code specific rrdidx question
You might compare your graph definition with the ones used for disks. The disk detail page can show more than one image. This might be a restriction on CUSTOM graphs instead of the ones that are built in. Henrik mentioned a while back that we could only have one graph on the detail page, and that it would be a big chore to change the code to allow more than one custom graph on the detail page. Evidently the trend page uses a different path. GLH -----Original Message----- From: Jeff Newman [mailto:user-e96740e73ca8@xymon.invalid] Sent: Wednesday, March 22, 2006 5:10 PM To: user-ae9b8668bcde@xymon.invalid Subject: [hobbit] Re: code specific rrdidx question UPDATE: I somewhat fixed the problem. This is for sure: A directory that has: sar,cpu0 , sar,cpu1 , sar,cpu2 , sar,cpu3 ...... sar,cpu10, ... sar,cpu15 will break on both the status and trends page after it does sar,cpu1 I changed all the RRD's/tests to do: sar,cpu00 , sar,cpu01 , sar,cpu02 , sar,cpu03 ..... sar,cpu15 and it now does not break. I still only see 0,1,2,3 on the button's page. The trends page shows all 16 cpu's. Any idea why the button page only shows the first 4? Again, I have sar::1 in the hobbitserver.cfg, so it generates 1 graph for each CPU Thanks, Jeff On 3/22/06, Jeff Newman <user-e96740e73ca8@xymon.invalid> wrote:
Sorry for the traffic, wanted to change this to say that maybe it isn't just (or plain isn't) RRDIDX, but maybe FNPATTERN as well. So add to my questions: Does FNPATTERN have a problem distinguishing between cpu1 and cpu10 -Jeff On 3/22/06, Jeff Newman <user-e96740e73ca8@xymon.invalid> wrote:All, I don't know C all that well, so I need some assistance. what are the limits/constraints of RRDIDX for graphing? I believe the answer is buried somewhere in hobbitgraph.c Here is my test case: I had: sar,cpu0.rrd sar,cpu1.rrd sar,cpu2.rrd sar,cpu3.rrd ... sar,cpu10.rrd sar,cpu11.rrd ... sar,cpu15.rrd hobbitgraph.cfg: [sar] FNPATTERN sar,(.*).rrd TITLE CPU sar YAXIS % DEF:usr at RRDIDX@=@RRDFN@:cpu at RRDIDX@pcntusr:AVERAGE DEF:sys at RRDIDX@=@RRDFN@:cpu at RRDIDX@pcntsys:AVERAGE DEF:wio at RRDIDX@=@RRDFN@:cpu at RRDIDX@pcntwio:AVERAGE DEF:idl at RRDIDX@=@RRDFN@:cpu at RRDIDX@pcntidl:AVERAGE AREA:usr at RRDIDX@#0000FF:@RRDPARAM at pcntusr GPRINT:usr at RRDIDX@:LAST: \: %5.1lf (cur) GPRINT:usr at RRDIDX@:MAX: \: %5.1lf (max) GPRINT:usr at RRDIDX@:MIN: \: %5.1lf (min) GPRINT:usr at RRDIDX@:AVERAGE: \: %5.1lf (avg)\n STACK:sys at RRDIDX@#FF0000:@RRDPARAM at pcntsys GPRINT:sys at RRDIDX@:LAST: \: %5.1lf (cur) GPRINT:sys at RRDIDX@:MAX: \: %5.1lf (max) GPRINT:sys at RRDIDX@:MIN: \: %5.1lf (min) GPRINT:sys at RRDIDX@:AVERAGE: \: %5.1lf (avg)\n STACK:wio at RRDIDX@#800080:@RRDPARAM at pcntwio GPRINT:wio at RRDIDX@:LAST: \: %5.1lf (cur) GPRINT:wio at RRDIDX@:MAX: \: %5.1lf (max) GPRINT:wio at RRDIDX@:MIN: \: %5.1lf (min) GPRINT:wio at RRDIDX@:AVERAGE: \: %5.1lf (avg)\n STACK:idl at RRDIDX@#008000:@RRDPARAM at pcntidl GPRINT:idl at RRDIDX@:LAST: \: %5.1lf (cur) GPRINT:idl at RRDIDX@:MAX: \: %5.1lf (max) GPRINT:idl at RRDIDX@:MIN: \: %5.1lf (min) GPRINT:idl at RRDIDX@:AVERAGE: \: %5.1lf (avg)\n hobbitserver.cfg has sar::1 (to only draw 1 graph per cpu) for GRAPHS, and just regular "sar" is in TEST2RRD sar is in the "multitest" section of hobbitcgi.cfg, as well as listed in the extra tests section of hobbitlaunch.cfg On a host with 4 CPU's (0,1,2,3) it draws 4 graphs On a host with 16
CPU's (0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15) It would draw a graph for 0,1, and broken boxes for 2,3, and no others. SO, I removed "10,11,12,13,14,15" and outputted them to a,b,c,d,e,f respectively. The graphs then drew 0,1,2,3 and no more. If I click on a specific graph, and change the "first=" to 5, it will draw a graph for CPU4, so somewhere, the first= line isn't getting generated correctly. So I guess 2 questions: 1 already stated (limitations of RRDIDX - does is get confused by 1 vs 10 for example) The other is where the first= line gets updated and why isn't it getting updated correctly? Any insights would be much apprieciated, as this is driving me nuts. Thanks, Jeff