- Xymon - Xymon mailman web

[hobbit] Monitoring Sql Databases
by dns1407＠yahoo.com 27 Jun '08

27 Jun '08

Hi Nikesh, I use the following vbscript to monitor MS SQL Servers.... ====Begin====== On Error Resume Next strAlarmState = "green" strTestName = "sqlserver" 'Physical Disk Settings '% Disk Read Time iPerDiskReadTimeWarn=5 iPerDiskReadTimeAlarm=10 '% Disk Write Time iPerDiskWriteTimeWarn=5 iPerDiskWriteTimeAlarm=10 'Current Disk Queue Length iCurDiskQueueLengthWarn=10 iCurDiskQueueLengthAlarm=20 'SQL Buffer Cache Hit Ratio iBufferCacheHitRatioWarn=90 iBufferCacheHitRatioAlarm=85 'SQL Lock Timeouts /sec iLockTimeoutsWarn=50 iLockTimeoutsAlarm=100 'SQL Dead Locks /sec iDeadLocksWarn=10000 iDeadLocksAlarm=100000 'User Connections iUserConnectionsWarn=150 iUserConnectionsAlarm=300 ' Master Database Settings Set aMasterDatabaseSetting = CreateObject("scripting.dictionary") aMasterDatabaseSetting.add "DataFileWarn", 150000 aMasterDatabaseSetting.add "DataFileAlarm", 200000 aMasterDatabaseSetting.add "LogFileWarn", 20000 aMasterDatabaseSetting.add "LogFileAlarm", 40000 ' Model Database Settings Set aModelDatabaseSetting = CreateObject("scripting.dictionary") aModelDatabaseSetting.add "DataFileWarn", 150000 aModelDatabaseSetting.add "DataFileAlarm", 200000 aModelDatabaseSetting.add "LogFileWarn", 20000 aModelDatabaseSetting.add "LogFileAlarm", 40000 'Main Array SET aSQLDataFileSettings = CreateObject("scripting.dictionary") aSQLDataFileSettings.add "master", aMasterDatabaseSetting aSQLDataFileSettings.add "model", aModelDatabaseSetting strOutput = "" Set ws = WScript.CreateObject("WScript.Shell") extPath = ws.RegRead("HKLM\SOFTWARE\BBWin\Output\") ' ======================================== ' Main Code Starts Here 'Physical Disk Set colDisk = GetObject("winmgmts:").InstancesOf("Win32_PerfRawData_PerfDisk_PhysicalDisk") strOutput = strOutput & vbcrlf &"Physical Disk Information:" & vbcrlf For each DiskInstance in ColDisk '% Disk Read Time strOutput = strOutput & CheckValue(GetWMIPercent("Win32_PerfRawData_PerfDisk_PhysicalDisk.Name","""" & DiskInstance.Name & """","PercentDiskReadTime","PercentDiskReadTime_Base"),DiskInstance.name & " % Disk Read Time",iPerDiskReadTimeWarn,iPerDiskReadTimeAlarm) '% Disk Write Time strOutput = strOutput & CheckValue(GetWMIPercent("Win32_PerfRawData_PerfDisk_PhysicalDisk.Name","""" & DiskInstance.Name & """","PercentDiskWriteTime","PercentDiskWriteTime_Base"),DiskInstance.name & " % Disk Write Time",iPerDiskWriteTimeWarn,iPerDiskWriteTimeAlarm) 'Current Disk Queue Length strOutput = strOutput & CheckValue(GetWMIValue("Win32_PerfRawData_PerfDisk_PhysicalDisk.Name","""" & DiskInstance.Name & """","CurrentDiskQueueLength"),DiskInstance.name & " Current Disk Queue Length",iCurDiskQueueLengthWarn,iCurDiskQueueLengthAlarm) Next ' SQL Server strOutput = strOutput & vbcrlf & "SQL Server Information:" & vbcrlf 'Buffer Cache Hit Ration strOutput = strOutput & CheckReverseValue(GetWMIPercent("Win32_PerfRawData_MSSQLSERVER_SQLServerBufferManager","@","Buffercachehitratio","Buffercachehitratio_Base"),"Buffer Cache Hit Ratio",iBufferCacheHitRatioWarn, iBufferCacheHitRatioAlarm) 'Lock Information ' strOutput = strOutput & CheckValue(GetWMIValue("Win32_PerfRawData_MSSQLSERVER_SQLServerLocks.Name","""_Total""","LockTimeoutsPersec"),"Locks Timeouts/sec",iLockTimeoutsWarn, iLockTimeoutsAlarm) strOutput = strOutput & CheckValue(GetWMIValue("Win32_PerfRawData_MSSQLSERVER_SQLServerLocks.Name","""_Total""","NumberofDeadlocksPersec"),"Number of