From Guillaume.Gourhand at alcatel-lucent.fr Fri Jul 13 08:36:13 2007 From: Guillaume.Gourhand at alcatel-lucent.fr (GOURHAND GUILLAUME) Date: Fri, 13 Jul 2007 14:36:13 +0200 Subject: [drraw-users] (no subject) Message-ID: <3309AFFCFD1EA9449FF537C182C320B168B7FF@FRVELSMBS13.ad2.ad.alcatel.com> Hello, I 'm trying to test the drraw front end for monitoring. I have some problems, consequently i would like to send a post on the forum. I am a beginner and i try to do some templates and dashboards. My problems: I choose for exemple two stations with the same metric ,then i would like to do a template where i can choose a platform and then display the two stations with the same metric on only one graph. Thanks. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://web.taranis.org/pipermail/drraw-users/attachments/20070713/5e37ae42/attachment.html From wernli at in2p3.fr Mon Jul 16 04:50:55 2007 From: wernli at in2p3.fr (Fabien Wernli) Date: Mon, 16 Jul 2007 10:50:55 +0200 Subject: [drraw-users] (no subject) In-Reply-To: <3309AFFCFD1EA9449FF537C182C320B168B7FF@FRVELSMBS13.ad2.ad.alcatel.com> References: <3309AFFCFD1EA9449FF537C182C320B168B7FF@FRVELSMBS13.ad2.ad.alcatel.com> Message-ID: <20070716085055.GB12319@ccswiss.in2p3.fr> On Fri, Jul 13, 2007 at 02:36:13PM +0200, GOURHAND GUILLAUME wrote: > I am a beginner and i try to do some templates and dashboards. My > problems: > I choose for exemple two stations with the same metric ,then i would > like to do a template where i can choose a platform and then display the > two stations with the same metric on only one graph. hmm... could you be a little more specific please? From jonran at gmail.com Tue Jul 17 10:28:25 2007 From: jonran at gmail.com (Jonny Anderson) Date: Tue, 17 Jul 2007 10:28:25 -0400 Subject: [drraw-users] External Links to Graphs Message-ID: <3df99e270707170728p64104f4djf1e7e30e76aa3a57@mail.gmail.com> Hi, I apologize if this has been answered already, but I could not uncover it by searching the archives. I would like to be able to link to graphs externally by names that I know in advance (some combination of graph type and hostname), but the graph names in drraw are Id's that I don't think I can know a priori, and they don't necessarily seem to exist until requested. I am tasked with identifying network performance anomalies in a separate database and system from RRD, but I want to be able to link to graphs that I am creating with templates in drraw from RRD's I am creating with the external DB's data. Is there a way to do this? Thanks very much in advance. Jon Anderson -------------- next part -------------- An HTML attachment was scrubbed... URL: http://web.taranis.org/pipermail/drraw-users/attachments/20070717/9c4ddcf6/attachment.html From jonran at gmail.com Tue Jul 17 11:38:02 2007 From: jonran at gmail.com (Jonny Anderson) Date: Tue, 17 Jul 2007 11:38:02 -0400 Subject: [drraw-users] External Links to Graphs In-Reply-To: <3df99e270707170728p64104f4djf1e7e30e76aa3a57@mail.gmail.com> References: <3df99e270707170728p64104f4djf1e7e30e76aa3a57@mail.gmail.com> Message-ID: <3df99e270707170838r5e951dcfm20e5901e9d7f1d90@mail.gmail.com> I think I answered my own question. By using templates, all I need is the template ID, which according to the documentation won't change, and then I can use the Base option in the URL to write the path to the RRD file, which I name by host, and also make use of the Start, End and other options to dynamically write URLs for handy searching. My only worry is that I might lose the template ID if my temp dir is wiped out, but I will test it out to make sure. If anyone has a better way, I would love to hear it. Jon On 7/17/07, Jonny Anderson wrote: > > Hi, > > I apologize if this has been answered already, but I could not uncover it > by searching the archives. > > I would like to be able to link to graphs externally by names that I know > in advance (some combination of graph type and hostname), but the graph > names in drraw are Id's that I don't think I can know a priori, and they > don't necessarily seem to exist until requested. I am tasked with > identifying network performance anomalies in a separate database and system > from RRD, but I want to be able to link to graphs that I am creating with > templates in drraw from RRD's I am creating with the external DB's data. > > Is there a way to do this? > > Thanks very much in advance. > > Jon Anderson > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://web.taranis.org/pipermail/drraw-users/attachments/20070717/448542ef/attachment.html From jkrauska at gmail.com Tue Jul 17 14:32:37 2007 From: jkrauska at gmail.com (Joel Krauska) Date: Tue, 17 Jul 2007 11:32:37 -0700 Subject: [drraw-users] Zooming? Message-ID: <469D0B45.3090508@gmail.com> Cacti has some nice javascript embedded that allows you to select a certain time window on a graph and zoom in to that time window. Any thoughts on adding this functionality to drraw? --joel From jkrauska at gmail.com Tue Jul 17 14:42:06 2007 From: jkrauska at gmail.com (Joel Krauska) Date: Tue, 17 Jul 2007 11:42:06 -0700 Subject: [drraw-users] More compatibility data Message-ID: <469D0D7E.1010709@gmail.com> At the bottom of http://web.taranis.org/drraw/ You have tables summarizing compatibility. I'd like to report that drraw appears to work fine on: SuSE 10.0 Which has: Perl: 5.8.7 CGI.pm: 3.10 Everything else is already reported. (apache, RRDs.pm) --joel From wernli at in2p3.fr Wed Jul 18 04:00:04 2007 From: wernli at in2p3.fr (Fabien Wernli) Date: Wed, 18 Jul 2007 10:00:04 +0200 Subject: [drraw-users] Zooming? In-Reply-To: <469D0B45.3090508@gmail.com> References: <469D0B45.3090508@gmail.com> Message-ID: <20070718080004.GF19978@ccswiss.in2p3.fr> On Tue, Jul 17, 2007 at 11:32:37AM -0700, Joel Krauska wrote: > Any thoughts on adding this functionality to drraw? hmmm... do you javascript? ;) From kalt at taranis.org Wed Jul 18 06:28:53 2007 From: kalt at taranis.org (Christophe Kalt) Date: Wed, 18 Jul 2007 06:28:53 -0400 Subject: [drraw-users] External Links to Graphs In-Reply-To: <3df99e270707170838r5e951dcfm20e5901e9d7f1d90@mail.gmail.com> References: <3df99e270707170728p64104f4djf1e7e30e76aa3a57@mail.gmail.com> <3df99e270707170838r5e951dcfm20e5901e9d7f1d90@mail.gmail.com> Message-ID: <20070718102853.GB29232@bzz.taranis.org> On Jul 17, Jonny Anderson wrote: | I think I answered my own question. Seems that you have, indeed. | My only worry is that I might