * Documentation:
+    - start-stop-daemon(1): Document behavior of --pidfile security checks.
+      Closes: #921557
   * Build system:
     - Check whether this dist is a release, based only on the version format.
       This will avoid having to do a two staged release to get a proper perl

Message #45 received at 921557-close@bugs.debian.org (full text, mbox, reply):

From: Guillem Jover <guillem@debian.org>

To: 921557-close@bugs.debian.org

Subject: Bug#921557: fixed in dpkg 1.19.5

Date: Sat, 23 Feb 2019 17:49:13 +0000

Source: dpkg
Source-Version: 1.19.5

We believe that the bug you reported is fixed in the latest version of
dpkg, which is due to be installed in the Debian FTP archive.

A summary of the changes between this version and the previous one is
attached.

Thank you for reporting the bug, which will now be closed.  If you
have further comments please address them to 921557@bugs.debian.org,
and the maintainer will reopen the bug report if appropriate.

Debian distribution maintenance software
pp.
Guillem Jover <guillem@debian.org> (supplier of updated dpkg package)

(This message was generated automatically at their request; if you
believe that there is a problem with it please contact the archive
administrators by mailing ftpmaster@ftp-master.debian.org)


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512

Format: 1.8
Date: Sat, 23 Feb 2019 18:00:54 +0100
Source: dpkg
Architecture: source
Version: 1.19.5
Distribution: unstable
Urgency: medium
Maintainer: Dpkg Developers <debian-dpkg@lists.debian.org>
Changed-By: Guillem Jover <guillem@debian.org>
Closes: 666147 883700 920880 920905 920972 920974 921031 921557 921942 921943 922039 922212 922407 922410 922799
Changes:
 dpkg (1.19.5) unstable; urgency=medium
 .
   [ Guillem Jover ]
   * start-stop-daemon: Make sure that we get a meaningful errno on
     parse_unsigned(), so that the error messages always make sense.
   * start-stop-daemon: Add new fatalv() and fatale() functions and use the
     latter for system errors, so that we are explicit on whether we want to
     use errno for error reporting or not.
   * start-stop-daemon: Always refuse to parse a world-writable pidfile,
     except when that is /dev/null.
   * dpkg: Print the current set of enabled force options on --force-help.
   * dpkg: Parse and set new DPKG_FORCE environment variable for subprocesses.
     Closes: #666147
   * dpkg-statoverride: Add support for --force-* options.
   * dpkg-statoverride: Switch from --force option to new --force-<thing>
     options. Deprecate --force option which will be considered an alias for
     --force-all for now.
   * dpkg, dpkg-statoverride: Add new option --refuse-security-mac to control
     SELinux. See #811037.
   * dpkg: Clarify error on unknown system user/group in statoverride database.
     Closes: #920880
   * dpkg-buildpackage: Clarify the warning/error on short OpenPGP key IDs.
     Closes: #922039
   * dpkg-maintscript-helper: Restrict find for dir-to-symlink move to
     -maxdepth 1. Thanks to Ralf Treinen <treinen@free.fr>. Closes: #922799
   * dpkg-genbuildinfo: Add support for a new Build-Tainted-By field in
     .buildinfo files. Suggested by Alexander E. Patrakov <patrakov@gmail.com>.
   * libdpkg: Clarify field names in error and warning messages.
   * libdpkg: Optimize error handling. Move the error reporting outside the
     involved functions so that we do not need to call gettext if there is no
     error, which has a significant performance cost.
   * libdpkg: Merge nicknames table into fieldinfos, to stop penalizing the
     lookup of non-obsolete fieldnames.
   * libdpkg: Print a more accurate warning for Revision nicknames.
   * libdpkg: Increase the pkg-hash bins size to 65521, to improve hash table
     performance at the cost of a bit more memory usage.
