June 2010 - Xymon - Xymon mailman web

TS - RRD Graph incorrect
by d.tom.schmitt＠L-3com.com 10 Jun '10

10 Jun '10

I am trying to graph the number of users on our Juniper SSL-VPN. I am sending the value pairs as: sslvpn : 227 ,etc. I am having a problem with the graph definition: [sslvpn] TITLE Count of Juniper Users YAXIS Users DEF:in=sslvpn.rrd:sslvpn:AVERAGE AREA:in#00CC00:Number of Current Users COMMENT: Licensed for 600 users\n GPRINT:in:LAST:Cur\: %6.2lf %s GPRINT:in:MAX: Max\: %6.2lf %s GPRINT:in:MIN: Min\: %6.1lf %s GPRINT:in:AVERAGE:Avg\: %6.2lf %s\n The Graph has a bad YAXIS. I need it to show the number of users and if possible a line in RED showing 600 as this is the License limit. I need assistance in getting the YAXIS values to be a number from zero(0) on up. Is it possible to also place a static line at the 600 value across the graph? Thanks, Tom Schmitt Senior IT Staff - R&D L-3 Communication Systems West 640 North 2200 West P.O. Box 16850 Salt Lake City, UT 84116 Phone (801) 594-3030 Cell (801) 231-7230 eFax (413) 480-6873 D.Tom.Schmitt at L-3Com.com \\\\||//// \ ~ ~ / | @ @ | --oOo---(_)---oOo--

3 2

Alerts not acknowledging
by david.baldwin＠ausport.gov.au 10 Jun '10

10 Jun '10

I have a frustrating situation where some alerts are not acknowledging. I don't know why some do and some don't. I am using "Acknowledge alerts" under the admin menu. When I complete the form for the appropriate alert I get the response on teh form: "Acknowledge sent for host servicedvsv.ausport.gov.au / test esx" Looking through the source of web/bb-ack.c I can see it is sending a message of the form: % bb localhost "hobbitdack 88362 60 msg user" Where 88362 is the cookie and 60 is the time to acknowledge. Trouble is, I get a message in hobbitd.log: 2010-06-10 12:01:02 Cookie 88362 not found, dropping ack 2010-06-10 12:06:43 Cookie 88362 not found, dropping ack 2010-06-10 12:09:39 Cookie 88362 not found, dropping ack The cookie is correct: % bb localhost "hobbitdboard host=servicedvsv.ausport.gov.au test=esx" servicedvsv.ausport.gov.au|esx|red||1275561583|1276135842|1276137642|0|0|10.1.0.161|88362|red Thu Jun 10 12:10:42 2010 In some cases, I can do a multiple acknowledge on various tests on the same host, and anything from all to none of the tests get acknowledged. :( Anyone else seen this? David. -- David Baldwin - IT Unit Australian Sports Commission www.ausport.gov.au Tel 02 62147830 Fax 02 62141830 PO Box 176 Belconnen ACT 2616 david.baldwin at ausport.gov.au Leverrier Street Bruce ACT 2617 ------------------------------------------------------------------------------------- Keep up to date with what's happening in Australian sport visit http://www.ausport.gov.au This message is intended for the addressee named and may contain confidential and privileged information. If you are not the intended recipient please note that any form of distribution, copying or use of this communication or the information in it is strictly prohibited and may be unlawful. If you receive this message in error, please delete it and notify the sender. -------------------------------------------------------------------------------------

3 2

Critical System Editor Tips.
by nfranken＠theunlimitedworld.co.za 10 Jun '10

10 Jun '10

Hi Guys/Gals I am busy setting up the critical systems and I am running into a brick wall. For example I add a server called TUW-x and the svcs to a level one priority and click search it tell me that there is no record for this host/service. I then fill out the other details and click the update record button. When I then search for this record I get the same error message. How do I know that this record was saved? I am pretty sure it was not as I check the Config Report(Critical) and get very empty report with zero hosts. Any ideas where I can look for errors regarding the critical setup. Regards Neil

5 8

message never goes away.
by rmcgraw＠purdue.edu 09 Jun '10

09 Jun '10

I think I have figure out my problem. I am not sure of the format of the file logfetch.status but it looked like the following: /var/adm/messages:0:0:0:0:0:0:0 I cleaned up some files and restarted the hobbit client on the host conon and now the file looks like /var/adm/messages:23485:23485:23436:23436:23436:23436:23387 It looks like the numbers are the number of characters in the file every x minutes, but just a guess. It seem to be working now so this can be closed. Robert > -----Original Message----- > From: McGraw, Robert P > Sent: Wednesday, June 09, 2010 4:03 PM > To: 'hobbit at hswn.dk' > Subject: message never goes away. > > I have one client that sends out the /var/adm/messages files but on the > server side it never seems to clear itself. > > On the display server the Critical entry was on June 8 at 13:28:00 but > the current time is June 9 15:38:52. I though it would only show the > error message for X amount of time. > > This is what the Display server shows under "msgs". > > Any idea why the error messages do not stop displaying? > > Thanks > > Robert > > > > _____________________________________________________________________ > Robert P. McGraw, Jr. > Manager, Computer System EMAIL: rmcgraw at purdue.edu > Purdue University ROOM: MATH-807 > Department of Mathematics PHONE: (765) 494-6055 > 150 N. University Street > West Lafayette, IN 47907-2067 >

