c3 | 2 Jul 08:06 2005

uninitialized value in grapher.cgi

I did a fresh install on FC4 and cricket-1.0.5.

These seem to be real common with cricket-1.0.5;
Use of uninitialized value in split at /home/cricket/cricket/grapher.cgi 
line 953.
 Use of uninitialized value in split at /home/cricket/cricket/grapher.cgi 
line 1381.
 Use of uninitialized value in split at /home/cricket/cricket/grapher.cgi 
line 1497.
Is the general fix, to roll back to grapher.cgi from older cricket versions?
Or is there a patch around to fix grapher.cgi that I missed?

TIA,
c3 

-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click
Brent | 2 Jul 11:13 2005
Picon

Re: uninitialized value in grapher.cgi

c3 wrote:
> I did a fresh install on FC4 and cricket-1.0.5.
> 
> These seem to be real common with cricket-1.0.5;
> Use of uninitialized value in split at /home/cricket/cricket/grapher.cgi 
> line 953.
> Use of uninitialized value in split at /home/cricket/cricket/grapher.cgi 
> line 1381.
> Use of uninitialized value in split at /home/cricket/cricket/grapher.cgi 
> line 1497.
> Is the general fix, to roll back to grapher.cgi from older cricket 
> versions?
> Or is there a patch around to fix grapher.cgi that I missed?
> 
> TIA,
> c3
> 
> 
> 
> -------------------------------------------------------
> SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
> from IBM. Find simple to follow Roadmaps, straightforward articles,
> informative Webcasts and more! Get everything you need to get up to
> speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click
> _______________________________________________
> cricket-users mailing list
> cricket-users <at> lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/cricket-users

Might check your config files and make sure they match the 1.0.5 spec
(Continue reading)

Jehad Arnold | 4 Jul 09:35 2005

policy-map

Hi

 

I would like to know if it is possible to graph conformed and drop rate of a Cisco policy-map as well as the class-maps inside of the policy-map.

 

sh policy-map interface gigabitEthernet 0/1.67

 GigabitEthernet0/1.67

 

  Service-policy output: p-vod-vas-out-GE-0/1.67

 

    Class-map: class-default (match-any)

      1288583034 packets, 734582778486 bytes

      30 second offered rate 6492000 bps, drop rate 0 bps

      Match: any

      Traffic Shaping

           Target/Average   Byte   Sustain   Excess    Interval  Increment

             Rate           Limit  bits/int  bits/int  (ms)      (bytes) 

          8500000/8500000   51000  204000    204000    24        25500   

 

        Adapt  Queue     Packets   Bytes     Packets   Bytes     Shaping

        Active Depth                         Delayed   Delayed   Active

        -      0         1265101216 473465601 318508526 1525732086 no

 

      Service-policy : c-vod-vas-out-GE-0/1.67

 

        Class-map: c-vod-vas-stream-prec2-GE-0/1.67 (match-all)

          14271145 packets, 9422857654 bytes

          30 second offered rate 28000 bps, drop rate 0 bps

          Match: access-group name c-vod-vas-stream-prec2-GE-0/1.67

          police:

              cir 256000 bps, bc 48000 bytes

            conformed 13880395 packets, 8878936619 bytes; actions:

              transmit

            exceeded 390199 packets, 543586321 bytes; actions:

              drop

            conformed 28000 bps, exceed 0 bps

 

        Class-map: c-vodacom-vas-prec2-GE-0/1.67 (match-all)

          40142089 packets, 12735626599 bytes

          30 second offered rate 40000 bps, drop rate 0 bps

          Match: access-group name c-vod-vas-prec2-GE-0/1.67

          police:

              cir 256000 bps, bc 48000 bytes

            conformed 39748346 packets, 12278729800 bytes; actions:

              transmit

            exceeded 388280 packets, 456245391 bytes; actions:

              drop

            conformed 40000 bps, exceed 0 bps

 

        Class-map: class-default (match-any)

          1234169802 packets, 712424295433 bytes

          30 second offered rate 6420000 bps, drop rate 0 bps

          Match: any

          Queueing

            Flow Based Fair Queueing

            Maximum Number of Hashed Queues 256

        (total queued/total drops/no-buffer drops) 0/5482314/0

          police:

              cir 8500000 bps, bc 1593750 bytes

            conformed 1216952298 packets, 692867372054 bytes; actions:

              transmit

            exceeded 17221027 packets, 19558739822 bytes; actions:

              drop

            conformed 6576000 bps, exceed 0 bps

 

thanx

 

 

Jehad Arnold

Mobile : +27 82 5200 169

Office : +27 21 940 9791

 

Jason Byrne | 5 Jul 06:56 2005
Picon

dividing one datasource by another


Dear Cricket Users

I've been using cricket and genDevconfig for sometime, I am looking for
The best way to divide two datasources and graph the result, I can
probably 
Do it with mtargets but this seems a bit late in the process ideally it
would 
Be simpiler in the view dictionary is this possible ?

