November 2006 - Xymon - Xymon mailman web

[hobbit] client-local.cfg question
by cdanger2k＠yahoo.com 20 Nov '06

20 Nov '06

Thanks, I appreciate the heads up! ----- Original Message ---- From: Charles Jones <jonescr at cisco.com> To: hobbit at hswn.dk Sent: Monday, November 20, 2006 12:37:49 PM Subject: Re: [hobbit] client-local.cfg question Yes. See the CLASS option in the bb-hosts man page. Basically you can define a CLASS (group) of hosts, and then in client-local.cfg you can do: [classname] log:`find /var/log/httpd -type f -maxdepth 1` size<10000000 -Charles Chuck Danger wrote: Thought I would try this again, maybe I caught people on a weekend... ----- Original Message ---- From: Chuck Danger <cdanger2k at yahoo.com> To: hobbit at hswn.dk Sent: Friday, November 17, 2006 4:25:31 PM Subject: [hobbit] client-local.cfg question Is there a way in client-local.cfg to group hosts by "like" hostnames? For example, if I have: web01 web02 web03 web04 web05 And I want to setup file monitoring in /var/log/httpd, is there some way to group them so that I don't have to define each one? I'd like to do something like: [web*] log:`find /var/log/httpd -type f -maxdepth 1` size<10000000 instead of having to do: [web01] ... [web02] ... [web03] ... Thanks- Jay

1 0

[hobbit] client-local.cfg question
by cdanger2k＠yahoo.com 20 Nov '06

20 Nov '06

Thought I would try this again, maybe I caught people on a weekend... ----- Original Message ---- From: Chuck Danger <cdanger2k at yahoo.com> To: hobbit at hswn.dk Sent: Friday, November 17, 2006 4:25:31 PM Subject: [hobbit] client-local.cfg question Is there a way in client-local.cfg to group hosts by "like" hostnames? For example, if I have: web01 web02 web03 web04 web05 And I want to setup file monitoring in /var/log/httpd, is there some way to group them so that I don't have to define each one? I'd like to do something like: [web*] log:`find /var/log/httpd -type f -maxdepth 1` size<10000000 instead of having to do: [web01] ... [web02] ... [web03] ... Thanks- Jay

2 1

Enable/Disable by page name instead of hostname(s)?
by jonescr＠cisco.com 20 Nov '06

20 Nov '06

Is this possible to do? Heres the situation I am trying to avoid. When there is going to be maintenance activities on all of the servers (let's say a kernel upgrade). Putting them all into disable mode literally floods the alert pagers and email with hundreds of messages (they have NOTICE tag). I have my Hobbit pages built thus that the main page has subpages for Prod, Stage, and Dev. Each one of those has sub-pages for Web, App, and Database tiers. It would be nice if there was a way to enable/disable/acknowledge by page name, or a pre-defined groupname. Then for example one could disable the "Prod" group, and only one disable notice alert gets sent, instead of one for every host. Alternatively, perhaps an option could be added to allow a host(s) to be "silently" enabled/disabled, such that no alerts are sent to addresses using the NOTICE tag. Other ways I can think of to deal with this: 1. Do not use NOTICE tag. Not really an option, but could temporarily remove it before doing an enable/disable of many hosts in order to reduce the spam. 2. Use DOWNTIME in bb-hosts, instead of using the interactive enable/disable. This would work but requires manual editing of the bb-hosts file, which takes away the usefulness of enabling/disabling from the web interface. Both of the above provide a workaround, but are not optimal. -Charles

2 1

[patch] debian hobbit client init script
by gn1＠sanger.ac.uk 20 Nov '06

20 Nov '06

Hi, As I said earlier we want the hobbit client script to remove the temporary files and kill the rogue processes when stopped, mainly so the hobbit-client installation/upgrade process through cfengine runs smoothly. Here is a patch to the debian hobbit client init script to do the following actions when the hobbit client is stopped: - remove the /var/run/hobbitclient-runtime.cfg file - kill the unwanted vmstat processes that were left running - remove any files owned by hobbit from /tmp Patch may be applied on 4.2.0 with allinone patch applied, using: cd hobbit-4.2.0 patch -p1 < /path/to/patch/hobbitclient_clean.patch Comments to help polishing this solution are more than welcome! Cheers Gildas

