July 2007 - Xymon - Xymon mailman web

DROP command
by hobbitdoubt＠gmail.com 06 Jul '07

06 Jul '07

Hello, I want to remove some tests that report clear colour from bb.html. I have used this command ~/server/bin/bb 127.0.0.1 "drop HOSTNAME column" but after some minutes the columns turn up again. How can I do this permanent? Thanks.

2 1

msgs file with html code
by vadud3＠gmail.com 05 Jul '07

05 Jul '07

Hi When I receive alerts related to messages file it comes with html code in my mail box like this. &red Critical entries in <a href="/hobbit-cgi/bb-hostsvc.sh?CLIENT=hobbit.client.com&SECTION=msgs:/var/adm/mess +ages">/var/adm/messages</a> &red Aug 17 15:39:55 hobbit.client.com WARNING: [ID 702911 user.error] this is a test message Full log <a href="/hobbit-cgi/bb-hostsvc.sh?CLIENT=hobbit.client.com&SECTION=msgs:/var/adm/messages">/var/adm/m +essages</a> Aug 17 15:39:55 hobbit.client.com WARNING: [ID 702911 user.error] this is a test message I am using hobbit client 4.2.0 and hobbit server 4.2.0. I sure did not play with any html code to make the output like this. Also I saw an email about keeping the full log link in the web gui for easy troubleshooting. But is that also apply to email. In otherwords can I just receive alert related lines and not the full log underneath. OR do I need to define SMS in hobbit-alerts.cfg page to achieve that? Thanks -- Asif Iqbal PGP Key: 0xE62693C5 KeyServer: pgp.mit.edu

2 3

Can't get NetApp CPU to graph
by jwprkr＠gmail.com 05 Jul '07

05 Jul '07

Hello, I have a problem that just doesn't make sense. I am monitoring the CPU of a Network Appliance server using SNMP queries from the Hobbit server. It's essentially what I am doing on an old BB server running LaRRD. I have added custom graphs to my BB server and they worked fine, so I have a basic understanding how the LaRRD parsing works. The CPU status message looks fine, but Hobbit is just not picking up the data and adding it to RRD, and therefore not graphing. I even tried copying the rrd file over from the BB server to see if that would help, but no dice. It graphs up to the point where I copied it from BB and stops. Other devices where I'm monitoring CPU the status message is identical, and it's graphing fine. Is there a way to debug RRD to see what's going wrong? I get very few useful messages in the RRD log files. Thanks in Advance, Jeff -- Jeff Parker jwprkr at gmail dot com

2 1

Availability Report anomally ....
by Geoff.Swavley＠aus.fujixerox.com 04 Jul '07

04 Jul '07

Hi All, We're using Hobbit to pretty much monitor most things - and it works fine. One of our sections likes to keep track of the availablility of their web page and they have been using hobbit for this and are happy with what it gives them, except for an anomally from last month. When the availability report is run for last month I get a green diamond (presuming this to mean that it's 100%). We know it's not 100% by simply clicking on the http diamond and performing a history of the service. Why doesn't the monthly report tell us what is obviously in the history??? It might be a case of resolution as the downtimes have only been 15seconds here , 50 seconds there, 25 seconds. Here are the screen shots to show this: http://tinyurl.com/yr2kz6 http://tinyurl.com/39q56p thanks Geoff :-) <http://www.fujixerox.com.au/> Geoff Swavley Infrastructure Specialist Business Services Fuji Xerox Australia Pty Limited Level 1, 101 Waterloo Road MACQUARIE PARK NSW 2113 AUSTRALIA Phone Fax email web (02) 9856 5468 (02) 9856 5003 Geoff.Swavley at aus.fujixerox.com <mailto:Geoff.Swavley at aus.fujixerox.com> www.FujiXerox.com.au <http://www.fujixerox.com.au/> This email (including any attachments) is confidential and may contain personal information within the Privacy Act 1988 (Cth). Fuji Xerox Australia make no warranties that it is free of viruses or has not been subject to unauthorised interference. If you have received it in error, please understand that confidentiality is not waived, please inform the sender by reply email, and delete it (and all copies) from your system.