1 0

Use of "TIME" Within hobbit-clients.cfg
by wiskbroom＠hotmail.com 09 Jun '10

09 Jun '10

I have several machines which run processes only during certain hours of the weekday, but never on holidays. I've got them defined in the following manner within my hobbit-clients.cfg file: HOST=myhost PROC "myAgent.pl -c myAgent.cfg" 1 "TEXT=SLIPPERY AGENT" TIME=W:0700:1600 For the most part, everything works as well as expected, except each weekday at, or around 4PM, I get an alarm, which subsequently clears up at 4:05PM. 1. Should I redefine this instead to TIME=W:0700:1555, or is there something else I can do? 2. How can I say, but not on holidays? Thanks all, .vp

1 0

ipv6 support
by vadud3＠gmail.com 09 Jun '10

09 Jun '10

Does hobbit support ipv6 addressing? -- Asif Iqbal PGP Key: 0xE62693C5 KeyServer: pgp.mit.edu

3 6

message never goes away.
by rmcgraw＠purdue.edu 09 Jun '10

09 Jun '10

I have one client that sends out the /var/adm/messages files but on the server side it never seems to clear itself. On the display server the Critical entry was on June 8 at 13:28:00 but the current time is June 9 15:38:52. I though it would only show the error message for X amount of time. This is what the Display server shows under "msgs". ----- System logs at Wed Jun 9 15:58:08 EDT 2010 red Critical entries in /var/adm/messages red Jun 8 13:28:00 conon.math.purdue.edu su: [ID 810491 auth.crit] 'su root' failed for admin on /dev/pts/3 Full log /var/adm/messages <...SKIPPED...> th.purdue.edu -- MARK -- Jun 6 19:19:14 conon.math.purdue.edu -- MARK -- Jun 6 19:39:14 conon.math.purdue.edu -- MARK -- Jun 6 19:59:13 conon.math.purdue.edu -- MARK -- Jun 6 20:19:13 conon.math.purdue.edu -- MARK -- Jun 6 20:39:13 conon.math.purdue.edu -- MARK -- Jun 6 20:59:13 conon.math.purdue.edu -- MARK -- Jun 6 21:19:13 conon.math.purdue.edu -- MARK -- Jun 6 21:39:13 conon.math.purdue.edu -- MARK -- : : Jun 9 13:58:53 conon.math.purdue.edu -- MARK -- Jun 9 14:18:53 conon.math.purdue.edu -- MARK -- Jun 9 14:38:53 conon.math.purdue.edu -- MARK -- Jun 9 14:58:53 conon.math.purdue.edu -- MARK -- Jun 9 15:18:53 conon.math.purdue.edu -- MARK -- Jun 9 15:38:52 conon.math.purdue.edu -- MARK -- ---- Any idea why the error messages do not stop displaying? Thanks Robert _____________________________________________________________________ Robert P. McGraw, Jr. Manager, Computer System EMAIL: rmcgraw at purdue.edu Purdue University ROOM: MATH-807 Department of Mathematics PHONE: (765) 494-6055 150 N. University Street West Lafayette, IN 47907-2067

1 0

http session cookies reuse
by mvadagnini＠h-care.eu 09 Jun '10

09 Jun '10

Hi list. I noticed that the file server/etc/cookie.session contains the last session cookies used for http testing via POST (i.e. JSESSIONID), but it seems that my webapp instantiates a new http session for every check. This makes very hard to know the REAL number of http sessions on the application server and introduces an offset in the statistics. I searched documentation and mailing list for a way to reuse the session cookie stored in that file, but found no information. Is there a way to reuse the same cookie in all the requests? my bb-hosts content is something like: 123.456.789.10 my.host post=webapp;http://my.host:8080/webapp/_dwr/call/plaincall/function.dwr;callCount=1&page=/webapp/&httpSessionId=&scriptSessionId=&c0-scriptName=function&c0-methodName=sendEvent&c0-id=0&c0-e1=string:default-web&c0-e2=string:monitoring&c0-e3=string:&c0-param0=Object_Object:{serviceId:reference:c0-e1,key:reference:c0-e2,params:reference:c0-e3}&batchId=1;check[[:space:]]ok HIDEHTTP Thanks Mat