Deadlocks/sec",iDeadLocksWarn, iDeadLocksAlarm) 'User Connections strOutput = strOutput & CheckValue(GetWMIValue("Win32_PerfRawData_MSSQLSERVER_SQLServerGeneralStatistics","@","UserConnections"),"Number of User Connections",iUserConnectionsWarn, iUserConnectionsAlarm) 'SQL Data and Log Files Set colDatabases = GetObject("winmgmts:").InstancesOf("Win32_PerfRawData_MSSQLSERVER_SQLServerDatabases") strOutput = strOutput & vbcrlf & "SQL Server Data and Log File Information:" & vbcrlf For each DatabaseInstance in colDatabases If aSQLDataFileSettings.exists(DatabaseInstance.name) then 'Check Values strOutput = strOutput & CheckValue(GetWMIValue("Win32_PerfRawData_MSSQLSERVER_SQLServerDatabases","""" & DatabaseInstance.name & """","DataFilesSizeKB"),DatabaseInstance.name & vbTab & "Data File Size(KB)",aSQLDataFileSettings.item(DatabaseInstance.name).item("DataFileWarn"), aSQLDataFileSettings.item(DatabaseInstance.name).item("DataFileAlarm")) strOutput = strOutput & CheckValue(GetWMIValue("Win32_PerfRawData_MSSQLSERVER_SQLServerDatabases","""" & DatabaseInstance.name & """","LogFilesSizeKB"),DatabaseInstance.name & vbTab & "Log File Size(KB)",aSQLDataFileSettings.item(DatabaseInstance.name).item("LogFileWarn"), aSQLDataFileSettings.item(DatabaseInstance.name).item("LogFileAlarm")) strOutput = strOutput & vbcrlf elseif DatabaseInstance.Name = "_Total" then ' Skipping Display of Totals at this stage. May re-add later ' strOutput = strOutput & vbcrlf & "Total Data File Size:" & vbTab & GetWMIValue("Win32_PerfRawData_MSSQLSERVER_SQLServerDatabases.Name","""" & DatabaseInstance.name & """","DataFilesSizeKB") & vbcrlf ' strOutput = strOutput & vbcrlf & "Total Log File Size:" & vbTab & GetWMIValue("Win32_PerfRawData_MSSQLSERVER_SQLServerDatabases.Name","""" & DatabaseInstance.name & """","LogFilesSizeKB") & vbcrlf else ' Write Data Information Directly Out without running the Value Check ' Datafile strOutput = strOutput & "&clear" & " " & DatabaseInstance.name & vbTab & "Data File Size(KB)" & ":" & vbTab & GetWMIValue("Win32_PerfRawData_MSSQLSERVER_SQLServerDatabases.Name","""" & DatabaseInstance.name & """","DataFilesSizeKB") & vbcrlf ' Logfile strOutput = strOutput & "&clear" & " " & DatabaseInstance.name & vbTab & "Log File Size(KB)" & ":" & vbTab & GetWMIValue("Win32_PerfRawData_MSSQLSERVER_SQLServerDatabases.Name","""" & DatabaseInstance.name & """","LogFilesSizeKB") & vbcrlf strOutput = strOutput & vbcrlf end if Next ' Write the file for BB WriteFile extPath, strTestName, strAlarmState, strOutput '=========================================================== ' FUNCTIONS and SUBS start here ' This is used to get a percentage value from WMI. It requires the value and the base objects. ' It then returns the percentage FUNCTION GetWMIPercent(strCollection,strInstance,strObject,strBaseObject) SET counterCollection = GETOBJECT("winmgmts:" & strCollection & "=" & strInstance) FOR EACH cntproperty IN counterCollection.properties_ IF cntproperty.name = strObject THEN iObjectValue = cntproperty ELSEIF cntproperty.name = strBaseObject THEN iObjectBaseValue = cntproperty END IF NEXT GetWMIPercent = ROUND(CDBL(iObjectValue) / CDBL(iObjectBaseValue) * CDBL(100),0) END FUNCTION ' This is used to pull a value from WMI. FUNCTION GetWMIValue(strCollection,strInstance,strObject) Set counterCollection = GetObject("winmgmts:" & strCollection & "=" & strInstance) FOR EACH cntproperty IN counterCollection.properties_ IF cntproperty.name = strObject THEN iObjectValue = cntproperty END