3 3

Monitoring and ESX host
by a_stranberg＠hotmail.com 04 Jul '07

04 Jul '07

Hello all, I am searching for folks that are successfully monitoring aVMware ESX3 host with either the hobbit client or BB client? I am inthe unfortunate situation of not having a test environment to test withand would appreciate getting a head start on others exeprience. Anyspecific RPM for hobbit that is known to work on the ESX3 host, detailson dependency packages that were required would be most welcome.Thanks-Aaron _________________________________________________________________ With Windows Live Hotmail, you can personalize your inbox with your favorite color. www.windowslive-hotmail.com/learnmore/personalize.html?locale=en-us&ocid=TX…

6 7

[hobbit] Re: Help - NCV graph issue
by bewhite＠fellowes.com 03 Jul '07

03 Jul '07

Hi all, I have yet a different problem with an NCV graph. I have a script which monitors are UPS. In the script it captures the current load. I have set-up the hobbitserver.cfg file and hobbitgraph.cfg files based on the examples and it is storing the data and producing the graph, but the numbers don't match. When I do an rrdtool dump I see the current data point is ok (like 29% and I see that 29 is the current datapoint). When it graphs it, it shows it as 8. I have tried playing with the data types and that doesn't help. Any ideas? Thanks ahead of time, Bruce -----Original Message----- From: Charles Jones [mailto:jonescr at cisco.com] Sent: Friday, June 29, 2007 3:23 PM To: hobbit at hswn.dk Subject: [hobbit] Re: Help - NCV graph issue Wow. YET AGAIN I send an email to the list with a problem, and figure it out immediately afterwards. The issue was using the wrong def name in the GPRINT. I know it looks like I'm a total idiot - I swear I looked at that section of hobbitgraph.cfg at least a dozen times and didn't notice it. Ught. Thank goodness it's Friday. -Charles To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk ---------------------------------------------------- Note: The information contained in this message may be privileged and confidential and protected from disclosure. If the reader of this message is not the intended recipient, or an employee or agent responsible for delivering this message to the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you have received this communication in error, please notify us immediately by replying to the message and deleting it from your computer. Thank you. Fellowes, Inc. ----------------------------------------------------

5 6

graphind bonded interfaces
by mlist＠cmcflex.com 03 Jul '07

03 Jul '07

I'm trying to add a graph to the trends column in which I graph my bond0 interfaces. I don't want the graph to have the eth interfaces. There is currently an rrd database (created from the hobbit client on each machine). The database file name is: ifstat.bond0.rrd. Using those names I edited the hobbitgraph.cfg file and included at the bottom this section: [bond0] TITLE bond0 bits YAXIS bits / second DEF:inbits=ifstat.bond0.rrd:bytesReceived:DERIVE DEF:outbits=ifstat.bond0.rrd:bytesSent:DERIVE AREA:inbits#00CCCC:Bond0 GPRINT:inbits:LAST: \: %5.1lf (cur) GPRINT:inbits:MAX: \: %5.1lf (max) GPRINT:inbits:MIN: \: %5.1lf (min) GPRINT:inbits:AVERAGE: \: %5.1lf (avg)\n GPRINT:outbits:LAST: \: %5.1lf (cur) GPRINT:outbits:MAX: \: %5.1lf (max) GPRINT:outbits:MIN: \: %5.1lf (min) GPRINT:outbits:AVERAGE: \: %5.1lf (avg)\n LINE2:outbits#FF0000 and in hobbitserver.cfg I appended bond0 to TEST2RRD. However, I'm not getting the graph generated on the trends page - not even a dead link. I know I've missed something simple - but not sure what. I don't think I'm supposed to append "bond0" to my -extra-tests= lines in hobbitlaunch.cfg. Is that right (although I did append it just to see what happens - and nothing). Any ideas? Matt

1 1

Estimating hardware needs for Hobbit server
by haertig＠avaya.com 03 Jul '07

03 Jul '07

Hi - I'm currently running my Hobbit server/Apache on Linux-based Dell Dimension desktop with the following: # grep MemTotal /proc/meminfo MemTotal: 2075380 kB # grep name /proc/cpuinfo model name : Intel(R) Pentium(R) 4 CPU 3.40GHz # It's not even breathing hard. It's just sitting there idle for the most part. (CPU load 0.02, physical mem used 35%, real mem 12%, nothing ever swapped) I found some older Sun hardware in the lab that I'm considering using as a replacement, but I don't know how it's specs would compare. I'm not familiar with Sun/Sparc hardware all that much. Can anyone comment on if what I've scrounged up might be adequate? At most, I'd expect to eventually monitor several dozen servers, but probably no more than 100. Ultra 5 Ultra 10 (350 MGz, 512 MB RAM) E420R (2 - 450 MHz CPUs, 2 GB RAM) Thanks!