1 0

FW: [hobbit] TS - Adding an ICON
by d.tom.schmitt＠L-3com.com 09 Jun '10

09 Jun '10

Added the setting of the ICON to the code for the Juniper SSLVPN pair. To set the ICON: BB=/home/xymon/server/bin/bb BBDISP-127.0.0.1 DEVICE=SSLVPN1 or SSLVPN2 COLUNM="online" COLOR=green for 'online' server =clear for 'offline' server MSG="<the message to show when you click the icon>" $BB $BBDISP "status $DEVICE.$COLUMN $COLOR `date` ${MSG} " This adds the new column called 'online' to both SSLVPN1 and 2 and sets the color to green or clear/white. Just add the code to the code below and also use 'mutt' to send an alert upon the change. I have the complete code if required. Thanks, Tom Schmitt Senior IT Staff - R&D L-3 Communication Systems West 640 North 2200 West P.O. Box 16850 Salt Lake City, UT 84116 Phone (801) 594-3030 Cell (801) 231-7230 eFax (413) 480-687 D.Tom.Schmitt at L-3Com.com \\\\||//// \ ~ ~ / | @ @ | --oOo---(_)---oOo-- -----Original Message----- From: Schmitt, D Tom @ CSW-SLC Sent: Monday, June 07, 2010 8:41 AM To: 'hobbit at hswn.dk' Subject: RE: [hobbit] TS - Adding an ICON Hope this helps! FYI: I found that the Primary server has all the information if you do an snmpbulkwalk -v2c -c public IP-of-server1 or 2 .1 >FILENAME The size of the file is very large for the online server compared to the offline server. I did the command above for both servers and then did a diff of the 2 files. I then looked for the Virtual IP address using: grep -I "<virtual-IP>" * If you find a better alternative, please let me know. I currently use the Juniper .cgi web page in the Xymon bb-hosts line to verify that there is a response. Both return the identical screen so you cannot tell which is online from this testing. This is what I am looking to do and then I want to toggle RED/YELLOW/GREEN icons on the same Xymon display line. The only device that seems to have the virtual IP address in the MIB is the live device. STRING=public # Set to your RO string #---------------------------------- # Loop through each Juniper SSL-VPN #---------------------------------- for IP in IP-of-server1 IP-of-server2 do #-------------------------------------- # Get the sysName - either IVE1 or IVE2 #-------------------------------------- NAME=`/usr/bin/snmpget -v2c -c $STRING $IP SNMPv2-MIB::sysName.0 2>&1 | \ awk -F'=' '{print $2}' | \ awk -F':' '{print $2}'` #------------------------------------------------------------- # Get the ipAdEntAddr - <Virtual-IP> or 'No Such Instance ...' #------------------------------------------------------------- VIRT=`/usr/bin/snmpget -v2c -c $STRING $IP IP-MIB::ipAdEntAddr.<Virtual-IP> 2>&1 | \ awk -F'=' '{print $2}' | \ awk -F':' '{print $2}'` if [ "$VIRT" = "" ]; then VIRT=" off-line" fi echo -e "IP: $IP NAME: $NAME Virtual IP: $VIRT" done Thanks, Tom Schmitt Senior IT Staff - R&D L-3 Communication Systems West 640 North 2200 West P.O. Box 16850 Salt Lake City, UT 84116 Phone (801) 594-3030 Cell (801) 231-7230 eFax (413) 480-6873 D.Tom.Schmitt at L-3Com.com \\\\||//// \ ~ ~ / | @ @ | --oOo---(_)---oOo-- -----Original Message----- From: W.J.M. Nelis [mailto:nelis at nlr.nl] Sent: Monday, June 07, 2010 1:42 AM To: hobbit at hswn.dk Subject: Re: [hobbit] TS - Adding an ICON Hello Tom, > I am using a MIB check to verify that a device (in a set of HA > devices) is the current one using the virtual IP address. > > I can tell from my scripting that the device has switched to the other > of the HA pair. > > (A Juniper SSL-VPN pair). > > I will be using NET-SNMP on the XYMON server to gather the info using > a crontab script. > Which MIB variable (OID) can be used to determine which device is the one using the virtual IP address? Regards, Wim Nelis. ************************************************************************ ******************************* The NLR disclaimer (http://www.nlr.nl/emaildisclaimer) is valid for NLR e-mail messages. ************************************************************************ ******************************* To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk

2 3

New client-side external script to monitor Bacula backup server with Xymon
by waa-hobbitml＠revpol.com 08 Jun '10

08 Jun '10

The script is a simple (and a little ugly) Bash shell script that uses Bacula's bconsole "status dir" command output to see if any jobs are waiting on operator intervention, and it also checks the last "x" number of jobs to see if any of them failed and reports both instances back to a Xymon monitoring server. For those who do not know, Bacula is an open-source, enterprise network backup system that is widely used. It may be found here: http://www.bacula.org/ I wrote the script a while back but modified it today to include checking for "intervention required" type issues. An entry for the script has been added to Xymonton (http://xymonton.trantor.org/doku.php/monitors#available_monitors) and may be found here: http://www.revpol.com/xymon_bacula_check_script There are currently no instructions on installing it, but they will be up shortly. Basic instructions for installing such a script may be found in the Xymon man pages as well. Hope this helps someone somewhere. :) (I cross-posted a similar notice to the bacula-users mailing list today) -- Bill Arlofski Reverse Polarity, LLC http://www.revpol.com/

4 4