November 2008 - Xymon - Xymon mailman web

FW: [hobbit] FW: RRD data points - still struggling with RRD and graph distortion
by Mike.Wood＠kci1.com 18 Nov '08

18 Nov '08

Hi All, I'm still struggling with RRD, and how Xymon / Hobbit uses it to graph. I've modified /hobbitd/do_rrd.c where each RRD has 2304 datapoints (and make / make install / restart hobbit). I've also used rrdtool resize vmstat.rrd 1 GROW 1728 (then copy the resize.rrd file back over vmstat.rrd) to match the datpoints. However, my graphs don't show up correctly. They get truncated like in the attached .jpg files. Can anyone help an RRD thick-head get me over the hump? Thanks! Mike Wood UNIX System Administrator Kinetic Concepts Inc. 5751 NW Parkway San Antonio, TX, 78249 -----Original Message----- From: Stef Coene [mailto:stef.coene at docum.org] Sent: Monday, November 10, 2008 4:15 PM To: hobbit at hswn.dk Subject: Re: [hobbit] FW: RRD data points - need help with graph distortion On Monday 10 November 2008, Wood, Mike wrote: > Looks like my graphs didn't come through on the E-mail. Trying different > attachment type... > > > From: Wood, Mike [mailto:Mike.Wood at kci1.com] > Sent: Monday, November 10, 2008 9:06 AM > To: hobbit at hswn.dk > Subject: [hobbit] FW: RRD data points - need help with graph distortion > > Hi All, > > I asked this question last week, but maybe I wasn't clear enough as to what > I was asking... - > Management wants to avoid graph distortion on my 60 day CPU Util graph if > possible. I think that more datapoints would address the problem. The > first graph looks good (30 day graph). The 2nd one distorts what is really > going on (60 day graph - same server). Is there any way to make the 2nd > graph more granular? Thanks for looking! - I have a perl script to update all the rrd files: - it uses rrdtool resize <rrd file> GROW to add extra datapoints so the rrd has 5760 data points / rra (per default it has 756 data points) - it adds also MAX and MIN rra's, this is more tricky because I export the rrd to xml, change the xml info and revert back to an rrd file. Unfortunatly, I can not share this script. But with use RRDs ; use File::Find (); you can get info about the file and find them ;) Stef To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk ***************************************************************************** "CONFIDENTIALITY NOTICE: This transmission (including any accompanying attachments) is confidential, is intended only for the individual or entity named above, and is likely to contain privileged, proprietary and confidential information that is exempt from disclosure requests under applicable law. If you are not the intended recipient, you are hereby notified that any disclosure, copying, distribution, use of or reliance upon any of the information contained in this transmission is strictly prohibited. Any inadvertent or unauthorized disclosure shall not compromise or waive the confidentiality of this transmission or any applicable attorney-client privilege. If you have received this transmission in error, please immediately notify us at postmaster at kci1.com." Kinetic Concepts, Inc. ****************************************************************************** To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk ***************************************************************************** "CONFIDENTIALITY NOTICE: This transmission (including any accompanying attachments) is confidential, is intended only for the individual or entity named above, and is likely to contain privileged, proprietary and confidential information that is exempt from disclosure requests under applicable law. If you are not the intended recipient, you are hereby notified that any disclosure, copying, distribution, use of or reliance upon any of the information contained in this transmission is strictly prohibited. Any inadvertent or unauthorized disclosure shall not compromise or waive the confidentiality of this transmission or any applicable attorney-client privilege. If you have received this transmission in error, please immediately notify us at postmaster at kci1.com." Kinetic Concepts, Inc. ******************************************************************************

1 0

Many hosts on the same line in the table main page
by supervision_opt＠hotmail.fr 18 Nov '08

18 Nov '08

Hi everybody, i just want where did i have to modified to put 4 to 5 hosts on the same line in the main page table, i means by default it give me only 2 host by line so i would like to have it 4 to 5 so i wouldn't have to scroll to the end of the webpage to see all my hosts. i did it in hard in the index.html and the bb.html but a couple minute after the bbgen (i think) overwrite it by a new one. ideas ?? thank for the help. _________________________________________________________________ Sur Windows Live Ideas, découvrez en exclusivité de nouveaux services en ligne... si nouveaux qu'ils ne sont pas encore sortis officiellement sur le marché ! http://ideas.live.com

3 4

page.log: Argument list too long
by stef.coene＠docum.org 18 Nov '08

18 Nov '08

Hi, I'm getting errors in page.log: 2008-11-17 04:18:02 Could not launch paging script /home/users/hobbit/server/ext/alerting.pl: Argument list too long This is the entry hobbit-alerts.cfg: HOST=* SCRIPT /home/users/hobbit/server/ext/alerting.pl FORMAT=TEXT RECOVERED REPEAT=1440 DURATION>10 The error is a procs check that went red. This server a lot of processe, so the variable BBALPHAMSG is quite large. Any idea what could be wrong ? Stef