Regard j

-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_idt77&alloc_id492&op=click
Rick Frey | 5 Jul 08:22 2005

Re: dividing one datasource by another

not a cricket expert, but....
I don't know of way to perform a mathematical function on two data 
sources other than to use mtargets (hopefully your datasources are in 
separate targets).  The grapher (scale) allows arithmetic operations on 
a single data source but I know of no means to combine multiple data 
sources in view or graph scope.

Jason Byrne wrote:

>Dear Cricket Users
>
>I've been using cricket and genDevconfig for sometime, I am looking for
>The best way to divide two datasources and graph the result, I can
>probably 
>Do it with mtargets but this seems a bit late in the process ideally it
>would 
>Be simpiler in the view dictionary is this possible ?
>
>Regard j
>
>
>
>-------------------------------------------------------
>SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
>from IBM. Find simple to follow Roadmaps, straightforward articles,
>informative Webcasts and more! Get everything you need to get up to
>speed, fast. http://ads.osdn.com/?ad_idt77&alloc_id492&opÌk
>_______________________________________________
>cricket-users mailing list
>cricket-users <at> lists.sourceforge.net
>https://lists.sourceforge.net/lists/listinfo/cricket-users
>  
>

--

-- 
Reclaim your Inbox by trying Thunderbird! 
<http://www.mozilla.org/products/thunderbird/>
and while you're at it
Get Firefox! <http://www.mozilla.org/products/firefox/>

-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click
c3 | 6 Jul 03:20 2005

Re: ***SPAM*** Re: uninitialized value in grapher.cgi

Thanks for the pointer, however, try as I may, I have not been able to find 
the 1.0.5 spec for the config files. I am working in the switch-ports config 
tree, if that helps limit which things I might be leaving out. Which 
document contains the 1.0.5 spec for switch-ports, or things instanced?

----- Original Message ----- 
From: "Brent" <overlordq <at> gmail.com>
To: "c3" <c3c1 <at> c3c1.com>
Cc: <cricket-users <at> lists.sourceforge.net>
Sent: Saturday, July 02, 2005 4:13 AM
Subject: ***SPAM*** Re: [cricket-users] uninitialized value in grapher.cgi

> c3 wrote:
>> I did a fresh install on FC4 and cricket-1.0.5.
>>
>> These seem to be real common with cricket-1.0.5;
>> Use of uninitialized value in split at /home/cricket/cricket/grapher.cgi 
>> line 953.
>> Use of uninitialized value in split at /home/cricket/cricket/grapher.cgi 
>> line 1381.
>> Use of uninitialized value in split at /home/cricket/cricket/grapher.cgi 
>> line 1497.
>> Is the general fix, to roll back to grapher.cgi from older cricket 
>> versions?
>> Or is there a patch around to fix grapher.cgi that I missed?
>>
>> TIA,
>> c3
>>
> Might check your config files and make sure they match the 1.0.5 spec
> 

-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click
Matt McLeod | 6 Jul 04:30 2005
Picon
Picon

Re: dividing one datasource by another

It's quite possible to use other datasources to derive the stuff to  
graph.

Basic trick is to include all the datasources you need in the view,  
and then you can refer to the other datasources when setting up the  
scale tag as ds<n> where <n> starts at 0.

e.g.:

view Memory
         elements = "Mem_Avail_Real Mem_Avail_Swap
                     Mem_Total_Real Mem_Total_Swap"

graph Mem_Avail_Swap
         draw-as         = LINE2
         legend          = "Swap"
         scale           = ds3,/,100,*,100,-,-1,*
         y-axis          = "Percent used"
         units           = "kB"
         si-units        = false

graph Mem_Avail_Real
         draw-as         = AREA
         legend          = "Real"
         scale           = ds2,/,100,*,100,-,-1,*
         y-axis          = "Percent used"
         units           = "kB"
         si-units        = false

graph Mem_Total_Swap
         draw-as         = LINE1
         legend          = "Total Swap"
         scale           = ds3,-,100,+
         y-axis          = "Percent used"
         units           = "kB"
         si-units        = false

graph Mem_Total_Real
         draw-as         = LINE1
         legend          = "Total Real"
         scale           = ds2,-,100,+
         y-axis          = "Percent used"
         units           = "kB"
         si-units        = false

The main disadvantage of this approach is that you wind up with any  
datasources you just wanted to use to derive the "real" graph also  
graphed.  I was using a trick where I'd set up a colour:

   color null ''

and then set the graphs I didn't want to see to use that, and this  
worked quite nicely on my test system, but for some reason it doesn't  
work on the production setup (test was Debian/x86 with whatever  
cricket/rrdtool/etc packages shipped with that, prod is RHAS/x86-64  
with current rrdtool/cricket).  If it works for you, though, it's a  
really handy way to work around this limitation.