IF NEXT GetWMIValue = iObjectValue END FUNCTION ' This is used to check the actual value against the warning and alarm. FUNCTION CheckValue(iObjectValue,strObjectDesc,iWarnValue,iAlarmValue) IF iWarnValue > iAlarmValue THEN CheckValue = "&red" & " " & strObjectDesc & ":" & vbTab & "Object is Misconfigured" IF strAlarmState <> "red" THEN strAlarmState = "red" END IF ELSE IF iObjectValue > iWarnValue THEN IF iObjectValue > iAlarmValue THEN CheckValue = "&red" & " " & strObjectDesc & ":" & vbTab & iObjectValue & vbcrlf SetAlarmStatus "red" ELSE CheckValue = "&yellow" & " " & strObjectDesc & ":" & vbTab & iObjectValue & vbcrlf SetAlarmStatus "yellow" END IF ELSE CheckValue = "&green" & " " & strObjectDesc & ":" & vbTab & iObjectValue & vbcrlf END IF END IF END FUNCTION ' This is used to check the actual value against the warning and alarm. ' This one the alarm will be a lower value than the warning. (Values Decrease rather than increase) FUNCTION CheckReverseValue(iObjectValue,strObjectDesc,iWarnValue,iAlarmValue) IF iWarnValue < iAlarmValue THEN CheckReverseValue = "&red" & " " & strObjectDesc & ":" & vbTab & "Object is Misconfigured" IF strAlarmState <> "red" THEN strAlarmState = "red" END IF ELSE IF iObjectValue < iWarnValue THEN IF iObjectValue < iAlarmValue THEN CheckReverseValue = "&red" & " " & strObjectDesc & ":" & vbTab & iObjectValue & vbcrlf SetAlarmStatus "red" ELSE CheckReverseValue = "&yellow" & " " & strObjectDesc & ":" & vbTab & iObjectValue & vbcrlf SetAlarmStatus "yellow" END IF ELSE CheckReverseValue = "&green" & " " & strObjectDesc & ":" & vbTab & iObjectValue & vbcrlf END IF END IF END FUNCTION ' This is called to set the overall alarm status. SUB SetAlarmStatus(strnewAlarmState) IF strnewAlarmState = "red" THEN strAlarmState = strnewAlarmState ELSEIF strnewAlarmState = "yellow" THEN IF strAlarmState <> "red" THEN strAlarmState = strnewAlarmState END IF END IF END SUB ' This SUB is used for outputting the file to the external's directory in bb SUB WriteFile(strExtPath, strTestName, strAlarmState, strOutput) Set fso = CreateObject("Scripting.FileSystemObject") strOutput = strAlarmState & " " & Date & " " & Time & vbcrlf & vbcrlf & strOutput & vbcrlf Set f = fso.OpenTextFile(strExtPath & "\" & strTestName , 8 , TRUE) f.Write strOutput f.Close Set fso = Nothing END SUB ===End==== In the registry of the Windows host you have to add: [HKEY_LOCAL_MACHINE\SOFTWARE\BBWin\Output] @="C:\\Program Files\\BBWin\\tmp" Success.... Kind regards, DNS ----- Original Message ---- From: Nikesh Maharaj <NMaharaj at tcta.co.za> To: hobbit at hswn.dk Sent: Tuesday, June 17, 2008 11:29:35 AM Subject: RE: [hobbit] Monitoring Sql Databases Hi, I just want to monitor sql2000 and sql2005 databases for sizes , jobs etc. will this script help ? -----Original Message----- From: Buchan Milne [mailto:bgmilne at staff.telkomsa.net] Sent: 17 June 2008 11:08 AM To: hobbit at hswn.dk Cc: Nikesh Maharaj Subject: Re: [hobbit] Monitoring Sql Databases On Tuesday 17 June 2008 08:52:27 Nikesh Maharaj wrote: > Hi Guys, > > I am aware that Hobbit can monitor and report on Sql Databases. Please > if I may ask, can someone who has this working already, please assist me > or point me in the direction I can get this implemented on my Hobbit > monitor ? You may want to be more specific on the database in question. We use dbcheck from http://sourceforge.net/projects/hobbit-perl-cl/ to monitor Oracle and MySQL. Regards, Buchan To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk

6 8

[bbwin-users] msgs - fields not being populated, %1 %2 %3.
by s_aiello＠comcast.net 27 Jun '08

27 Jun '08

On Monday 16 June 2008, s_aiello at comcast.net wrote: > Hello All, > > I normally run bbwin 0.9, but I have just notice a strange error with msgs > report output. Seem some fields event log message are not getting > populated. I just upgraded a test server to the 0.12 and I see the same > issue there as well. An example of this problem would be: > > bbwin msgs output snippet: > application: warning - 2008/06/16 09:20:11 - ASP.NET 2.0.50727.0 (1309) - > n/a "Event code: %1 Event message: %2 Event time: %3 Event time (UTC): %4 > Event ID: %5 > > eventlog message seen from the server's event viewer, snippet: > Event code: 3005 > Event message: An unhandled exception has occurred. > Event > time: 6/16/2008 9:20:11 AM > Event time (UTC): 6/16/2008 1:20:11 PM > Event ID: > 9f92c98a8695411c99fa81faf5362d50 > > Thoughts, advice ? > ~Steve > has anyone seen the above issue, just curious if i am completely crazy or not. thanks, `steve

2 1

[hobbit] Can hobbit server comprehense COMBO keywords like BB server
by Myra.Haubrich＠ec.gc.ca 27 Jun '08