2 3

configure dependencies: supress triggered actions, when another action is triggered before?
by lists＠nanl.de 17 Nov '08

17 Nov '08

Hey, first - thanks for such an easy to use monitoring-tool! Anyway, I've a a little issue which is really bothering me. We have physical servers and virtual servers (vmware / xen) installed on them. When a physical server is not reachable (reset, blackout, etc.) all virtual-servers and services installed on them are not reacheable, too. In such a case I'm getting over 50 SMS and even more emails, because we're monitoring each service running in our network. Is there a way to configure dependencies? Such, "ping on virtual-server x, ping on virtual-server y, ssh on virtual-server z" depends on "ping physical-server foo", so that if the ping-test on the physical-server <foo> fails, all other actions, which depend on that, are suppressed? Thanks a lot in advance! Have a nice day, mirko

2 1

gaps in hobbit graphs
by Mike.Wood＠kci1.com 17 Nov '08

17 Nov '08

Hi All, Any idea what would / could cause gaps in graphing like in the attached picture? It happens periodically on various systems, but only on a few particular graphs (like CPU utilization). Thanks for looking! Mike Wood UNIX System Administrator Kinetic Concepts Inc. 5751 NW Parkway San Antonio, TX, 78249 E-mail: mike.wood at kci1.com Office: (210) 255-6382 Mobile: (210) 825-5134 ***************************************************************************** "CONFIDENTIALITY NOTICE: This transmission (including any accompanying attachments) is confidential, is intended only for the individual or entity named above, and is likely to contain privileged, proprietary and confidential information that is exempt from disclosure requests under applicable law. If you are not the intended recipient, you are hereby notified that any disclosure, copying, distribution, use of or reliance upon any of the information contained in this transmission is strictly prohibited. Any inadvertent or unauthorized disclosure shall not compromise or waive the confidentiality of this transmission or any applicable attorney-client privilege. If you have received this transmission in error, please immediately notify us at postmaster at kci1.com." Kinetic Concepts, Inc. ******************************************************************************

4 4

afs_fsmon ext-script
by martin.flemming＠desy.de 17 Nov '08

17 Nov '08

Hi, Tracy et all ... Nice that you are on the list again, dosen't you disappear from it for one or two years .... So i take another chance to ask you, for your configuration of the various graphs of your afs_fsmon-script ... > It was also simple to have all the numbers rxstats generates fed into an > rrd, so pretty graphs are available, although calls waiting for a thread is > the only one monitored Unfortunatley not for me, :-( Thanks & Cheers, Martin On Fri, 11 Jul 2008, Martin Flemming wrote: > > Hi, Tracy et all ! > > .. don't know, if you are still on the list, > that's why you explicit on the Cc ... > > I use your afs_fsmon monitoring-script > > ftp://ftp.iastate.edu/pub/bb/ext/afs_fsmon > > to check our AFS-Fileserver with much success .. :-) > > But i want also to have the output not only as status-messages > besides as Grpahic and have tested much but without success ... :-( > > And now i read (not the first time, but never recognize) under > > http://workshop.openafs.org/afsbpw06/talks/gendalia/gendalia-web.html > > > that you've a configuration to do that ! > > Monitoring > > * We use hobbit for monitoring > * When the http connections on the various web servers start going red en > masse... it's usually an afs server problem > * So I added "rxdebug $HOST 7000 -noconn -rxstats" to my afs monitoring > script, and started alerting on too many calls waiting for a thread. > * It was also simple to have all the numbers rxstats generates fed into an > rrd, so pretty graphs are available, although calls waiting for a thread is > the only one monitored > > Can i/we (hobbit-community) share it with you ?! > > .. thinking about hobbitgraph.cfg and maybe NCV-entries in hobbitserver.cfg .. anymore ? > > Thanks & Cheers > > Martin > > P.S.: > Does anybody got further ext-scripts for checking AFS > or monitor the performence ? > > It will also welcome ! :-) > > ______________________________________________________ > Martin Flemming > DESY / IT office : Building 2b / 008a > Notkestr. 85 phone : 040 - 8998 - 4667 > 22603 Hamburg mail : martin.flemming at desy.de > ______________________________________________________ >