lose the template ID if my temp dir is wiped | out, but I will test it out to make sure. The ID will only change if you clone the template and decide to use the new clone. Stuff in the temp directory can be deleted any time, it doesn't matter. | If anyone has a better way, I would love to hear it. Nope, that's it. From kalt at taranis.org Wed Jul 18 06:31:19 2007 From: kalt at taranis.org (Christophe Kalt) Date: Wed, 18 Jul 2007 06:31:19 -0400 Subject: [drraw-users] More compatibility data In-Reply-To: <469D0D7E.1010709@gmail.com> References: <469D0D7E.1010709@gmail.com> Message-ID: <20070718103119.GC29232@bzz.taranis.org> On Jul 17, Joel Krauska wrote: | At the bottom of http://web.taranis.org/drraw/ | You have tables summarizing compatibility. | | I'd like to report that drraw appears to work fine on: [...] This process is somewhat automated, it uses the data that is submitted by clicking on the 'Add this installation to the compatibility report' link in your index. From kalt at taranis.org Wed Jul 18 06:32:46 2007 From: kalt at taranis.org (Christophe Kalt) Date: Wed, 18 Jul 2007 06:32:46 -0400 Subject: [drraw-users] Zooming? In-Reply-To: <469D0B45.3090508@gmail.com> References: <469D0B45.3090508@gmail.com> Message-ID: <20070718103246.GD29232@bzz.taranis.org> On Jul 17, Joel Krauska wrote: | Cacti has some nice javascript embedded that allows you to select a | certain time window on a graph and zoom in to that time window. i guess Cacti has made some progress from when i installed it and was horrified enough to start writing drraw. | Any thoughts on adding this functionality to drraw? Sounds like a neat feature, it's a matter of someone (you?) looking into it. From nokess at upmc.edu Thu Jul 19 10:59:04 2007 From: nokess at upmc.edu (Nokes, Sam X) Date: Thu, 19 Jul 2007 10:59:04 -0400 Subject: [drraw-users] Unable to graph with VDEF operators LSLSLOPE, LSLINT, and LSLCORREL Message-ID: <66985FA02F58D14E97230077F3388BF903250E7C@1upmc-msx14.acct.upmchs.net> We have drraw v2.2a3 implemented on Red Hat and are successful in generating graphs that depict Data Sources from RRAs that we have captured. Typical CDEFs and VDEFs such as Average and 95th percentile also produced graphs as one would expect. However, we are having difficulty displaying a projection with the Least Squares Line VDEF operators LSLSLOPE, LSLINT, and LSLCORREL. We suspect a usage issue, however we are also questioning whether the operators are usable with the 'beater' code. Here is the rrd invocation from drraw we have at the moment: rrdtool graph - \ --start=end-86400 \ --title=CPU Projection \ --vertical-label=Utilization \ --imgformat=PNG \ --width=600 \ --base=1000 \ --height=200 \ --interlaced \ DEF:a=/data/nmon2web/html/020256D3D/Shared/c0prd01/rrddata//cpu_all.rrd: UserPct:AVERAGE \ DEF:b=/data/nmon2web/html/020256D3D/Shared/c0prd01/rrddata//cpu_all.rrd: SysPct:AVERAGE \ DEF:c=/data/nmon2web/html/020256D3D/Shared/c0prd01/rrddata//cpu_all.rrd: WaitPct:AVERAGE \ CDEF:G=a,b,c,+,+ \ VDEF:D=G,LSLSLOPE \ VDEF:H=G,LSLINT \ VDEF:F=G,LSLCORREL \ LINE1:G#FFA500: Usr+Sys+Wait \ LINE1:D#000080:Slope \ LINE1:H#FF0000: \ LINE1:F#008000: and the graph that is generated: Realizing v2.2a3 is beta, and noting a comment by Alex: "In RRDtool 1.2 I made a start with VDEF processing. VDEFs aren't arrays like CDEFs are, they are single values. The work is far from finished, however certain functions are already usable."( http://www.vandenbogaerdt.nl/rrdtool/total.php ), the questions follow: (1) Are the VDEF operators LSLSLOPE, LSLINT, and LSLCORREL usable in drraw v2.2a3? (2) Should the vnames be placed into a linear equation: y=LSLSLOPE*LSLCORREL+LSLINT ? We tried to add a CDEF definition substituting VDEF vnames in a vpn format (D,F,*,H,+) and received an error from drraw stating "rpn expressions without DEF or CDEF variables are not supported" (3) Has anyone generated a least squares projection with rrdtool similar to the following? If so what is the rrd invocation? -------------- next part -------------- An HTML attachment was scrubbed... URL: http://web.taranis.org/pipermail/drraw-users/attachments/20070719/54edde9d/attachment-0001.html -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: image/gif Size: 34984 bytes Desc: ATT16011.gif Url : http://web.taranis.org/pipermail/drraw-users/attachments/20070719/54edde9d/attachment-0001.gif -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: image/jpeg Size: 206976 bytes Desc: Outlook.jpg Url : http://web.taranis.org/pipermail/drraw-users/attachments/20070719/54edde9d/attachment-0001.jpe From vitroth+ at cmu.edu Fri Jul 20 14:34:54 2007 From: vitroth+ at cmu.edu (David Nolan) Date: Fri, 20 Jul 2007 14:34:54 -0400 Subject: [drraw-users] Unable to graph with VDEF operators LSLSLOPE, LSLINT, and LSLCORREL In-Reply-To: <66985FA02F58D14E97230077F3388BF903250E7C@1upmc-msx14.acct.upmchs.net> References: <66985FA02F58D14E97230077F3388BF903250E7C@1upmc-msx14.acct.upmchs.net> Message-ID: <1cbe50830707201134i509f45b2sf5fe629c3cf4fc9@mail.gmail.com> On 7/19/07, Nokes, Sam X wrote > > Realizing v2.2a3 is beta, and noting a comment by Alex: "In RRDtool 1.2 I > made a start with VDEF processing. VDEFs aren't arrays like CDEFs are, they > are single values. The work is far from finished, however certain functions > are already usable."( http://www.vandenbogaerdt.nl/rrdtool/total.php ), > the questions follow: > (1) Are the VDEF operators LSLSLOPE, LSLINT, and LSLCORREL usable in drraw > v2.2a3? > (2) Should the vnames be placed into a linear equation: > y=LSLSLOPE*LSLCORREL+LSLINT ? > We tried to add a CDEF definition substituting VDEF vnames in a vpn > format (D,F,*,H,+) and > received an error from drraw stating "rpn expressions without DEF or > CDEF variables are not supported" > (3) Has anyone generated a least squares projection with rrdtool similar > to the following? If so what is the rrd invocation? > Sam, I've never used these functions in rrdtool before, but my reading of the documentation indicates that you are close with statement 2 above. I believe you need to define the VDEF variables as you're doing, and then define a CDEF using those variables and the automatic COUNT variable, i.e. CDEF:trend=D,COUNT,*,H,+ (using the variables you defined above), and then graph the CDEF. -David -------------- next part -------------- An HTML attachment was scrubbed... URL: http://web.taranis.org/pipermail/drraw-users/attachments/20070720/38446604/attachment.html