27 Jun '08

========= On Thu, Jun 26, 2008 at 09:42:19AM -0400, Haubrich,Myra [Ontario] wrote: > Thank you for your reply! > > I've tried with > > # nc localhost 1984 < test.txt > > Where test.txt contents: > combo > > status server1,xx,xx,xx,xx.cpu green Tue Jun 24 19:59:34 UTC 2008 up: 5 > days, 2 users, 119 procs, load=17 > > LOAD AVG on server1,xx,xx,xx,xx is 17 > > status server1,xx,xx,xx,xx.disk green Tue Jun 24 19:59:34 UTC 2008 - > Disk partitions on server1,xx,xx,xx,xx OK > > /dev/cciss/c0d1p1 100790004 9644576 86025516 11% /home > /dev/cciss/c0d0p3 5036316 44476 4736008 1% /tmp > /dev/cciss/c0d0p1 101086 19502 76365 21% /boot > /dev/cciss/c0d0p6 40670524 2182592 36421976 6% / > /dev/cciss/c0d0p2 20161204 988424 18148640 6% /var You have some new-line problems there. The layout of a combo message should be combo<newline> status ........ <newline> <newline> status ...... So your first blank line between the "combo" and the "status" should not be there. Not sure if this is what causes the problem, though... It could also be that your "test.txt" file has DOS/Windows style line termination (carriage-return + newline), this would also cause problems. Hobbit might be a bit more finicky about this than BB, but it was tested with the Big Brother "bb" client. And Hobbit does use combo messages internally, e.g. for all of the network test status messages (try running "bbcmd bbtest-net --no-update" and see the messages that get sent to the Hobbit server). The data you're sending looks pretty much like a standard Unix client report. Any particular reason why you're not just using the normal Hobbit client and the "bb" utility to send the data ? Regards, Henrik ====== The client host (a RHEL server) is behind firewalls, and I tried with hobbitfetch and msgcache, the normal hobbit way. I've checked with "lsof -i:1984" to make sure that msgcache is running on the client and hobbitfetch is running on hobbit server. It didn't work, most likely the firewall blocked it. Below is the configurations that I learned from the forum. Any suggestions? > =========== > > CLIENT SIDE > > /etc/sysconfig/hobbit-cleint > Hobbitserver="xx.xx.xx.xx 127.0.0.1" <-- hobbit server IP > > hobbitclient.cfg > BBDISP="127.0.0.1" <-- Changed from Hobbit server IP to localhost > > > clientlaunch.cfg > [msgcache] > # DISABLED <-- Comment it > ENVFILE $HOBBITCLIENTHOME/etc/hobbitclient.cfg > CMD $HOBBITCLIENTHOME/bin/msgcache --no-daemon > --pidfile=$HOBBITCLIENTHOME/logs/msgcache.pid > LOGFILE $HOBBITCLIENTHOME/logs/msgcache.log > > [client] > ENVFILE $HOBBITCLIENTHOME/etc/hobbitclient.cfg > CMD $HOBBITCLIENTHOME/bin/hobbitclient.sh --local <-- Add > --local > LOGFILE $HOBBITCLIENTHOME/logs/hobbitclient.log > INTERVAL 5m > > > SERVER SIDE > > hobbitlaunch.cfg > [hobbitfetch] > # DISABLED <-- Comment it > ENVFILE > /home/users/hobbit/application/server/etc/hobbitserver.cfg > CMD $BBHOME/bin/hobbitfetch --server=XX.XX.XX.XX --no-daemon > --pidfile=$BBSERVERLOGS/hobbitfetch.pid <-- replace the IP by your > Hobbit server one > LOGFILE $BBSERVERLOGS/hobbitfetch.log > > bb-hosts > zz.zz.zz.zz fqdn # conn hobbitfetch pulldata > > ========== Anyhow, a perl script on client host sends status data as combo message to hobbit server via ssh. The combo message is now working with hobbit server. Thank you so much for your help!!! MH.

2 2

bug report: hostname can't have "-" (dash) in bb-hosts file
by tj_yang＠hotmail.com 26 Jun '08

26 Jun '08

I have a few hostnames like following. group-compress<H3><I> Serial Console Server </I></H3> 0.0.0.0 linux2-sol.test.com # noconn 0.0.0.0 linux2-sol.test.com # noconn 0.0.0.0 linux3-sol.test.com # noconn 0.0.0.0 linux1-sol.test.com # noconn They won't shown up in the final html page unless I changed them to "_" underline characher like following. group-compress<H3><I> Serial Console Server </I></H3> 0.0.0.0 linux2_sol.test.com # noconn 0.0.0.0 linux2_sol.test.com # noconn 0.0.0.0 linux3_sol.test.com # noconn 0.0.0.0 linux1_sol.test.com # noconn T.J. Yang _________________________________________________________________ The other season of giving begins 6/24/08. Check out the i’m Talkathon. http://www.imtalkathon.com?source=TXT_EML_WLH_SeasonOfGiving

5 5

[hobbit] Hobbit compatible temperature monitor
by bewhite＠fellowes.com 26 Jun '08

