Hi guys,

Is anyone using bindgraph on Etch? I have it installed (from the Debian
repos) on two Etch boxes. Took some fiddling, since the default install
of bind doesn't log queries, but that sorted, I now have the daemon
making the rrd file, but the bindgraph.cgi doesn't seem to work
correctly. It loads the page, but no images.

It does the same thing on both machines. Has anyone else had trouble
with this? From what I can see the code is very similar to with
mailgraph.cgi and queuegraph.cgi and they both work fine.

Is there a way to do a trace/debug of a cgi script? It's not logging
anything at all.

Thanks


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

Le Friday 27 July 2007 14:06:08 koffiejunkie, vous avez écrit:
> Hi guys,
>
> Is anyone using bindgraph on Etch? I have it installed (from the Debian
> repos) on two Etch boxes. Took some fiddling, since the default install
> of bind doesn't log queries, but that sorted, I now have the daemon
> making the rrd file, but the bindgraph.cgi doesn't seem to work
> correctly. It loads the page, but no images.
>
> It does the same thing on both machines. Has anyone else had trouble
> with this? From what I can see the code is very similar to with
> mailgraph.cgi and queuegraph.cgi and they both work fine.
>
> Is there a way to do a trace/debug of a cgi script? It's not logging
> anything at all.
>
> Thanks

Strange !
I just installed it today, and I found what is the problem.
It is not compatible with rrdtool 1.2.
I found it trying to display directly the image (look in the HTMl source).
RRD says it doesn't understand the data format...

I backported the testing/unstable version 0.2a.
I saw this bug in BTS which was corrected.

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQBGqkipDltnDmLJYdARAlBmAJ9nlsRdW+YV5uj3+eDFQf WzasYnEwCgrN2D
iAy03x25QcwRQ6W48Dv0M1c=
=0bC3
-----END PGP SIGNATURE-----

Gilles Mocellin wrote:
> Strange !
> I just installed it today, and I found what is the problem.
> It is not compatible with rrdtool 1.2.
> I found it trying to display directly the image (look in the HTMl source).
> RRD says it doesn't understand the data format...
>
> I backported the testing/unstable version 0.2a.
> I saw this bug in BTS which was corrected.

Hi Gilles,

Thanks for this. I looked at the code for a long time, but it's greek
to me - my perl is limited to "hello" world...

I'm just on the way out for the weekend, will look into this again
tomorrow night. As far as I remember I don't even get the images
generated (that is triggered by the .cgi, isn't it?).

# ./bindgraph.cgi
Use of uninitialized value in substitution (s///) at
/usr/share/perl/5.8/File/Basename.pm line 338.
fileparse(): need a valid pathname at ./bindgraph.cgi line 181

When I strace it, I get this snipped some of the "file not found" lines:

stat64("/usr/lib/perl/5.8/re.pm", {st_mode=S_IFREG|0644, st_size=1340,
....}) = 0
open("/usr/lib/perl/5.8/re.pm", O_RDONLY|O_LARGEFILE) = 5
ioctl(5, SNDCTL_TMR_TIMEBASE or TCGETS, 0xbfffec48) = -1 ENOTTY
(Inappropriate ioctl for device)
_llseek(5, 0, [0], SEEK_CUR) = 0
read(5, "package re;\n\nour $VERSION = 0.05"..., 4096) = 1340
read(5, "", 4096) = 0
close(5) = 0
read(4, "[:>\\]])?(.*)/s);\n $dirpath ||"..., 4096) = 4096
brk(0x8235000) = 0x8235000
read(4, ".\'. fileparse() would return \'f"..., 4096) = 3069
read(4, "", 4096) = 0
close(4) = 0
read(3, "t_date {\n\tmy ($when) = @_;\n\n\tmy "..., 4096) = 2195
read(3, "", 4096) = 0
close(3) = 0
uname({sys="Linux", node="rimwards", ...}) = 0
write(2, "Use of uninitialized value in su"..., 100Use of uninitialized
value in substitution (s///) at /usr/share/perl/5.8/File/Basename.pm
line 338.
) = 100
[snip]
stat64("/usr/share/perl/5.8/Carp/Heavy.pm", {st_mode=S_IFREG|0644,
st_size=5747, ...}) = 0
open("/usr/share/perl/5.8/Carp/Heavy.pm", O_RDONLY|O_LARGEFILE) = 3
ioctl(3, SNDCTL_TMR_TIMEBASE or TCGETS, 0xbffff598) = -1 ENOTTY
(Inappropriate ioctl for device)
_llseek(3, 0, [0], SEEK_CUR) = 0
read(3, "# Carp::Heavy uses some variable"..., 4096) = 4096
brk(0x8256000) = 0x8256000
read(3, "s($called, $caller, $cache);\n "..., 4096) = 1651
read(3, "", 4096) = 0
close(3) = 0
write(2, "fileparse(): need a valid pathna"..., 63fileparse(): need a
valid pathname at ./bindgraph.cgi line 181
) = 63
exit_group(255) = ?
Process 28639 detached


Is this what you got too?