From nokess at upmc.edu Fri Jul 20 17:01:19 2007 From: nokess at upmc.edu (Nokes, Sam X) Date: Fri, 20 Jul 2007 17:01:19 -0400 Subject: [drraw-users] Unable to graph with VDEF operators LSLSLOPE, LSLINT, and LSLCORREL In-Reply-To: <1cbe50830707201134i509f45b2sf5fe629c3cf4fc9@mail.gmail.com> Message-ID: <66985FA02F58D14E97230077F3388BF903250E80@1upmc-msx14.acct.upmchs.net> Appreciate the assist David. Unfortunately I received the familiar "rpn expressions without DEF or CDEF variables are not supported" error Here is the rrd invocation generated by drraw for review: rrdtool graph - \ --start=end-86400 \ --title=CPU Projection \ --vertical-label=Utilization \ --imgformat=PNG \ --width=600 \ --base=1000 \ --height=200 \ --interlaced \ DEF:a=/data/nmon2web/html/020256D3D/Shared/c0prd01/rrddata//cpu_all.rrd: UserPct:AVERAGE \ DEF:b=/data/nmon2web/html/020256D3D/Shared/c0prd01/rrddata//cpu_all.rrd: SysPct:AVERAGE \ DEF:c=/data/nmon2web/html/020256D3D/Shared/c0prd01/rrddata//cpu_all.rrd: WaitPct:AVERAGE \ CDEF:G=a,b,c,+,+ \ VDEF:D=G,LSLSLOPE \ VDEF:H=G,LSLINT \ VDEF:F=G,LSLCORREL \ CDEF:E=D,COUNT,*,H,+ \ LINE1:G#FFA500: Usr+Sys+Wait \ LINE1:E#FF0000: Projection ________________________________ From: drraw-users-bounces at taranis.org [mailto:drraw-users-bounces at taranis.org] On Behalf Of David Nolan Sent: Friday, July 20, 2007 2:35 PM To: drraw-users at taranis.org Subject: Re: [drraw-users] Unable to graph with VDEF operators LSLSLOPE,LSLINT, and LSLCORREL On 7/19/07, Nokes, Sam X wrote Realizing v2.2a3 is beta, and noting a comment by Alex: "In RRDtool 1.2 I made a start with VDEF processing. VDEFs aren't arrays like CDEFs are, they are single values. The work is far from finished, however certain functions are already usable."( http://www.vandenbogaerdt.nl/rrdtool/total.php ), the questions follow: (1) Are the VDEF operators LSLSLOPE, LSLINT, and LSLCORREL usable in drraw v2.2a3? (2) Should the vnames be placed into a linear equation: y=LSLSLOPE*LSLCORREL+LSLINT ? We tried to add a CDEF definition substituting VDEF vnames in a vpn format (D,F,*,H,+) and received an error from drraw stating "rpn expressions without DEF or CDEF variables are not supported" (3) Has anyone generated a least squares projection with rrdtool similar to the following? If so what is the rrd invocation? Sam, I've never used these functions in rrdtool before, but my reading of the documentation indicates that you are close with statement 2 above. I believe you need to define the VDEF variables as you're doing, and then define a CDEF using those variables and the automatic COUNT variable, i.e. CDEF:trend=D,COUNT,*,H,+ (using the variables you defined above), and then graph the CDEF. -David -------------- next part -------------- An HTML attachment was scrubbed... URL: http://web.taranis.org/pipermail/drraw-users/attachments/20070720/2c171d88/attachment.html From nokess at upmc.edu Fri Jul 20 17:24:57 2007 From: nokess at upmc.edu (Nokes, Sam X) Date: Fri, 20 Jul 2007 17:24:57 -0400 Subject: [drraw-users] Unable to graph with VDEF operators LSLSLOPE, LSLINT, and LSLCORREL In-Reply-To: <66985FA02F58D14E97230077F3388BF903250E80@1upmc-msx14.acct.upmchs.net> Message-ID: <66985FA02F58D14E97230077F3388BF903250E81@1upmc-msx14.acct.upmchs.net> On the hand, when I add POP per Alex's answer at https://lists.oetiker.ch/pipermail/rrd-users/2006-September/011643.html ir works! Here is the successful invocation from drraw: rrdtool graph - \ --start=end-86400 \ --title=CPU Projection \ --vertical-label=Utilization \ --imgformat=PNG \ --width=600 \ --base=1000 \ --height=200 \ --interlaced \ DEF:a=/data/nmon2web/html/020256D3D/Shared/c0prd01/rrddata//cpu_all.rrd: UserPct:AVERAGE \ DEF:b=/data/nmon2web/html/020256D3D/Shared/c0prd01/rrddata//cpu_all.rrd: SysPct:AVERAGE \ DEF:c=/data/nmon2web/html/020256D3D/Shared/c0prd01/rrddata//cpu_all.rrd: WaitPct:AVERAGE \ CDEF:G=a,b,c,+,+ \ VDEF:D=G,LSLSLOPE \ VDEF:H=G,LSLINT \ VDEF:F=G,LSLCORREL \ CDEF:E=G,POP,D,COUNT,*,H,+ \ LINE1:G#FFA500: Usr+Sys+Wait \ LINE1:E#FF0000: Projection Now if I could only extend the end time beyond today we're set... ________________________________ From: drraw-users-bounces at taranis.org [mailto:drraw-users-bounces at taranis.org] On Behalf Of Nokes, Sam X Sent: Friday, July 20, 2007 5:01 PM To: David Nolan; drraw-users at taranis.org Subject: Re: [drraw-users] Unable to graph with VDEF operators LSLSLOPE,LSLINT, and LSLCORREL Appreciate the assist David. Unfortunately I received the familiar "rpn expressions without DEF or CDEF variables are not supported" error Here is the rrd invocation generated by drraw for review: rrdtool graph - \ --start=end-86400 \ --title=CPU Projection \ --vertical-label=Utilization \ --imgformat=PNG \ --width=600 \ --base=1000 \ --height=200 \ --interlaced \ DEF:a=/data/nmon2web/html/020256D3D/Shared/c0prd01/rrddata//cpu_all.rrd: UserPct:AVERAGE \ DEF:b=/data/nmon2web/html/020256D3D/Shared/c0prd01/rrddata//cpu_all.rrd: SysPct:AVERAGE \ DEF:c=/data/nmon2web/html/020256D3D/Shared/c0prd01/rrddata//cpu_all.rrd: WaitPct:AVERAGE \ CDEF:G=a,b,c,+,+ \ VDEF:D=G,LSLSLOPE \ VDEF:H=G,LSLINT \ VDEF:F=G,LSLCORREL \ CDEF:E=D,COUNT,*,H,+ \ LINE1:G#FFA500: Usr+Sys+Wait \ LINE1:E#FF0000: Projection ________________________________ From: drraw-users-bounces at taranis.org [mailto:drraw-users-bounces at taranis.org] On Behalf Of David Nolan Sent: Friday, July 20, 2007 2:35 PM To: drraw-users at taranis.org Subject: Re: [drraw-users] Unable to graph with VDEF operators LSLSLOPE,LSLINT, and LSLCORREL On 7/19/07, Nokes, Sam X wrote Realizing v2.2a3 is beta, and noting a comment by Alex: "In RRDtool 1.2 I made a start with VDEF processing. VDEFs aren't arrays like CDEFs are, they are single values. The work is far from finished, however certain functions are already usable."