26 Jun '08

We have a NetBotz appliance from APC which supports temperature probes. I use snmpwalk to query the MIB and produce a hobbit test which looks like: NetBotz Overall Status: normal(0) NetBotz Enclosure Status: normal(0) Module Pod Status: normal(0) Location 1 Temp at: DC Temp. DCH2 Location 1 Temp Status: normal(0) Location 1 Temperature : 76F Location 2 Temp at: Temp. East Location 2 Temp Status: normal(0) Location 2 Temperature : 70F Location 3 Temp at: Temp. West Location 3 Temp Status: normal(0) Location 3 Temperature : 77F Humidity Status: normal(0) Humidity: 34 I have never been able to get custom graphs working, so I am not graphing the output. But getting custom graphs is not high on my priority list. ......Bruce -----Original Message----- From: Tim McCloskey [mailto:morphine at zwizard.com] Sent: Wednesday, June 25, 2008 11:50 AM To: hobbit at hswn.dk Subject: Re: [hobbit] Hobbit compatible temperature monitor If your platform supports it you can use ipmitool for various internals. Actual server room temp would need some external device such as the ones others have mentioned. Example: On a Sun X4150 box one of the sensors is Ambient Temp0. temp="`/usr/sfw/bin/ipmitool sensor reading 'Ambient Temp0' | awk '{print $NF}'`" Other hardware has different sensor names, some just have (temp) sensors for CPU die temps. Regards, Buchan Milne wrote: > On Wednesday 25 June 2008 17:28:01 Eric Boyd wrote: > >> I am looking to add temperature graphs to hobbit. >> >> What should I use to monitor and graph the temperature of our server room? >> None of our servers report any temperature readings in the client data. >> > > What servers are these? Any recent HP or Dell servers, if you have their > respective agents (HPASM or OMSA) installed, should allow you to monitor > their temperatures via snmp, e.g. with Devmon + Hobbit you should be able to > get graphs for each temperature probe. > > >> We have some APC battery backups that have environmental monitors built >> in, >> > > Can you get the temperature data via SNMP? The existing templates for APC > devices in devmon don't have any 'temp' test, but if the data is there that > can be fixed. > > > Regards, > Buchan > > 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 ---------------------------------------------------- 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 5

Question about Linux Memory alert
by tlewick＠tradebotsystems.com 26 Jun '08

26 Jun '08

Hi all. I've setup the hobbit-clients.cfg with the following for memory for our linux servers... PAGE=NY/NY-Linux MEMPHYS 100 101 MEMSWAP 60 80 MEMACT 90 95 I've done this because the Physical Memory on the linux machines always shows quite high due to Linux's memory management, and thus far its worked great. But today we started receiving alerts on a host that the physical memory is 99%. The actual memory is only 10% used, so my question is why am I getting alerts on this host. Thanks, Taylor

2 1

[hobbit] Can hobbit server comprehense COMBO keywords like BB server
by Myra.Haubrich＠ec.gc.ca 26 Jun '08

26 Jun '08

Thank you for your reply! I've tried with # nc localhost 1984 < test.txt Where test.txt contents: combo status server1,xx,xx,xx,xx.cpu green Tue Jun 24 19:59:34 UTC 2008 up: 5 days, 2 users, 119 procs, load=17 LOAD AVG on server1,xx,xx,xx,xx is 17 status server1,xx,xx,xx,xx.disk green Tue Jun 24 19:59:34 UTC 2008 - Disk partitions on server1,xx,xx,xx,xx OK /dev/cciss/c0d1p1 100790004 9644576 86025516 11% /home /dev/cciss/c0d0p3 5036316 44476 4736008 1% /tmp /dev/cciss/c0d0p1 101086 19502 76365 21% /boot /dev/cciss/c0d0p6 40670524 2182592 36421976 6% / /dev/cciss/c0d0p2 20161204 988424 18148640 6% /var status server1,xx,xx,xx,xx .procs green Tue Jun 24 19:59:34 UTC 2008 All processes are OK &green cron >=1 - 1 instance running &green bbrun >=1 - 2 instances running ... Activity of Telvent HTTPIN Log @ 2008-06-24 19:59:36: GREEN Log file last updated on 2008-06-24 19:55:03. Contents of Telvent HTTPIN Log @ 2008-06-24 19:59:36: GREEN No errors found. I've checked that the nc command went through, but no update on hobbit website, and no logs of anysort. However, it worked with BB server. Is there any suggested way to feed hobbit server combo data? Thank you very much in advance! MH =========== On Tue, Jun 24, 2008 at 05:01:42PM -0400, Haubrich,Myra [Ontario] wrote: > I am wondering if hobbit server can understand the keywork "combo" at > the beginning of a combo report file like bb server does? You could try it - and you would find that it works just like the Big Brother server. It's purely for compatibility with Big Brother based clients, though. The cpu-time needed to process a combo-message as opposed to multiple individual status-messages is nearly identical. Regards, Henrik