   * libdpkg: Blank packages that are not-installed with unknown selection.
     Closes: #922410
   * libdpkg, dpkg: Print the dpkg database directory on access errors.
     Closes: #883700
   * Perl modules:
     - Dpkg::Vendor::Debian: Add support for merged-usr-via-symlinks tainted
       tag. Suggested by Alexander E. Patrakov <patrakov@gmail.com>.
     - Dpkg::Vendor::Debian: Add support for usr-local-has-* tainted tags.
     - Dpkg::Source::Package: Add a missing use Dpkg::Source::Format.
       Diagnosed by Ian Jackson <ijackson@chiark.greenend.org.uk>.
       Closes: #921031
   * Documentation:
     - start-stop-daemon(1): Document behavior of --pidfile security checks.
       Closes: #921557
     - dpkg(1): Document the unknown selection state. Closes: #922407
   * Code internals:
     - dpkg: Move SELinux fallback label to the SELinux specific code path.
     - dpkg: Simplify maintscript_set_exec_context().
     - dpkg: Move force options support into its own file.
     - dpkg: Do not hardcode the program name in the --force-help output.
     - dpkg: Switch force options from individual variables to bit fields.
     - dpkg: Switch from a char to an enum to track the force options types.
     - dpkg: Switch to set the default force option from the forceinfos array.
     - libdpkg: New benchmark programs and infrastructure.
     - libdpkg: Add new dpkg_error_move() function.
     - libdpkg: Add new dpkg_has_error() function.
     - libdpkg: Move status names from parse errors to arguments.
     - libdpkg: Use va_arg copy instead of the original on a vasprintf() call.
     - libdpkg: Include <string.h> in pager.c.
       Reported by Y <sevener@cock.li>. Closes: #922212
     - libdpkg: Use pkg_set_want() instead of a direct assignment.
   * Build system:
     - Check whether this dist is a release, based only on the version format.
       This will avoid having to do a two staged release to get a proper perl
       distribution tarball.
   * Packaging:
     - autopkgtest: Add file to Depends fields.
     - autopkgtest: Clarify behavior on root/non-root requirement.
     - Bump Standards-Version to 4.3.0 (no changes required).
     - Remove now unused assert usertag description.
     - Remove trailing whitespace from changelog.
     - Remove Origin and Bugs fields from control file.
     - Include a bug-script to report on tainted merged-usr-via-symlinks.
   * Test suite:
     - libdpkg: Fix unit test for file_slurp(). Closes: #920974
       Diagnosed by Frank Schaefer <kelledin@gmail.com>.
 .
   [ Updated programs translations ]
   * Dutch (Frans Spiesschaert). Closes: #921942
   * German (Sven Joachim).
   * Simplified Chinese (Zhou Mo). Closes: #920972
 .
   [ Updated scripts translations ]
   * German (Helge Kreutzmann).
 .
   [ Updated man pages translations ]
   * Dutch (Frans Spiesschaert). Closes: #921943
   * French (Jean-Pierre Giraud). Closes: #920905
   * German (Helge Kreutzmann).
Checksums-Sha1:
 c93c51dab02db4617f7ec02bc8b42a4f11fb1424 2103 dpkg_1.19.5.dsc
 fdb7768844d6f48a1ca7a76455303018d9532bf3 4699096 dpkg_1.19.5.tar.xz
 5f0ab39ac04b23f54f2529b6042fb3c6d00d9980 7180 dpkg_1.19.5_amd64.buildinfo
Checksums-Sha256:
 a799548f5b991d6b3701d4fea798811aa50cef9c6e87a89a14522e2a206d39e1 2103 dpkg_1.19.5.dsc
 ed327c9973610a695e1950613861c734974d9476e76ef713724558601ce63544 4699096 dpkg_1.19.5.tar.xz
 c11fb3eed26354384d45a26a31f4f00640617ffaa25ad159422c7e5a142a8c0d 7180 dpkg_1.19.5_amd64.buildinfo
Files:
 fc9d8e7c820ddb1e99be56403123da6e 2103 admin required dpkg_1.19.5.dsc
 74b2dc1265730c6f9bea614c64584f4d 4699096 admin required dpkg_1.19.5.tar.xz
 b708a0f6cc11e8db8736831a65fe6ace 7180 admin required dpkg_1.19.5_amd64.buildinfo

-----BEGIN PGP SIGNATURE-----
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=/hnh
-----END PGP SIGNATURE-----

Send a report that this bug log contains spam.