( http://www.vandenbogaerdt.nl/rrdtool/total.php ), the questions follow: (1) Are the VDEF operators LSLSLOPE, LSLINT, and LSLCORREL usable in drraw v2.2a3? (2) Should the vnames be placed into a linear equation: y=LSLSLOPE*LSLCORREL+LSLINT ? We tried to add a CDEF definition substituting VDEF vnames in a vpn format (D,F,*,H,+) and received an error from drraw stating "rpn expressions without DEF or CDEF variables are not supported" (3) Has anyone generated a least squares projection with rrdtool similar to the following? If so what is the rrd invocation? Sam, I've never used these functions in rrdtool before, but my reading of the documentation indicates that you are close with statement 2 above. I believe you need to define the VDEF variables as you're doing, and then define a CDEF using those variables and the automatic COUNT variable, i.e. CDEF:trend=D,COUNT,*,H,+ (using the variables you defined above), and then graph the CDEF. -David -------------- next part -------------- An HTML attachment was scrubbed... URL: http://web.taranis.org/pipermail/drraw-users/attachments/20070720/1b92d6d8/attachment-0001.html From nokess at upmc.edu Fri Jul 20 17:38:03 2007 From: nokess at upmc.edu (Nokes, Sam X) Date: Fri, 20 Jul 2007 17:38:03 -0400 Subject: [drraw-users] Unable to graph with VDEF operators LSLSLOPE, LSLINT, and LSLCORREL In-Reply-To: <66985FA02F58D14E97230077F3388BF903250E81@1upmc-msx14.acct.upmchs.net> Message-ID: <66985FA02F58D14E97230077F3388BF903250E82@1upmc-msx14.acct.upmchs.net> I may have it with the following invocation for say a one month projection (let me know if someone has a more efficient version?): rrdtool graph - \ --start=end-38d \ --end=+31d \ --title=CPU Projection \ --vertical-label=Utilization \ --imgformat=PNG \ --width=600 \ --base=1000 \ --height=200 \ --interlaced \ DEF:a=/data/nmon2web/html/020256D3D/Shared/c0prd01/rrddata//cpu_all.rrd: UserPct:AVERAGE \ DEF:b=/data/nmon2web/html/020256D3D/Shared/c0prd01/rrddata//cpu_all.rrd: SysPct:AVERAGE \ DEF:c=/data/nmon2web/html/020256D3D/Shared/c0prd01/rrddata//cpu_all.rrd: WaitPct:AVERAGE \ CDEF:G=a,b,c,+,+ \ VDEF:D=G,LSLSLOPE \ VDEF:H=G,LSLINT \ VDEF:F=G,LSLCORREL \ CDEF:E=G,POP,D,COUNT,*,H,+ \ LINE1:G#FFA500: Usr+Sys+Wait \ LINE1:E#FF0000: Projection Here is the resultant graph: ________________________________ From: Nokes, Sam X Sent: Friday, July 20, 2007 5:25 PM To: Nokes, Sam X; David Nolan; drraw-users at taranis.org Subject: RE: [drraw-users] Unable to graph with VDEF operators LSLSLOPE,LSLINT, and LSLCORREL On the hand, when I add POP per Alex's answer at https://lists.oetiker.ch/pipermail/rrd-users/2006-September/011643.html ir works! Here is the successful invocation from drraw: rrdtool graph - \ --start=end-86400 \ --title=CPU Projection \ --vertical-label=Utilization \ --imgformat=PNG \ --width=600 \ --base=1000 \ --height=200 \ --interlaced \ DEF:a=/data/nmon2web/html/020256D3D/Shared/c0prd01/rrddata//cpu_all.rrd: UserPct:AVERAGE \ DEF:b=/data/nmon2web/html/020256D3D/Shared/c0prd01/rrddata//cpu_all.rrd: SysPct:AVERAGE \ DEF:c=/data/nmon2web/html/020256D3D/Shared/c0prd01/rrddata//cpu_all.rrd: WaitPct:AVERAGE \ CDEF:G=a,b,c,+,+ \ VDEF:D=G,LSLSLOPE \ VDEF:H=G,LSLINT \ VDEF:F=G,LSLCORREL \ CDEF:E=G,POP,D,COUNT,*,H,+ \ LINE1:G#FFA500: Usr+Sys+Wait \ LINE1:E#FF0000: Projection Now if I could only extend the end time beyond today we're set... ________________________________ From: drraw-users-bounces at taranis.org [mailto:drraw-users-bounces at taranis.org] On Behalf Of Nokes, Sam X Sent: Friday, July 20, 2007 5:01 PM To: David Nolan; drraw-users at taranis.org Subject: Re: [drraw-users] Unable to graph with VDEF operators LSLSLOPE,LSLINT, and LSLCORREL Appreciate the assist David. Unfortunately I received the familiar "rpn expressions without DEF or CDEF variables are not supported" error Here is the rrd invocation generated by drraw for review: rrdtool graph - \ --start=end-86400 \ --title=CPU Projection \ --vertical-label=Utilization \ --imgformat=PNG \ --width=600 \ --base=1000 \ --height=200 \ --interlaced \ DEF:a=/data/nmon2web/html/020256D3D/Shared/c0prd01/rrddata//cpu_all.rrd: UserPct:AVERAGE \ DEF:b=/data/nmon2web/html/020256D3D/Shared/c0prd01/rrddata//cpu_all.rrd: SysPct:AVERAGE \ DEF:c=/data/nmon2web/html/020256D3D/Shared/c0prd01/rrddata//cpu_all.rrd: WaitPct:AVERAGE \ CDEF:G=a,b,c,+,+ \ VDEF:D=G,LSLSLOPE \ VDEF:H=G,LSLINT \ VDEF:F=G,LSLCORREL \ CDEF:E=D,COUNT,*,H,+ \ LINE1:G#FFA500: Usr+Sys+Wait \ LINE1:E#FF0000: Projection ________________________________ From: drraw-users-bounces at taranis.org [mailto:drraw-users-bounces at taranis.org] On Behalf Of David Nolan Sent: Friday, July 20, 2007 2:35 PM To: drraw-users at taranis.org Subject: Re: [drraw-users] Unable to graph with VDEF operators LSLSLOPE,LSLINT, and LSLCORREL On 7/19/07, Nokes, Sam X wrote Realizing v2.2a3 is beta, and noting a comment by Alex: "In RRDtool 1.2 I made a start with VDEF processing. VDEFs aren't arrays like CDEFs are, they are single values. The work is far from finished, however certain functions are already usable."( http://www.vandenbogaerdt.nl/rrdtool/total.php ), the questions follow: (1) Are the VDEF operators LSLSLOPE, LSLINT, and LSLCORREL usable in drraw v2.2a3? (2) Should the vnames be placed into a linear equation: y=LSLSLOPE*LSLCORREL+LSLINT ? We tried to add a CDEF definition substituting VDEF vnames in a vpn format (D,F,*,H,+) and received an error from drraw stating "rpn expressions without DEF or CDEF variables are not supported" (3) Has anyone generated a least squares projection with rrdtool similar to the following? If so what is the rrd invocation? Sam, I've never used these functions in rrdtool before, but my reading of the documentation indicates that you are close with statement 2 above. I believe you need to define the VDEF variables as you're doing, and then define a CDEF using those variables and the automatic COUNT variable, i.e. CDEF:trend=D,COUNT,*,H,+ (using the variables you defined above), and then graph the CDEF. -David -------------- next part -------------- An HTML attachment was scrubbed... URL: http://web.taranis.org/pipermail/drraw-users/attachments/20070720/1c8d0895/attachment-0001.html -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: image/jpeg Size: 36308 bytes Desc: Outlook.jpg Url : http://web.taranis.org/pipermail/drraw-users/attachments/20070720/1c8d0895/attachment-0001.jpe From vitroth+ at cmu.edu Sat Jul 21 10:41:41 2007 From: vitroth+ at cmu.edu (David Nolan) Date: Sat, 21 Jul 2007 10:41:41 -0400 Subject: [drraw-users] Unable to graph with VDEF operators LSLSLOPE, LSLINT, and LSLCORREL In-Reply-To: <66985FA02F58D14E97230077F3388BF903250E82@1upmc-msx14.acct.upmchs.net> References: <66985FA02F58D14E97230077F3388BF903250E81@1upmc-msx14.acct.upmchs.net> <66985FA02F58D14E97230077F3388BF903250E82@1upmc-msx14.acct.upmchs.net> Message-ID: <1cbe50830707210741q7d9aca8fl79d1e3d66d3d2e56@mail.gmail.com> On 7/20/07, Nokes, Sam X wrote: > > I may have it with the following invocation for say a one month > projection (let me know if someone has a more efficient version?): > I believe you could also use "now - X days" and "start + Y days" for your start/end values. But its basically the same approach. -David -------------- next part -------------- An HTML attachment was scrubbed... URL: http://web.taranis.org/pipermail/drraw-users/attachments/20070721/add6305a/attachment.html From ulises.best at yahoo.co.uk Wed Jul 25 14:28:01 2007 From: ulises.best at yahoo.co.uk (ulises best) Date: Wed, 25 Jul 2007 19:28:01 +0100 (BST) Subject: [drraw-users] Help needed "404 Not found error" Message-ID: <144570.90139.qm@web26707.mail.ukl.yahoo.com> Hi mates I am quite ashemed to ask you for help with this matter(as it's said that it is extremely easy to install) but i couldn't install drraw in my linux box and obtained the following error when trying : http://localhost/nagios/cgi-bin/drraw Not found the requested url /nagios/cgi-bin/drraw was not found on this server I configured it and installed CGI.pm , rddtool with RRD perl module and nothing happens. I try to use drraw to view nagios performance data. Can anyone tell me why this message appear? i am using Apache 2.0.59 maybe the error could be that i am using the incorrect cgi file in the web server. i configured with the path /usr/lib/cgi-bin/drraw and changed permissions for the user 'nobody' for access the directory: chown -R nobody /var/lib/drraw Thanks in advance Ulises --------------------------------- Yahoo! Answers - Get better answers from someone who knows. Tryit now. --------------------------------- Yahoo! Mail is the world's favourite email. Don't settle for less, sign up for your freeaccount today. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://web.taranis.org/pipermail/drraw-users/attachments/20070725/6b8212d4/attachment.html From ulises.best at yahoo.co.uk Thu Jul 26 13:04:09 2007 From: ulises.best at yahoo.co.uk (ulises best) Date: Thu, 26 Jul 2007 18:04:09 +0100 (BST) Subject: [drraw-users] access_log & error_log files for "404 error" In-Reply-To: <3a3ebfee0707260526vc7040a7p304aff931d99a667@mail.gmail.com> Message-ID: <327802.79504.qm@web26705.mail.ukl.yahoo.com> Hi Richard This is how my error_log file looks like : E [25/Jul/2007:13:36:35 -0500] Creating missing directory "/var/run/cups/certs" and the same type of lines appear with only the date and hour varying , and here is the access_log : localhost - - [25/Jul/2007:07:51:27 -0500] "POST /HTTP/1.1" 200 354 CUPS -Get-Classes client-error-not-found and as the previous one, the same type of lines which maybe happens because of the fact i was trying to access many times...have you got any idea what it would mean?? Regards Ulises Richard Viana wrote: Ulises its probably an apache config issue. what do your access and error logs state about this request? On 7/25/07, ulises best wrote: Hi mates I am quite ashemed to ask you for help with this matter(as it's said that it is extremely easy to install) but i couldn't install drraw in my linux box and obtained the following error when trying : http://localhost/nagios/cgi-bin/drraw Not found the requested url /nagios/cgi-bin/drraw was not found on this server I configured it and installed CGI.pm , rddtool with RRD perl module and nothing happens. I try to use drraw to view nagios performance data. Can anyone tell me why this message appear? i am using Apache 2.0.59 maybe the error could be that i am using the incorrect cgi file in the web server. i configured with the path /usr/lib/cgi-bin/drraw and changed permissions for the user 'nobody' for access the directory: chown -R nobody /var/lib/drraw Thanks in advance Ulises --------------------------------- Yahoo! Answers - Get better answers from someone who knows. Try it now. --------------------------------- Yahoo! Mail is the world's favourite email. Don't settle for less, sign up for your free account today. _______________________________________________ drraw-users mailing list drraw-users at taranis.org http://web.taranis.org/mailman/listinfo/drraw-users --------------------------------- Yahoo! Mail is the world's favourite email. Don't settle for less, sign up for your freeaccount today. --------------------------------- Yahoo! Mail is the world's favourite email. Don't settle for less, sign up for your freeaccount today. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://web.taranis.org/pipermail/drraw-users/attachments/20070726/5e269f15/attachment.html From gentoo at mats-gustafsson.se Fri Aug 24 06:00:44 2007 From: gentoo at mats-gustafsson.se (gentoo@mats-gustafsson.se) Date: Fri, 24 Aug 2007 12:00:44 +0200 (CEST) Subject: [drraw-users] How to get rid of SI units in legend? Message-ID: <65389.81.232.42.100.1187949644.squirrel@webmail1.one.com> When generating a graph showing my barometer readings the Y-axis shows values between 1017 and 1021 mBar, which is perfectly ok, but the legend shows the min, max, average and last values on the format 1.0175k. I'd rather like to have them in the format 1017.5. My guess is that it is because the format is specified with %s. When looking into the drraw.conf I found that it is possible to specify the number of digits to be used and also the number of decimals, But is there anyway to get rid of %s, or is there any other way to get the format I want? Thanks for your help and thanks for a very nice tool! Best regards, Mats From kalt at taranis.org Fri Aug 24 08:35:00 2007 From: kalt at taranis.org (Christophe Kalt) Date: Fri, 24 Aug 2007 08:35:00 -0400 Subject: [drraw-users] How to get rid of SI units in legend? In-Reply-To: <65389.81.232.42.100.1187949644.squirrel@webmail1.one.com> References: <65389.81.232.42.100.1187949644.squirrel@webmail1.one.com> Message-ID: <20070824123500.GB24843@bzz.taranis.org> i think this is a limitation from RRDtool which decides on its own how to do this. You should check the documentation for semi-obscure options. Christophe From greg at pando.com Fri Aug 31 14:52:39 2007 From: greg at pando.com (Gregory Davis) Date: Fri, 31 Aug 2007 14:52:39 -0400 (EDT) Subject: [drraw-users] Old Subject from Wiki Wishlist Message-ID: <1865961222.1121188586359273.JavaMail.root@dkny.pando.com> The feature request below can be added with the following change. -- Gregory Davis line 3207 and 3208 -values=>[@gt], -default=>param("${ds}_tType")), checkbox(-name=>"${ds}_tSTACK", -values=>'Y', -label=>'')), can be replaced with: -values=>[@gt], -default=>param("${ds}_tType")), br, checkbox(-name=>"${ds}_tSTACK", -values=>'Y', -label=>'<', -title=>'Whether to stack these elements on top of each other'), checkbox(-name=>"${ds}_STACK", -values=>'Y', -label=>'^', -title=>'Whether to stack these elements on top of the previous one')), from: http://web.taranis.org/pipermail/drraw-users/2007q1/000095.html On Jan 21, Greg Davis wrote: | Thanks for your help. About the regex stacking, i may not have been | clear on the question. What im trying to do is stack two template based | DSs on a single graph each template based DS would contain the data from | about 40 rrds for a total of 80 actual DSs. So lets say template DS A | would be 40 stacked areas in blue and green and Template DS B would be | another 40 stacked areas in red and yellow stacked on top of Template A. | When I add the second template DS it always begins the stack at 0. So | far I've had to Add template A and then use 40 regular DSs stacked on | top of template A's output. Is this the only way to do it? Which version of drraw are you using? With version 2.1.1, i'm able to do this by using "STACK" as the type in the drop down list, but this doesn't let one control what is stacked (it's based on the previous entry). With 2.2a2, you can't really do it. If you want a headache, read on. you could do this: Say your template DS A procudes a0, a1, ... a39 Set Element to "UN,0,$,IF" and Formula to "+" This gives a new CDEF a which is a0 + a1 + ... + a39 (skipping NaNs). Now, if you template DS A produces b0, b1, ... b39 Set the CDEF to "b0,$,EQ,a,0,IF,$,+" The problem is you "b0,$,EQ" compares the value of b0 with $ (which is b0 or b1 or ... b39), but you really want to compare the variable names, not their values. So while it may work most of the time, there'll be cases when it doesn't. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://web.taranis.org/pipermail/drraw-users/attachments/20070831/1f8d4e08/attachment.html From siiludus at gmail.com Fri Sep 7 14:22:03 2007 From: siiludus at gmail.com (Andi) Date: Fri, 7 Sep 2007 21:22:03 +0300 Subject: [drraw-users] Stacking DS templates Message-ID: <5648b5570709071122l63bd9bbbxa368f32128bbbd53@mail.gmail.com> Hello I am using version 2.2a4 and I unable to stack an DS template to previous data sources - first DS from the DS template is always with out STACK keyword. Is this a bug or feature? Also is there a way to reorder or sort the matches of DS template? This would be useful when stacking. PS! Why is google not indexing the drraw-users archives? -- Andi L?hmus siiludus [at] gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://web.taranis.org/pipermail/drraw-users/attachments/20070907/1ee321aa/attachment-0001.html From jkrauska at gmail.com Wed Sep 12 00:54:23 2007 From: jkrauska at gmail.com (Joel Krauska) Date: Tue, 11 Sep 2007 21:54:23 -0700 Subject: [drraw-users] drraw "cluster"? Message-ID: <46E770FF.1000003@gmail.com> I know everytime I do a feature request, I get a "Great, why don't you add that and send us a patch", but just to go on record, here's another idea. I've got drraw running in multiple datacenters showing rrds that are locally collected. (host load levels, disk usage trends, etc) This is fairly manageable for 2 datacenters, but with more, it would be nice for the separate instances of drraw to connect to eachother and exchange information. The simplest form would allow me to regex graphs and see data from a rrd at one site in the same template as an rrd at another site. (would require some multi-site drraw graph, template and dashboard mixing) The most complex form might allow me to combine rrd stats on the same graph with data sets from multiple locations. (would probably require some rrdtool rewrite) It's perhaps possible to get around this using a shared filesystem for rrds, but it would be nice to think about how drraw itself could accomplish this.... Anyone else have this problem/need? Thanks, Joel From jkrauska at gmail.com Wed Sep 12 00:56:09 2007 From: jkrauska at gmail.com (Joel Krauska) Date: Tue, 11 Sep 2007 21:56:09 -0700 Subject: [drraw-users] Bad link on webpage Message-ID: <46E77169.1010409@gmail.com> version 2.2a4 [2007/05/30] - http://web.taranis.org/drraw/dist/drraw-2.2a4.tar.gz Clicking on "version 2.2a4" grabs a3 --joel From vitroth at cmu.edu Wed Sep 12 11:01:23 2007 From: vitroth at cmu.edu (David Nolan) Date: Wed, 12 Sep 2007 11:01:23 -0400 Subject: [drraw-users] drraw "cluster"? In-Reply-To: <46E770FF.1000003@gmail.com> References: <46E770FF.1000003@gmail.com> Message-ID: <1cbe50830709120801p3a6489cq78268d167adc0524@mail.gmail.com> On 9/12/07, Joel Krauska wrote: > It's perhaps possible to get around this using a shared filesystem for rrds, but it would > be nice to think about how drraw itself could accomplish this.... > > Anyone else have this problem/need? > Absolutely, but I don't think this is a problem that drraw should try to solve. I'd like to see a more generalized solution for distributed RRD collection & graphing. I have got more devices/hosts I want to graph at my primary location then I can collect with a single machine, let alone graph... I'm doing the shared filesystem via NFS for graphing. Collection on two hosts, exported to a third for graphing. And I shudder every time I think about it. Tobias Oetiker (author of RRD) has been looking at adding an rrd daemon which handles the file IO, buffering and smoothing it, maybe even caching it for read access. Once that happens I think distributed RRD access should be fairly easy. On the other hand... Great, why don't you add that and send us a patch. :) -David From kalt at taranis.org Wed Sep 12 11:10:45 2007 From: kalt at taranis.org (Christophe Kalt) Date: Wed, 12 Sep 2007 11:10:45 -0400 Subject: [drraw-users] Bad link on webpage In-Reply-To: <46E77169.1010409@gmail.com> References: <46E77169.1010409@gmail.com> Message-ID: <20070912151045.GB28895@bzz.taranis.org> On Sep 11, Joel Krauska wrote: | version 2.2a4 [2007/05/30] - http://web.taranis.org/drraw/dist/drraw-2.2a4.tar.gz | | Clicking on "version 2.2a4" grabs a3 Thanks for the note, that's the problem with manually editing web pages. From vitroth+ at cmu.edu Wed Sep 12 16:23:21 2007 From: vitroth+ at cmu.edu (David Nolan) Date: Wed, 12 Sep 2007 16:23:21 -0400 Subject: [drraw-users] drraw "cluster"? In-Reply-To: <1cbe50830709120801p3a6489cq78268d167adc0524@mail.gmail.com> References: <46E770FF.1000003@gmail.com> <1cbe50830709120801p3a6489cq78268d167adc0524@mail.gmail.com> Message-ID: <1cbe50830709121323t9ddca51tc7d094652b71038b@mail.gmail.com> On 9/12/07, Joel Krauska wrote: > It's perhaps possible to get around this using a shared filesystem for rrds, but it would > be nice to think about how drraw itself could accomplish this.... > > Anyone else have this problem/need? > Absolutely, but I don't think this is a problem that drraw should try to solve. I'd like to see a more generalized solution for distributed RRD collection & graphing. I have got more devices/hosts I want to graph at my primary location then I can collect with a single machine, let alone graph... I'm doing the shared filesystem via NFS for graphing. Collection on two hosts, exported to a third for graphing. And I shudder every time I think about it. Tobias Oetiker (author of RRD) has been looking at adding an rrd daemon which handles the file IO, buffering and smoothing it, maybe even caching it for read access. Once that happens I think distributed RRD access should be fairly easy. On the other hand... Great, why don't you add that and send us a patch. :) -David From ericslaw at gmail.com Wed Sep 12 16:46:02 2007 From: ericslaw at gmail.com (Eric Law) Date: Wed, 12 Sep 2007 16:46:02 -0400 Subject: [drraw-users] drraw "cluster"? In-Reply-To: <1cbe50830709121323t9ddca51tc7d094652b71038b@mail.gmail.com> References: <46E770FF.1000003@gmail.com> <1cbe50830709120801p3a6489cq78268d167adc0524@mail.gmail.com> <1cbe50830709121323t9ddca51tc7d094652b71038b@mail.gmail.com> Message-ID: Joel Krauska, What you might be really asking is for loose integration across multiple disparate RRD installations? I'm using cricket and have long surpassed the ability to collect and store data using single hosts. I've even been pushing IO limits on my shared storage. Among the possible solutions (aside from just adding more/better/faster hardware), are: * reduce IO load on RRD updates (ie: keep more in memory for longer periods, or optimize for updates like the rrd daemon alluded to would do). * integrate disparate RRD respositories at the UI level (the closest solution applicable to the problem mentioned in the original email). Imagine a drraw URL call that would export the data in csv/xml, much like RRD XPORT does today.... Then imagine the drraw UI that would 'fetch' data (via the export feature) from a _remote_ drraw UI and merge results together in a single display. maybe the fetch is done via perl sockets/LWP within the CGI, or maybe the fetch is done via AJAX at the browser..... It could happen... it could work... it could exist before the RRD Daemon becomes a reality (unless Tobi is up to something I'm not aware of ;-) ) Anyone want to work on such a beast? -Eric On 9/12/07, David Nolan wrote: > > On 9/12/07, Joel Krauska wrote: > > It's perhaps possible to get around this using a shared filesystem for > rrds, but it would > > be nice to think about how drraw itself could accomplish this.... > > > > Anyone else have this problem/need? > > > > Absolutely, but I don't think this is a problem that drraw should try > to solve. I'd like to see a more generalized solution for distributed > RRD collection & graphing. I have got more devices/hosts I want to > graph at my primary location then I can collect with a single machine, > let alone graph... I'm doing the shared filesystem via NFS for > graphing. Collection on two hosts, exported to a third for graphing. > And I shudder every time I think about it. > > Tobias Oetiker (author of RRD) has been looking at adding an rrd > daemon which handles the file IO, buffering and smoothing it, maybe > even caching it for read access. Once that happens I think > distributed RRD access should be fairly easy. > > On the other hand... Great, why don't you add that and send us a > patch. :) > > -David > _______________________________________________ > drraw-users mailing list > drraw-users at taranis.org > http://web.taranis.org/mailman/listinfo/drraw-users > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://web.taranis.org/pipermail/drraw-users/attachments/20070912/bc3fd0c9/attachment.html From wernli at in2p3.fr Thu Sep 13 03:26:05 2007 From: wernli at in2p3.fr (Fabien Wernli) Date: Thu, 13 Sep 2007 09:26:05 +0200 Subject: [drraw-users] drraw "cluster"? In-Reply-To: References: <46E770FF.1000003@gmail.com> <1cbe50830709120801p3a6489cq78268d167adc0524@mail.gmail.com> <1cbe50830709121323t9ddca51tc7d094652b71038b@mail.gmail.com> Message-ID: <20070913072605.GB25953@ccswiss.in2p3.fr> On Wed, Sep 12, 2007 at 04:46:02PM -0400, Eric Law wrote: > * reduce IO load on RRD updates (ie: keep more in memory for longer periods, > or optimize for updates like the rrd daemon alluded to would do). I believe Tobias Oetiker has put quite some work on this matter: http://oss.oetiker.ch/rrdtool-trac/wiki/RRDtool13 > Imagine a drraw URL call that would export the data in csv/xml, much like > RRD XPORT does today.... you're not talking about exporting the actual rrd here are you? more like exporting the parts that need to be plotted, right? > Anyone want to work on such a beast? let's wait for rrdtool 1.3 first ;) From greg at pando.com Thu Sep 13 09:35:40 2007 From: greg at pando.com (Gregory Davis) Date: Thu, 13 Sep 2007 09:35:40 -0400 (EDT) Subject: [drraw-users] drraw "cluster"? In-Reply-To: <20070913072605.GB25953@ccswiss.in2p3.fr> Message-ID: <1813612210.26981189690540231.JavaMail.root@dkny.pando.com> :) I have the same issue. Take a look at the rrdpoller and rrdpollerd perl projects. Olivier Poitrey, the author has implemented some of the RRDs functions over rpc-xml. I don't think it will plug right in to DRRAW but it seems like a good place to start. -greg ----- Original Message ----- From: "Fabien Wernli" To: drraw-users at taranis.org Sent: Thursday, September 13, 2007 3:26:05 AM (GMT-0500) America/New_York Subject: Re: [drraw-users] drraw "cluster"? On Wed, Sep 12, 2007 at 04:46:02PM -0400, Eric Law wrote: > * reduce IO load on RRD updates (ie: keep more in memory for longer periods, > or optimize for updates like the rrd daemon alluded to would do). I believe Tobias Oetiker has put quite some work on this matter: http://oss.oetiker.ch/rrdtool-trac/wiki/RRDtool13 > Imagine a drraw URL call that would export the data in csv/xml, much like > RRD XPORT does today.... you're not talking about exporting the actual rrd here are you? more like exporting the parts that need to be