2 1

Custom graphing problem
by Mike.Wood＠kci1.com 26 Jun '08

26 Jun '08

Hi All, I need a bit of help with a custom graph. I'm going through the "howtograph.html", but I never get any results at all on the hobbit server. On my remote AIX client "dallas", I have this script: ------------------ #!/usr/bin/sh # Get numperm, minperm, maxperm, and calculate computational memory numperm=`vmstat -v|grep numperm|awk '{print $1}'` minperm=`vmstat -v|grep minperm|awk '{print $1}'` maxperm=`vmstat -v|grep maxperm|awk '{print $1}'` MSGFILE=$BBTMP/memperm.txt print "minperm: $minperm" > $MSGFILE print "maxperm: $maxperm" >> $MSGFILE print "numperm: $numperm" >> $MSGFILE $BB $BBDISP "status $MACHINE.memperm green `date` `cat $MSGFILE` " exit 0 ------------------- And this stanza in the clientlaunch.cfg: ------------------- [memperminfo] ENVFILE $HOBBITCLIENTHOME/etc/hobbitclient.cfg CMD $HOBBITCLIENTHOME/ext/memperm INTERVAL 5m ------------------- I have verified that the script is running on the client. However, I never get a memperm column on the Hobbit Server. What am I missing? Thanks! Mike Wood UNIX System Administrator Kinetic Concepts, Inc. 6103 Farinon Drive 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. ******************************************************************************

2 2

Trouble with Hobbit Alerts
by ncavanagh＠bimedia.net 26 Jun '08

26 Jun '08

Hello all, I hope that someone might be able to help me here. I have Hobbit server 4.2.0 installed on a Pc running Ubuntu 8. The server is configured to monitor a couple of test servers (one internal, and one external). Everything seems to function fine except for the alerts. My /etc/hobbit/hobbit-alerts.cfg file has the following lines: HOST=bimload1 SERVICE=cpu,disk,memory,procs,conn SCRIPT /usr/bin/sendalerts.pl n8 MAIL me at bimedia.net When I try to test this using: ./bbcmd hobbitd_alert --test bimload1 cpu I get the following: 2008-06-26 08:34:07 Using default environment file /usr/lib/hobbit/ client/etc/hobbitserver.cfg 00027518 2008-06-26 08:34:07 send_alert bimload1:cpu state Paging 00027518 2008-06-26 08:34:07 Matching host:service:page 'bimload1:cpu:' against rule line 120 00027518 2008-06-26 08:34:07 *** Match with 'HOST=bimload1 SERVICE=cpu,disk,memory,procs,conn' *** 00027518 2008-06-26 08:34:07 Matching host:service:page 'bimload1:cpu:' against rule line 120 00027518 2008-06-26 08:34:07 *** Match with 'HOST=bimload1 SERVICE=cpu,disk,memory,procs,conn' *** 00027518 2008-06-26 08:34:07 Script alert with command '/usr/bin/ sendalerts.pl' and recipient n8 00027518 2008-06-26 08:34:07 Matching host:service:page 'bimload1:cpu:' against rule line 120 00027518 2008-06-26 08:34:07 *** Match with 'HOST=bimload1 SERVICE=cpu,disk,memory,procs,conn' *** 00027518 2008-06-26 08:34:07 Mail alert with command 'mailx -s "Hobbit [12345] bimload1:cpu CRITICAL (RED)" me at bimedia.net' But when I check my mail logs and syslogs and hobbit logs it looks like the sendalert.pl script never fires off, and the email is never sent. I can send email using mailx from the command line and i can run the sendalert.pl script form the command line, and both produce the expected results. I have even sent the permissions for the sendalrt.pl script to be 777 (yes this is a security risk but for now I just wanted to make sure it was not a permissions issue). Why is the script not running not the email being sent via hobbit-alert? Am I missing anything? I would greatly appreciate any assistance in this matter Thank you, N8

