January 2007 - Xymon - Xymon mailman web

NCV, devmon and rrd and graphing problem
by mdingeldey＠sbcglobal.net 09 Jan '07

09 Jan '07

I am currently running hobbit 4.2.0/all-in-one patch on a Compaq DL360 using FC6. At the present time, I am attempting to use devmon to feed data to hobbit and create a graph form the resulting information. The only test I am running is the if_load test; I am running it against a Cisco 2600 using the base 2600 template. I am running devmon 0.2.2 At the current time, NCV creates and updates the if_load.rrd file. The if_load column appears on the hobbit page, and the if_load detail displays a table with the correct flow rates on the interfaces. However, I am unable to create any graphs of the data. At this point, I am assuming the devmon is feeding the correct data to ncv. My test router has five interfaces (loopback, one etherent, a serial and two frame-relay subinterfaces) so how do I verify that the resulting if_load.rrd file is in fact correct? >From what I have checked, everything looks like it should work. I'd like to go farther with verification on the rrd side, but the if_load.rrd file structure (and the tableborder) has me scratching my head. Could the bottom line be that I didn't define the data type for NCV correctly? The reason I ask is that the only time anything did not complain was when I defined the data type as a counter. I used "*:Counter" because I did not know what the names were going to be. Also, could the if_load file names be part of the problem? Any help or insight would be greatly appreciated! TIA! I currently have the following: TEST2RRD = "blah,blah,blah.....,if_load=ncv) NCV_if_load="*:Counter" GRAPH="blah,blah,blah,...,if_load) [if_load] FNPATTERN if_load.(.*).rrd TITLE load test YAXIS Percent DEF:loadin (at) RRDIDX@= (at) RRDFN@:loadin:AVERAGE DEF:loadout (at) RRDIDX@= (at) RRDFN@:loadout:AVERAGE LINE2:loadin (at) RRDIDX@# (at) COLOR@:@RRDPARAM@ load in GPRINT:loadin (at) RRDIDX@:LAST: \: %5.1lf (cur) GPRINT:loadin (at) RRDIDX@:MAX: \: %5.1lf (max) GPRINT:loadin (at) RRDIDX@:MIN: \: %5.1lf (min) GPRINT:loadin (at) RRDIDX@:AVERAGE: \: %5.1lf (avg)\n LINE2:loadout (at) RRDIDX@# (at) COLOR@:@RRDPARAM@ load out GPRINT:loadout (at) RRDIDX@:LAST: \: %5.1lf (cur) GPRINT:loadout (at) RRDIDX@:MAX: \: %5.1lf (max) GPRINT:loadout (at) RRDIDX@:MIN: \: %5.1lf (min) >From the dump of if_load.rrd: <rrd> <version> 0003 </version> <step> 300 </step>  <lastupdate> 1168130749 </lastupdate> <!!-- 2007-01-06 19:45:49 EST --> <ds> <name> tableborder </name> <type> COUNTER </type> <minimal_heartbeat> 600 </minimal_heartbeat> <min> 0.0000000000e+00 </min> <max> NaN </max>  <last_ds> 1 </last_ds> <value> 0.0000000000e+00 </value> <unknown_sec> 0 </unknown_sec> </ds>  <rra> <cf> AVERAGE </cf> <pdp_per_row> 1 </pdp_per_row>  <params> <xff> 5.0000000000e-01 </xff> </params> <cdp_prep> <ds> <primary_value> 0.0000000000e+00 </primary_value> <secondary_value> 0.0000000000e+00 </secondary_value> <value> NaN </value> <unknown_datapoints> 0 </unknown_datapoints> </ds> </cdp_prep> <database> <SNIP>

2 2

[hobbit] hobbit on Solaris 10 non-global Zone
by mndud3＠gmail.com 09 Jan '07

09 Jan '07

Install the client on the non-global zone. I haven't tried it with resource management on the zones (aka containers), but with a simple non-global zone, you get duplicate monitoring of cpu and memory usage. Disk usage is also a duplicate unless you have given the non-global zone control of any devices. ... and prtdiag doesn't report anything in the non-global zone, so it's a good idea to watch the global zone even if you don't have any apps running there. Everything else (ports, procs, files, etc) works just fine. You could try to watch everything from the client in the global zone, but I suspect there would be issues with that. In particular, if you want to watch sshd, for instance. The client on the global zone would see sshd for each of the zones just as "sshd". If one zone wasn't running any sshd's, but the others had extra sshd's running, hobbit would be none the wiser. -Bill. -----Original Message----- From: Nicolas Dorfsman [mailto:ndorfsman at gmail.com] Sent: Tuesday, January 09, 2007 9:00 AM To: hobbit at hswn.dk Subject: [hobbit] hobbit on Solaris 10 non-global Zone Hi, I have to monitor a Sol10 non-global Zone. Any hint ? TIA Nicolas To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk

1 0

[hobbit] Antwort: [hobbit] RE: Antwort: [hobbit] RE: V0.2.0 of adminscripts
by michael.nemeth＠lmco.com 09 Jan '07

09 Jan '07

> perl -e '$date=time(); print "$date\n";' even work with perl 4 on my hp-ux 11 system!!!!! Works on the windoze system with perl too. I though it was a basic perl fuction! -- +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ | _p_ Mike Nemeth | ___| |_____ email(w) michael.nemeth at lmco.com Work: 856 359-1425 |><___________) | Home Page:http://www.geocities.com/mjnemeth/ | Work Page:http://scorpion/~nemethm/ | Work Page:http://ortfsb/~mnemeth/ |++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