The other hitch is that it seems that the avg/max figures aren't  
calculated correctly when you do this.  I don't really care about  
that, but you might.  Not sure why it happens, haven't had time to  
debug it yet.

Matt

--

-- 
Matt McLeod
Systems Programmer
Information Infrastructure
The University of Melbourne

-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click
Ethan Torretta | 6 Jul 20:03 2005

x86 linux supports liveHead3

Hi,

In lib/RRD/Format.pm under the aix/x86 architecture there is no
liveHead3 format definition, but the win32 and solaris format works just
fine.  This of course helps enormously to have defined, since it
eliminates a ton of undef errors and enables the last-update summary
display with rrdtool 1.2.9.  I've attached the one line patch.

Ethan

diff --unified ~cricket/src/cricket-1.0.5/lib/RRD/Format.pm
~cricket/cricket/lib/RRD/Format.pm:

--- /home/cricket/src/cricket-1.0.5/lib/RRD/Format.pm   2004-01-20
18:11:09.000000000 -0800
+++ /home/cricket/cricket/lib/RRD/Format.pm     2005-07-06
10:50:18.000000000 -0700
 <at>  <at>  -123,6 +123,7  <at>  <at> 
         $self->{'cdpDef'} = "d L x4 x64";

         $self->{'liveHead'} = "L";
+        $self->{'liveHead3'} = "L L";
         $self->{'rraPtr'} = "L";
         $self->{'element'} = "d";

-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_idt77&alloc_id492&op=click
Jason Signalness | 7 Jul 20:51 2005
Picon

trouble with mtargets when more than 11 targets listed

Hello,

I'm having a strange problem.  We're using version 1.0.4pre2.

I have an mtarget defined as follows:

target overall_total
    display-name        = "Overall Total"
    short-desc          = "Total associated stations, for all tower sites"
    mtargets            = "radio10w1;radio10w1;radio10w1;radio10w1;
                                    radio10w1;radio10w1;radio10w1;radio10w1;
                                    radio10w1;radio10w1;radio10w1;radio10w1"
    mtargets-ops        = "sum()"
    unknown-is-zero     = true

If I remove one instance of "radio10w1" from the list, the graph works.  
If I add it back in, the graph does not work.  The duplicates are 
because I'm testing.

What happens is the grapher.cgi spits out the HTML.  The part with the 
textual information about the total and averages does appear fine.  But 
the images don't load.  It just sits there for a very long time.

The Apache error log doesn't provide anything too out of the ordinary 
for Cricket, but I did notice these:
 Premature end of script headers: grapher.cgi, referer: http:// . . . 
/smartbridge-radios/overall_total?view=associated%20stations

Am I hitting some kind of limit?  Is there a maximum number of mtargets 
that can be specified?

Thanks in advance,
Jason

-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click
Brent Van Dussen | 7 Jul 21:35 2005

Re: trouble with mtargets when more than 11 targets listed

Is there a particular reason all of the target names are the same?  That 
seems kind of odd.

-Brent

At 11:51 AM 7/7/2005, Jason Signalness wrote:
>Hello,
>
>I'm having a strange problem.  We're using version 1.0.4pre2.
>
>I have an mtarget defined as follows:
>
>target overall_total
>    display-name        = "Overall Total"
>    short-desc          = "Total associated stations, for all tower sites"
>    mtargets            = "radio10w1;radio10w1;radio10w1;radio10w1;
>                                    radio10w1;radio10w1;radio10w1;radio10w1;
>                                    radio10w1;radio10w1;radio10w1;radio10w1"
>    mtargets-ops        = "sum()"
>    unknown-is-zero     = true
>
>If I remove one instance of "radio10w1" from the list, the graph works.
>If I add it back in, the graph does not work.  The duplicates are because 
>I'm testing.
>
>What happens is the grapher.cgi spits out the HTML.  The part with the 
>textual information about the total and averages does appear fine.  But 
>the images don't load.  It just sits there for a very long time.
>
>The Apache error log doesn't provide anything too out of the ordinary for 
>Cricket, but I did notice these:
>Premature end of script headers: grapher.cgi, referer: http:// . . . 
>/smartbridge-radios/overall_total?view=associated%20stations
>
>Am I hitting some kind of limit?  Is there a maximum number of mtargets 
>that can be specified?
>
>Thanks in advance,
>Jason
>
>
>-------------------------------------------------------
>SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
>from IBM. Find simple to follow Roadmaps, straightforward articles,
>informative Webcasts and more! Get everything you need to get up to
>speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click
>_______________________________________________
>cricket-users mailing list
>cricket-users <at> lists.sourceforge.net
>https://lists.sourceforge.net/lists/listinfo/cricket-users

-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click

Gmane