1 0

[hobbit] Service SMTP on xxxxxx is not OK : Unexpected servic e response
by bewhite＠fellowes.com 26 Jun '08

26 Jun '08

We had a similar problem with our exchange server until I changed the bb-services file to be: [smtp] send "helo\r\nquit\r\n" expect "220" options banner port 25 Now the exchange response to smtp is much more stable, but my linux sendmail servers complain more often. I guess I need to make a smtpex with the above and restore the original smtp definition, which seemed to work better with Linux. Why can't MS just do things to standards? ......Bruce _____ From: Geoff Hallford [mailto:geoff.hallford at gmail.com] Sent: Wednesday, June 25, 2008 11:53 AM To: hobbit at hswn.dk Subject: Re: [hobbit] Service smtp on xxxxxx is not OK : Unexpected service response I have also seen this happen when the response is slow from the mail server to show the 220 message. It does show it but is slow for some reason. I create a new service called smtp-slow by adding the following and then configure the bb-hosts entry with 'smtp-slow' instead of smtp. Just a thought. This way it doesn't require a 220 response but if it gets a valid "connection" it is okay. If it can't open a connection, it goes red. It does limit the amount of app validation but it is the only thing that worked in my case. [smtp-slow] send "mail\r\nquit\r\n" options banner port 25 On Wed, Jun 25, 2008 at 12:31 PM, Andrew Clarke <andrew at bccit.co.uk <mailto:andrew at bccit.co.uk> > wrote: As it happens, there are PIXs between the hobbit server & all the Exchange servers. However, they've all been there for ages (years in some cases). I'll have a chat with our resident Cisco guru & see if he made any changes at our end around 6th June, when things stopped working. Andrew Clarke Support and Technical Development Officer BCC IT Solutions DDI: 0871 288 2036 Tel: +44 (0)1239 710823 http://www.bccit.co.uk/ <http://www.bccit.co.uk/> Providers of business to business IT Solutions & Services Back-up - Anti Virus - Security - Servers - Wireless Networks - VPN - Remote Monitoring - Project Consultation Hardware Provision Servers & Workstations - Disaster Recovery - Proactive Diagnostics - System Migration - Fibre CAT5e Server Administration - Forensics - Grant Appraisal - Network Management - System Appraisals - Support Contracts ... For full e-mail terms & conditions of use, please follow the link to the bcc web site http://www.bccitsolutions.co.uk/index.php?option=com_content <http://www.bccitsolutions.co.uk/index.php?option=com_content&task=view& amp;id=24&Itemid=1> &task=view&id=24&Itemid=1 BCC IT Solutions is a trading name of European Computer Units Limited, Registered in Cardiff, Company Registration No. 2651835 Registered Office: Unit B Station Road, Newcastle Emlyn, Carmarthenshire, Wales, SA38 9BX -----Original Message----- From: Rob MacGregor [mailto:rob.macgregor at gmail.com <mailto:rob.macgregor at gmail.com> ] You wouldn't happen to have a Cisco PIX/ASA between the hobbit host and the mail server, with it configured to do "fixup"? Those are known for breaking things ;) Rob MacGregor To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk <mailto:hobbit-unsubscribe at hswn.dk> -- 'If my answers frighten you then you should cease asking scary questions.' --Sam Jackson ---------------------------------------------------- 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. ----------------------------------------------------

4 4