plotted, right? > Anyone want to work on such a beast? let's wait for rrdtool 1.3 first ;) _______________________________________________ drraw-users mailing list drraw-users at taranis.org http://web.taranis.org/mailman/listinfo/drraw-users -------------- next part -------------- An HTML attachment was scrubbed... URL: http://web.taranis.org/pipermail/drraw-users/attachments/20070913/d0df8553/attachment-0001.html From ericslaw at gmail.com Thu Sep 13 10:36:36 2007 From: ericslaw at gmail.com (Eric Law) Date: Thu, 13 Sep 2007 10:36:36 -0400 Subject: [drraw-users] drraw "cluster"? In-Reply-To: <20070913072605.GB25953@ccswiss.in2p3.fr> References: <46E770FF.1000003@gmail.com> <1cbe50830709120801p3a6489cq78268d167adc0524@mail.gmail.com> <1cbe50830709121323t9ddca51tc7d094652b71038b@mail.gmail.com> <20070913072605.GB25953@ccswiss.in2p3.fr> Message-ID: RRD XPORT is not the same as DUMP... it's more like FETCH. You basically use it like GRAPH but the values get printed rather than drawn into a PNG. Think of it as getting your graph data in XML/CSV rather than PNG. If more RRD web interfaces did this, then I'd actually have access to network data that I could harvest this data and include it in my own graphs (OS and Business metrics) and the whole user experience would improve. oooOOooo 1.3 beta is out... [thanks!] the single wiki page doesn't mention much about the daemon and delayed writes, but references MMAP IO, which mentions writing only modified blocks to disk... I thought that was what it did before. I also see yet-another-change of libs, so the build instructions are ever changing.... oh well. My problem is that even with 4GB of RAM across several servers, I cannot hope to keep even most of the RRDs in memory, so while the in-core performance has improved, I still get hammered on IO. I've been resorting to pre-processing my RRD updates into batches sorted by which RRD they update, which improves performance notably. loose integration at the UI level is still desirable when you have several large remote datacenters and tons of datacenter specific metrics provide little value if that datacenter is down. Site generic data still needs to be consolidated, but can be done by fetch-n-store in a single central location, but the datacenter specific doesn't need to be central if the UI can still fetch information from a remote UI. I suppose this could be done without XPORT if you 'layer' transparent PNGs.... or just show datacenter specific data on the page with separate PNGs (I do that now too). On 9/13/07, Fabien Wernli wrote: > > On Wed, Sep 12, 2007 at 04:46:02PM -0400, Eric Law wrote: > > * reduce IO load on RRD updates (ie: keep more in memory for longer > periods, > > or optimize for updates like the rrd daemon alluded to would do). > > I believe Tobias Oetiker has put quite some work on this matter: > > http://oss.oetiker.ch/rrdtool-trac/wiki/RRDtool13 > > > Imagine a drraw URL call that would export the data in csv/xml, much > like > > RRD XPORT does today.... > > you're not talking about exporting the actual rrd here are you? more like > exporting the parts that need to be plotted, right? > > > Anyone want to work on such a beast? > > let's wait for rrdtool 1.3 first ;) > > _______________________________________________ > drraw-users mailing list > drraw-users at taranis.org > http://web.taranis.org/mailman/listinfo/drraw-users > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://web.taranis.org/pipermail/drraw-users/attachments/20070913/461254f9/attachment.html From aditya at grot.org Wed Sep 19 14:08:20 2007 From: aditya at grot.org (R.P. Aditya) Date: Wed, 19 Sep 2007 18:08:20 +0000 Subject: [drraw-users] printing last update and last graph generation time on graphs Message-ID: <20070919180820.GA75358@mighty.grot.org> I'd like to be able to print the following two things on every graph: - the value of rrdtool last $RRD (when the last sample was) - the value of "time" when the graph is generated I can't think of a nice way to do this using the existing template definitions, and before I try to patch drraw to do this, does anyone else have a hint or an existing method to do this with drraw? Thanks, Adi From kalt at taranis.org Thu Sep 20 08:04:01 2007 From: kalt at taranis.org (Christophe Kalt) Date: Thu, 20 Sep 2007 08:04:01 -0400 Subject: [drraw-users] printing last update and last graph generation time on graphs In-Reply-To: <20070919180820.GA75358@mighty.grot.org> References: <20070919180820.GA75358@mighty.grot.org> Message-ID: <20070920120401.GA28945@bzz.taranis.org> On Sep 19, R.P. Aditya wrote: | I'd like to be able to print the following two things on every graph: | | - the value of rrdtool last $RRD (when the last sample was) What would that be when you have multiple RRD files? Christophe From aditya at grot.org Thu Sep 20 08:56:34 2007 From: aditya at grot.org (R.P. Aditya) Date: Thu, 20 Sep 2007 12:56:34 +0000 Subject: [drraw-users] printing last update and last graph generation time on graphs In-Reply-To: <20070920120401.GA28945@bzz.taranis.org> References: <20070919180820.GA75358@mighty.grot.org> <20070920120401.GA28945@bzz.taranis.org> Message-ID: <20070920125634.GD86314@mighty.grot.org> On Thu, Sep 20, 2007 at 08:04:01AM -0400, Christophe Kalt wrote: > On Sep 19, R.P. Aditya wrote: > | I'd like to be able to print the following two things on every graph: > | > | - the value of rrdtool last $RRD (when the last sample was) > > What would that be when you have multiple RRD files? checkbox per RRD to print the time? or if that is too complex, not do it when there are multiple files... Thanks, Adi From kalt at taranis.org Thu Sep 20 11:51:30 2007 From: kalt at taranis.org (Christophe Kalt) Date: Thu, 20 Sep 2007 11:51:30 -0400 Subject: [drraw-users] printing last update and last graph generation time on graphs In-Reply-To: <20070920125634.GD86314@mighty.grot.org> References: <20070919180820.GA75358@mighty.grot.org> <20070920120401.GA28945@bzz.taranis.org> <20070920125634.GD86314@mighty.grot.org> Message-ID: <20070920155130.GB3641@bzz.taranis.org> On Sep 20, R.P. Aditya wrote: | > What would that be when you have multiple RRD files? | | checkbox per RRD to print the time? | | or if that is too complex, not do it when there are multiple files... i think flexibility is more useful, but i was wondering whether you'd thought of this and what you thought. i think it could probably be added under the additional GPRINTs section.