1 0

Antwort: [hobbit] RE: Antwort: [hobbit] RE: V0.2.0 of adminscripts
by michael.nemeth＠lmco.com 09 Jan '07

09 Jan '07

True but I believe that solaris comes with perl and old one but one should handle the date stuff. Atlest the one system I found here that does have an update perl reports sun pkg SUNWpl5u, Can any one confirm this? And can anyone supply the code to do a timestamp like the GNU date +%s ? a one liner hopefull (like perl -e '$date=localtime(time); print "$date \n" ;' ) >Yes but than must perl available. >I think i include an switch in the next version which test what is >available and what not. -- +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ | _p_ Mike Nemeth | ___| |_____ email(w) michael.nemeth at lmco.com Work: 856 359-1425 |><___________) | Home Page:http://www.geocities.com/mjnemeth/ | Work Page:http://scorpion/~nemethm/ | Work Page:http://ortfsb/~mnemeth/ |++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

3 3

snapshot 070109
by lars.ebeling＠leopg9.no-ip.org 09 Jan '07

09 Jan '07

To solve this problem on HP-UX 11.11 add #include <limits.h> to holidays.c gcc -g -O -D_REENTRANT -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -DHPUX -I. -I /home/hobbit/snapshot/include -I/usr/local/include -I. -I../include -c -o hol idays.o holidays.c holidays.c: In function `load_holidays': holidays.c:162: error: `PATH_MAX' undeclared (first use in this function) holidays.c:162: error: (Each undeclared identifier is reported only once holidays.c:162: error: for each function it appears in.) make[1]: *** [holidays.o] Error 1 make[1]: Leaving directory `/home/hobbit/snapshot/lib' make: *** [lib-build] Error 2 Regards Lars Ebeling http://leopg9.no-ip.org Hobbithobbyist "It is better to keep your mouth shut and appear stupid than to open it and remove all doubt." -- Mark Twain

1 0

[hobbit] Antwort: [hobbit] RE: Antwort: [hobbit] RE: V0.2.0 of adminscripts
by Johann.Eggers＠teleatlas.com 09 Jan '07

09 Jan '07

> -----Original Message----- > From: Michael Nemeth [mailto:michael.nemeth at lmco.com] > Sent: Dienstag, 9. Januar 2007 14:34 > To: hobbit > Subject: [hobbit] Antwort: [hobbit] RE: Antwort: [hobbit] RE: V0.2.0 of > adminscripts > > True but I believe that solaris comes with perl and old one but one > should handle the date stuff. > Atlest the one system I found here that does have an update perl reports > sun pkg SUNWpl5u, > Can any one confirm this? And can anyone supply the code to do a > timestamp like the GNU date +%s ? > a one liner hopefull (like perl -e '$date=localtime(time); print "$date > \n" ;' ) > Instead of using localtime() just use time()... perl -e '$date=time(); print "$date\n";' 1168350264 Johann > > >Yes but than must perl available. > > >I think i include an switch in the next version which test what is > >available and what not. > > -- > +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ > | _p_ Mike Nemeth > | ___| |_____ email(w) michael.nemeth at lmco.com Work: 856 359-1425 > |><___________) | Home > Page:http://www.geocities.com/mjnemeth/ > | Work Page:http://scorpion/~nemethm/ > | Work Page:http://ortfsb/~mnemeth/ > |++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ > > > To unsubscribe from the hobbit list, send an e-mail to > hobbit-unsubscribe at hswn.dk >

1 0

Antwort: [hobbit] RE: V0.2.0 of adminscripts
by michael.nemeth＠lmco.com 09 Jan '07

09 Jan '07

No not really. Gnu date which is in coreutils but I think can be done in perl maybe as a one liner or if not a short perlscript. >Hi, >sorry i love solaris ;-). >is there no possibility to get an timestamp? >By >Stefan -- +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ | _p_ Mike Nemeth | ___| |_____ email(w) michael.nemeth at lmco.com Work: 856 359-1425 |><___________) | Home Page:http://www.geocities.com/mjnemeth/ | Work Page:http://scorpion/~nemethm/ | Work Page:http://ortfsb/~mnemeth/ |++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

2 1

Server multiple times,what is wrong?
by stefan.freisler＠aspect-online.de 09 Jan '07

09 Jan '07

4 6

Disable netstat test
by toaster＠imber.com 08 Jan '07

08 Jan '07

Greet, Just had a new issue pop up and looking for some suggestions/help. We had to enable a firewall to do BGP routing for a new Colo Install. Once the firewall received the routes from the edge router hobbit went insane. What does insane mean? Well, it no longer sent updates the hobbit server. Also, if I tried to restart hobbit, hobbit consumes 100% of the CPU. I get the following in the logs : 2007-01-08 18:02:28 Write error while sending message to bbd at 172.16.2.10:1984 2007-01-08 18:02:28 Whoops ! bb failed to send message - write error I have already allocated more memory on the FreeBSD box. What is most likely happening is hobbit runs the netstat command and expects to get a few routes not 250,000 of them! And this is where it is hanging. Any idea how to disable the netstat test in the hobbit client? Or perhaps another way around this? Thanks, Lee

2 2

SMS alerts
by mprice＠hst.nasa.gov 08 Jan '07

08 Jan '07

Hello, Has anyone come across any easy SMS alerting scripts and free services to dial out to them to send the SMS alerts. thanks, michael -- Michael A. Price Performance Network Engineering NASA/GSFC Code 440.8/LMB Greenbelt, Maryland 20770 Phone: 240-684-1356 Cell: 410-507-7476 e-mail: mprice at hst.nasa.gov ----------------------------------------------

1 0