3 6

[hobbit] Hobbit: Run Scripts
by sigurdur＠raforninn.is 14 Nov '08

14 Nov '08

Hi. Why not use a cron job then? Afaik, hobbit is not designed to be a deployment server. Regards, Sigurður Guðbrandsson. ________________________________ From: Sanu Mathew [mailto:mathew.sanu at gmail.com] Sent: 4. nóvember 2008 20:47 To: hobbit at hswn.dk Subject: Re: [hobbit] Hobbit: Run Scripts Hi Gary, This script needs to be run on all machines monitored by hobbit. The script can be placed on all servers (a tedious job though - unless we cant run the script from the hobbit server to the clients). The script gives a snapshot of the server, which helps us in case of rebuilds. Its not in response to a monitored condition, but is a generic script which gives the snapshot, and would run regardless of any specific condition. Regards, Sanu On Wed, Nov 5, 2008 at 12:45 AM, Gary Baluha <gumby3203 at gmail.com> wrote: Is the script to be run only on the machines monitored by Hobbit, and is it already on those machines? What is the script trying to do? If the script is in response to a monitored condition, the best way to accomplish that would be different than if it just runs at a specific time regardless of any specific condition, which is why I ask. On Tue, Nov 4, 2008 at 3:08 PM, Sanu Mathew <mathew.sanu at gmail.com> wrote: Hi, I am looking at being able to run a global script on all servers that are monitored by Hobbit. How can i accomplish this task? (The script is created by us and I am looking at incorporating this script to run at a specific time weekly) Any help/suggestion would be great. Regards, Sanu

34 62

[hobbit] Hobbit - Alert
by bewhite＠fellowes.com 14 Nov '08

14 Nov '08

I use the following in my hoobit-clients.cfg file to trap EMS messages: LOG %/var/adm/syslog/syslog.log "%(?-i) EMS" color=red I get paged whenever an EMS message appears in any of my HPUX systems syslog.log file. One question, have you configured hobbit (xymon) to send alerts when a yellow condition is encountered? ......Bruce _____ From: Sanu Mathew [mailto:mathew.sanu at gmail.com] Sent: Monday, October 20, 2008 4:24 PM To: hobbit at hswn.dk Subject: Re: [hobbit] Hobbit - Alert Yes, the file is readable globally. On Tue, Oct 21, 2008 at 2:23 AM, Larry Barber <lebarber at gmail.com <mailto:lebarber at gmail.com> > wrote: Can your Hobbit client read the file in question? Thanks, Larry Barber On Mon, Oct 20, 2008 at 3:45 PM, Sanu Mathew <mathew.sanu at gmail.com <mailto:mathew.sanu at gmail.com> > wrote: Hi, I would like to monitor HP servers for EMS alerts, and i have added the below entry in hobbit-clients.cfg. CLASS=hp-ux LOG /var/adm/syslog/syslog.log %EMS COLOR=yellow LOG /var/adm/syslog/syslog.log vmunix COLOR=yellow IGNORE=vx_nospace A similar entry is also there in client-local.cfg file as well as listed below. [hp-ux] log:/var/adm/syslog/syslog.log:102400 Is there anything that i have missed, as i do not receive an alert? Any suggestion at the earliest would be really really helpful, thanks in advance. Regards, Sanu ---------------------------------------------------- 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. ----------------------------------------------------

1 0

HTTP error 0
by Aaron.Li＠ehealth-china.com 14 Nov '08

14 Nov '08

Hi all When hobbit monitors the https site, it always gets the error " HTTP error 0" like below, I appreciate a lot if anyone can give the solution, thanks https://x.x.x.x/ - HTTP error 0

5 12

Issue trying to monitor VIP's
by jr＠e-integration.net 14 Nov '08

14 Nov '08

I'm having an issue with my hobbit server monitoring virtual IP's. When I click on the info diamond it causes an internal server error. I'm not sure why. The two hosts that are part of the VIP have hobbit clients running on them and their info diamonds are fine as all of their other tests. The VIP only has three columns conn, info, and trends. What's weird is only a couple of the VIPs aren't working correctly, the others seem to be working fine. I just want to fix the issue so that clicking the info diamond won't cause an internal server error. Here's what I've tried so far: On hobbit server: hobbit 4.2.0 running on RHEL 4 update 5 Remove VIP from bb-host ./bb localhost "drop name-of-vip" Wait like 20 minutes and re-add This doesn't work. Example of working info request via logs: http://netmon4/hobbit-cgi/bb-hostsvc.sh?HOST=name-of-vip&SERVICE=info 10.106.6.55 - - [29/Oct/2008:11:33:51 -0700] "GET /hobbit-cgi/bb-hostsvc.sh?HOST=name-of-vip&SERVICE=info HTTP/1.1" 200 11239 Example of failed info request: http://netmon4/hobbit-cgi/bb-hostsvc.sh?HOST=name-of-vip&SERVICE=info [Wed Oct 29 11:38:19 2008] [error] [client 10.106.6.55] Premature end of script headers: bb-hostsvc.sh Further investigation: cat /opt/hobbit/cgi-bin/bb-hostsvc.sh #!/bin/sh # This is a wrapper for the Hobbit hobbitsvc.cgi script . /opt/hobbit/server/etc/hobbitcgi.cfg exec /opt/hobbit/server/bin/hobbitsvc.cgi $CGI_SVC_OPTS This CGI is a 32-bit binary and I would need to look at source to troubleshoot this further. ..Hoping to avoid that. Any suggestions would be greatly appreciated. Thanks.

1 0