John A. Martin wrote:
> With regard to the \0 please see 'man sysklogd' from sysklogd-1.4.1-17
> which, as of this writing, is in stable and testing.  The 1.4.1-17.1
> in unstable does not appear to touch the present issue.
>
> ,----
> INSTALLATION CONCERNS
>        There is probably one important consideration when installing
> this ver- sion of syslogd.  This version of syslogd is dependent on 
> proper  for- matting  of  messages  by  the syslog function.  The
> functioning of the syslog function in the shared libraries changed
> somewhere in the region of  libc.so.4.[2-4].n.   The  specific change
> was to null-terminate the message before transmitting it to the /dev/log
>  socket.   Proper  func- tioning  of this version of syslogd is
> dependent on null-termination of the message.
>
>        This problem will typically manifest itself if  old  statically 
> linked binaries  are being used on the system.  Binaries using old
> versions of the syslog function will cause empty lines to be logged
> followed by the message  with  the  first  character in the message
> removed.  Relinking these binaries to newer versions of the shared
> libraries  will  correct this problem.
> `----
>
> FWIW Bind9-9.3.2-2 nameserver logging suffers the same problems as
> libmail-spf-query-perl when using syslog-ng with unix-stream input.
>
> Has libc changed in this respect?

Recently?  I doubt it, but I really don't know for sure.

> If [libc hasn't changed in this respect,] why should Sys::Syslog change?

That is a very good question that Brendan O'Dea <[EMAIL PROTECTED]> (the 
maintainer of the Perl Debian packages) and Wojtek Rzasa 
<[EMAIL PROTECTED]> (the reporter of bug #345157[1]) should be asked.

> If Sys::Syslog restored the \0 would syslog-ng be happy with 
> unix-stream?  Who would be unhappy?

I have no idea.

> You guys are package maintainers, I am not.  Though I think Bug
> #356700 should be reassigned yet again, I am not sure it is my place
> to do the reassignment.

It is you who reported the bug.  This issue is still not resolved, and if, 
assuming it hadn't been reported yet, you felt the need to report it, then 
you should by all means feel free to reassign it yourself to whatever 
package you believe it is rooted in (unless of course some package 
maintainer has already accepted the issue to be in their jurisdiction).

Julian.

References:
 1. http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=345157

Attachment: pgp1pNMkB9J7X.pgp
Description: PGP signature

Reply via email to