February 2006 - Xymon - Xymon mailman web

Graph appears only on the test column but nor on the trends page
by christian.maxeiner＠usd.de 06 Feb '06

06 Feb '06

Hi all, I am using hobbit 4.1.2p1 and I monitor some MS IIS webservers with "iishealth-Script from deadcat. When I click on the "iishealth" button on a specific host I see the graph for iis connections and sessions. This graph is missing under the "trends" view. There is only a dead link for iishealth. All other graphs appear normally. Any ideas? Thanks in advance. Chris

1 0

drop host not working
by stef.coene＠docum.org 05 Feb '06

05 Feb '06

Hi, I want to remove some purple tests from a host. But the command ~/server/bin/bb 127.0.0.1 "drop server check" is not doing what it should do. The test is still displayed as purple. So, can I do this manually? The problem is I know how BigBrother is storing the information and how I can do some stuff manually (like update the html pages). For hobbit, I can nog find such information. Where is the status stored? How can I manually update the html pages? Stef

2 2

Layout of webpages
by stef.coene＠docum.org 05 Feb '06

05 Feb '06

Hi, Is it possible to put a host on 2 lines with some checks on the other and some checks on the other line? My hobbit server is an aix server and I want to put the aix tests together with the other aix hosts and the hobbit related tests on sperate line. Stef

1 1

hobbit and ttcp
by great.dilla＠gmail.com 05 Feb '06

05 Feb '06

Fellows, Anyone who has had success with hobbit-ttcp and its graphs? I wonder how to make ttcp run as a daemon and have it as a server-side service for hobbit on selected machines. This was how i found out that our Solaris 9 machine ran very slowly but with 100Mbps full-duplex network. ftp, telnet and other network services ran too painfully slow (ftp of ~200kb/s). I was wondering if maybe this check could be incorporated to others as well. Thanks in advance. Cheers, Great

1 0

[hobbit] Hobbit_d crashing - help!
by kmckinnis＠macrovision.com 04 Feb '06

04 Feb '06

Thanks Henrik. Interestingly, I am running 4.1.2p1... Fresh install. I apologize for not noting that in my original email. I dropped the test and haven't seen it crash again yet. I appreciate the reassurance that alerts are not lost, that's what matters most :) I was more worried about it happening again, as I couldn't find a cause. -----Original Message----- From: Henrik Stoerner [mailto:henrik at hswn.dk] Sent: Saturday, February 04, 2006 12:17 AM To: hobbit at hswn.dk Subject: Re: [hobbit] Hobbit_d crashing - help! On Fri, Feb 03, 2006 at 05:09:25PM -0800, Kimberly McKinnis wrote: > This is bad. Especially since this system goes live at 4am. > > hobbitd_alert > - Program crashed > > Fatal signal caught! > > How can I prevent this? Is this a bug somewhere? Yes, it's a bug. It only affects alerts - it's the "hobbitd_alert" module that crashes, not hobbitd itself. If this is version 4.1.2, then there's a known bug which is fixed in 4.1.2p1 that has this effect. It's mostly an annoyance, though - alerts are not lost because of this, since hobbitlaunch will immediately re-start the alert task. > I don't see anything in the logs, unless I'm looking in the wrong place. > /var/logs/hobbit logs are pretty empty, save for notifications.log. hobbitlaunch.log should have a notice about the task terminating with a signal, and you should have a core file in the ~hobbit/data/tmp/ directory. Henrik To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk

1 0

[hobbit] multiple graphs from an ncv test
by gendalia＠iastate.edu 04 Feb '06

04 Feb '06

In message <4175.1138438339 at entilzha.ait.iastate.edu>, Tracy Di Marco White writes: } }I have a test that picks up about 30 things and puts them in an rrd. }I was graphing one originally, and that one still works, but I'm not }sure how to get everything graphed. }"TRENDS:*,afs:afs|afs1|afs2" only shows the original afs graph on }both the afs test page and the trends page, and the trends page shows }"hobbit graph afs1" & "hobbit graph afs2". Is there more to having }multiple graphs work for a test? Maybe this isn't something that is supposed to work? My test returns: Sat Feb 4 03:05:16 2006 free packets: 1614 calls waiting: 0 green threads idle: 34 server connections: 1 client connections: 6 peer structs: 7 call structs: 2 free calls: 1 packet allocation failures: 0 calls: 43083 allocs: 132855 read data: 43086 read ack: 89763 read dup: 0 read spurious: 0 read busy: 0 read abort: 0 read ackall: 0 read challenge: 2 read response: 0 sent data: 172338 sent resent: 0 sent ack: 3599 sent busy: 0 sent abort: 0 sent ackall: 0 sent challenge: 0 sent response: 2 Instance upclientetc, currently running normally. Instance fs, currently running normally. Auxiliary status is: file server running. green ----- I currently have a graph for "calls waiting", but not for anything else. I don't know much about how to set up graphs, so I may well have it all wrong, but I'm not sure if it's just not possible now. }My hobbitgraphs.cfg also has: }[afs] } DEF:calls=afs.rrd:callswaiting:AVERAGE } YAXIS calls waiting for thread } TITLE AFS fileservers calls waiting for thread } COMMENT:\n } GPRINT:calls:LAST:Calls waiting for thread \: %5.1lf (cur) } GPRINT:calls:MAX: \: %5.1lf (max) } GPRINT:calls:MIN: \: %5.1lf (min) } GPRINT:calls:AVERAGE: \: %5.1lf (avg)\n } }[afs1] } DEF:freePackets=afs.rrd:freepackets:AVERAGE } YAXIS free RX packet buffers } TITLE Number of free RX packet buffers } COMMENT:\n } GPRINT:buffers:LAST:buffers \: %5.1lf (cur) } GPRINT:buffers:MAX:buffers \: %5.1lf (cur) } GPRINT:buffers:MIN:buffers \: %5.1lf (cur) } GPRINT:buffers:AVERAGE:buffers \: %5.1lf (cur) } }[afs2] } DEF:threadsIdle=afs.rrd:threadsIdle:AVERAGE } YAXIS idle threads } TITLE Number of idle threads } COMMENT:\n } GPRINT:threads:LAST:threads\: %5.1lf (cur) } GPRINT:threads:MAX:threads\: %5.1lf (cur) } GPRINT:threads:MIN:threads\: %5.1lf (cur) } GPRINT:threads:AVERAGE:threads\: %5.1lf (cur) } }(I'm not sure I have these set up correctly.) Thanks, Tracy J. Di Marco White Information Technology Services Iowa State University

1 0

Hobbit_d crashing - help!
by kmckinnis＠macrovision.com 04 Feb '06

04 Feb '06

This is bad. Especially since this system goes live at 4am. hobbitd_alert - Program crashed Fatal signal caught! How can I prevent this? Is this a bug somewhere? I don't see anything in the logs, unless I'm looking in the wrong place. /var/logs/hobbit logs are pretty empty, save for notifications.log. sc-isus-boot hobbit # cat hobbitd.log 2006-02-03 12:32:43 Setup complete sc-isus-boot hobbit #

2 1

Messages file not reporting
by ecroft＠openratings.com 03 Feb '06

03 Feb '06

I was able to have log file reporting under Big Brother and the df-msgstab would tell what to alert on. Is that functionality gone in Hobbit? System logs at Thu Feb 2 14:11:38 EST 2006 : No log data available The client did not report any logfile data -- Edward M. Croft Sr. Systems Engineer Open Ratings, Inc. 200 West Street Waltham, MA 02451-1121

5 10

hobbit
by anilth＠hi.is 03 Feb '06

03 Feb '06

1 0

[hobbit] Alerting on Server Reboot
by MilburnJA＠dot.il.gov 02 Feb '06

02 Feb '06

If the servers are part of an Active Directory domain, you can use a policy to run a script at startup. I use one that writes an event log entry in a management server and sends an email to the admins. -----Original Message----- From: PNIXON at ci.somerville.ma.us [mailto:PNIXON at ci.somerville.ma.us] Sent: Thursday, February 02, 2006 10:03 AM To: hobbit at hswn.dk Subject: RE: [hobbit] Alerting on Server Reboot Doh! Forgot to mention this is primarily for Windows Servers (I know, I know, not my group :) --Pat -----Original Message----- From: Dan Vande More [mailto:bigdan at gmail.com] Sent: Thursday, February 02, 2006 10:54 AM To: hobbit at hswn.dk Subject: Re: [hobbit] Alerting on Server Reboot Pat, One of the things I've done in the past is have a init.d script (For linux at least), that emails your account that the server just booted. Something like this: #!/bin/sh echo "`hostname -f` just rebooted!" | mail -s "`hostname -f` just rebooted!" admin at domain.com EOF Maybe you could just append that to your hobbit client's init scripts? In any case, I'd be interested to see what others have to say as well. -Dan On 2/2/06, PNIXON at ci.somerville.ma.us <PNIXON at ci.somerville.ma.us> wrote: > Hey all, > Does anyone have a recommended solution for triggering an alert when > a server is rebooted? > > Assuming it comes back up within the ping cycle, the conn test still > passes. > > I'm aware that the CPU turns yellow due to the reboot, but is it > possible to build an alert rule to trigger on ServerX CPU = Yellow? > > Thanks for any feedback... > > --Pat > > > To unsubscribe from the hobbit list, send an e-mail to > hobbit-unsubscribe at hswn.dk > > > To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk

1 0