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

Le Saturday 28 July 2007 08:49:50 koffiejunkie, vous avez écrit:
> Gilles Mocellin wrote:
> > Strange !
> > I just installed it today, and I found what is the problem.
> > It is not compatible with rrdtool 1.2.
> > I found it trying to display directly the image (look in the HTMl
> > source). RRD says it doesn't understand the data format...
> >
> > I backported the testing/unstable version 0.2a.
> > I saw this bug in BTS which was corrected.
>
> Hi Gilles,
>
> Thanks for this. I looked at the code for a long time, but it's greek
> to me - my perl is limited to "hello" world...
>
> I'm just on the way out for the weekend, will look into this again
> tomorrow night. As far as I remember I don't even get the images
> generated (that is triggered by the .cgi, isn't it?).
>
> # ./bindgraph.cgi
> Use of uninitialized value in substitution (s///) at
> /usr/share/perl/5.8/File/Basename.pm line 338.
> fileparse(): need a valid pathname at ./bindgraph.cgi line 181
>
> When I strace it, I get this snipped some of the "file not found" lines:
>
> stat64("/usr/lib/perl/5.8/re.pm", {st_mode=S_IFREG|0644, st_size=1340,
> ...}) = 0
> open("/usr/lib/perl/5.8/re.pm", O_RDONLY|O_LARGEFILE) = 5
> ioctl(5, SNDCTL_TMR_TIMEBASE or TCGETS, 0xbfffec48) = -1 ENOTTY
> (Inappropriate ioctl for device)
> _llseek(5, 0, [0], SEEK_CUR) = 0
> read(5, "package re;\n\nour $VERSION = 0.05"..., 4096) = 1340
> read(5, "", 4096) = 0
> close(5) = 0
> read(4, "[:>\\]])?(.*)/s);\n $dirpath ||"..., 4096) = 4096
> brk(0x8235000) = 0x8235000
> read(4, ".\'. fileparse() would return \'f"..., 4096) = 3069
> read(4, "", 4096) = 0
> close(4) = 0
> read(3, "t_date {\n\tmy ($when) = @_;\n\n\tmy "..., 4096) = 2195
> read(3, "", 4096) = 0
> close(3) = 0
> uname({sys="Linux", node="rimwards", ...}) = 0
> write(2, "Use of uninitialized value in su"..., 100Use of uninitialized
> value in substitution (s///) at /usr/share/perl/5.8/File/Basename.pm
> line 338.
> ) = 100
> [snip]
> stat64("/usr/share/perl/5.8/Carp/Heavy.pm", {st_mode=S_IFREG|0644,
> st_size=5747, ...}) = 0
> open("/usr/share/perl/5.8/Carp/Heavy.pm", O_RDONLY|O_LARGEFILE) = 3
> ioctl(3, SNDCTL_TMR_TIMEBASE or TCGETS, 0xbffff598) = -1 ENOTTY
> (Inappropriate ioctl for device)
> _llseek(3, 0, [0], SEEK_CUR) = 0
> read(3, "# Carp::Heavy uses some variable"..., 4096) = 4096
> brk(0x8256000) = 0x8256000
> read(3, "s($called, $caller, $cache);\n "..., 4096) = 1651
> read(3, "", 4096) = 0
> close(3) = 0
> write(2, "fileparse(): need a valid pathna"..., 63fileparse(): need a
> valid pathname at ./bindgraph.cgi line 181
> ) = 63
> exit_group(255) = ?
> Process 28639 detached
>
>
> Is this what you got too?

No, but I didn't try to execute the cgi on the command line.
I got RRD errors. The rrd command that generates images did not work for a
formatting reason.
The error was well known to me, since I had several time problems with
mailgraph / amavis-stats which are tools very similar to bindgraph.

Just try the newer version, no need to backport it since it's just a perl
script. You can use pinning or just downlad appart the deb file from testing.

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQBGrOFpDltnDmLJYdARAkgoAKC/0iyAX9i3Ld3uxr9ajau83HPxnwCfeBsE
ErbK+m3tf5YfpybfvhfU0Ek=
=gu9K
-----END PGP SIGNATURE-----

Gilles Mocellin wrote:
> Just try the newer version, no need to backport it since it's just a perl
> script. You can use pinning or just downlad appart the deb file from testing.

Aah, awesome. The package from testing works like a charm!

Thanks for the .


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

Gilles Mocellin wrote:
> Just try the newer version, no need to backport it since it's just a perl
> script. You can use pinning or just downlad appart the deb file from testing.

Aah, awesome. The package from testing works like a charm!

Thanks for the .


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

koffiejunkie wrote:
> Gilles Mocellin wrote:
>> Just try the newer version, no need to backport it since it's just a
>> perl script. You can use pinning or just downlad appart the deb file
>> from testing.
>
> Aah, awesome. The package from testing works like a charm!
>
> Thanks for the .
>
>


Damn, I thought I could diff the two and try to wedge the changes into
amavis-stats - then I realised amavis-stats is writtein in php :-(


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org