4 3

[hobbit] Estimating hardware needs for Hobbit server
by bewhite＠fellowes.com 03 Jul '07

03 Jul '07

If you put hobbit on that E420R, you'll have more then enough to monitor your environment. I have my hobbit server on an HP- RP5400 2x540 MHz - 6 GB ram. I'm monitoring 311 "hosts" (90 printers, 12 HPUX/Linux machines and the windows servers). The hobbit server has been averaging 54% idle and a .7 load over the last 4 months. Most of the monitoring on the windows servers is not via the hobbit client but via SNMP walking the HP Insight manager MIB on the windows server. So the hobbit server is doing most of that work. .....Bruce -----Original Message----- From: Haertig, David F (Dave) [mailto:haertig at avaya.com] Sent: Monday, July 02, 2007 5:50 PM To: hobbit at hswn.dk Subject: [hobbit] Estimating hardware needs for Hobbit server Hi - I'm currently running my Hobbit server/Apache on Linux-based Dell Dimension desktop with the following: # grep MemTotal /proc/meminfo MemTotal: 2075380 kB # grep name /proc/cpuinfo model name : Intel(R) Pentium(R) 4 CPU 3.40GHz # It's not even breathing hard. It's just sitting there idle for the most part. (CPU load 0.02, physical mem used 35%, real mem 12%, nothing ever swapped) I found some older Sun hardware in the lab that I'm considering using as a replacement, but I don't know how it's specs would compare. I'm not familiar with Sun/Sparc hardware all that much. Can anyone comment on if what I've scrounged up might be adequate? At most, I'd expect to eventually monitor several dozen servers, but probably no more than 100. Ultra 5 Ultra 10 (350 MGz, 512 MB RAM) E420R (2 - 450 MHz CPUs, 2 GB RAM) Thanks! ---------------------------------------------------- Note: The information contained in this message may be privileged and confidential and protected from disclosure. If the reader of this message is not the intended recipient, or an employee or agent responsible for delivering this message to the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you have received this communication in error, please notify us immediately by replying to the message and deleting it from your computer. Thank you. Fellowes, Inc. ----------------------------------------------------

2 1

Monitoring and ESX host
by jdustin＠usm.maine.edu 03 Jul '07

03 Jul '07

>>> On 7/2/2007 at 12:03 PM, Aaron Stranberg <a_stranberg at hotmail.com> wrote: > Hello all, I am searching for folks that are successfully monitoring > aVMware ESX3 host with either the hobbit client or BB client? I am inthe > unfortunate situation of not having a test environment to test withand would > appreciate getting a head start on others exeprience. Anyspecific RPM for > hobbit that is known to work on the ESX3 host, detailson dependency packages > that were required would be most welcome.Thanks-Aaron Here is a snippet of the code I run on ESX3: my $esxtop = "sudo /usr/bin/esxtop -b -d $interval"; open(CMD,"$esxtop |") || die "error - could not open $esxtop $!\n"; my $count = 0; while( <CMD> ) { chomp; s/\"//g; my (@row) = split/\,/; $count++; if ( $count <= 2 ) { next; } # skip 1st few results, just titles and 100% values my $util = int($row[9]); # round to INT, because RRD does not like floating-point &SendHobbit($util); } This is a PERL script, called by the "standard" Hobbit client running inside the service console. The only ESX-specific parameter I am watching is overall CPU utilization. The subroutine SendHobbit "prints" the data to tcp/1984 with the following line: print SOCK "status $MACHINE.pcpu $COLOR $datenow\npcpuBusy : $util\n"; I was able to compile the Hobbit client on a RedHat box, then ZIP the client and manually install on ESX. Good luck! -- -- Jon Dustin - Network Specialist University of Southern Maine Portland, ME

2 1