2 2

Hobbit
by pretocj2＠staff.intekom.com 20 Nov '06

20 Nov '06

Hi Im am a sysadmin fo a ISP tipe environment. We have been using Nagios as our monitoring tool. I do admit that I like Nagios and prefer Nagios over using the Hobbit. Unfortuanately some people with too much say in our organisation decided to use Hobbit. These people are big hobbit fans and I even found them on mailing lists trying to convince Nagios users to switch to hobbit. Since one of my main functions is maintaining the monitoring environment, this decision was a bit of a let down. Hobbit is easier to configure, I give you that. We are running Nagios and Hobbit side by side and obviously when new servers are build nagios is left out and the kickstarts are built to install hobbit. But I could do some comparisons and I started collecting data that makes hobbit look not so reliable as Nagios. Nagios also has some functionality that I could not find in hobbit. I have never been on any hobbit mailing list. Since the decision was made, and that is now the road we are going. I would like to see some changes/additions, or make some suggestions to the hobbit. Where can I send my suggestions to? I am currently a bit anti hobbit, for good reason, but that could change, I hope. Regards CJ

3 2

no mailq graph?
by stefan.freisler＠aspect-online.de 20 Nov '06

20 Nov '06

Hi, i don't get an mailq graph. What is to do? rrd file will not created I use Hobbit 4.2.0 Part of my hobbitserver.cfg TEST2RRD="cpu=la,disk,inode,qtree,$PINGCOLUMN=tcp,http=tcp,dns=tcp,dig=tcp,time=ntpstat,vmstat,iostat,netstat,temperature,apache,apache1=apache,ntp=clock,bind,sendmail,mailq,nmailq=mailq,socks,bea,iishealth,citrix,bbgen,bbtest,bbproxy,hobbitd,files,procs=processes,ports,clock,lines,temp,oracle,sar-io,sar-rpca,sar-rqlla,memory,sar-cpu,sar-cswitch,sar-kernel,r-huk24-a,r-huk24-b,r-huk24-c,r-div,r-allianz,r-axakfz,r-directline,r-ineas,r-sdirekt,r-ineas,spam" # This defines which RRD files to include on the "trends" column webpage, # and the order in which they appear. GRAPHS="la,disk,inode,qtree,files,processes,memory,users,vmstat,iostat,tcp.http,tcp,ncv,netstat,ifstat,mrtg::1,ports,temperature,ntpstat,apache,bind,sendmail,mailq,socks,bea,iishealth,citrix,bbgen,bbtest,bbproxy,hobbitd,clock,lines,temp,oracle,sar-io,sar-rpca,sar-rqlla,sar-memory,sar-cpu,sar-cswitch,sar-kernel,r-directline,r-huk24,r-huk24b,r-huk24c,r-div,r-allianz,r-axakfz,r-directline,r-ineas,r-sdirekt,response,spam" thx for Help Stefan

1 0

[hobbit] dbcheck
by fduranti＠q8.it 20 Nov '06

20 Nov '06

To run it on the client you've to change 2 variables at the start of dbcheck.pl file: $Hobbit_fd_lib::HOBBIT = 1; $Hobbit_fd_lib::USEBBHGREP = 0; Can you run a dbcheck.pl -dd (it should create some debugging data that could be usefull) and send it to me ? It's possible that the "client" part that don't use bbhostgrep is not working right for some reason, I didn't tested it really much because I use dbcheck.pl on the hobbit server from which I monitor all the databases with just one script. Francesco > -----Original Message----- > From: Iain Conochie [mailto:iain at shihad.org] > Sent: Monday, November 20, 2006 2:27 PM > To: hobbit at hswn.dk > Subject: RE: [hobbit] dbcheck > > On Mon, 2006-11-20 at 08:19 -0500, Aiello, Steve (GE, Corporate, > consultant) wrote: > > Did you add the dbcheck flag to the device entry in the > bb-hosts file ? > > This is the bb-hosts file on the client > > 192.168.1.87 breed.int.servista.com # noconn dbcheck > > > Iain > > > > > > > > -----Original Message----- > > > From: Iain Conochie [mailto:iain at shihad.org] > > > Sent: Monday, November 20, 2006 8:10 AM > > > To: hobbit at hswn.dk > > > Subject: [hobbit] dbcheck > > > > > > > > > Morning, > > > > > > I have having a little issue getting the dbcheck script > to check > > > our mysql database. I am running the script from the client side. > > > Below is the config for dbcheck.ini. Local config replaced with > > > XXXXXXX > > > > > > [default] > > > warningout =send > > > conn = conn > > > include = > > > defaultgroup = > > > timeoutdbconn = 20s > > > timeoudbtselect = 1h > > > fastcheck = 30m > > > mediumcheck = 2h > > > slowcheck = 6h > > > bbdisplay = XXXXXX.XXXXXXX.XXXXXXXXXXXX.XXXXXX > > > bbsendtype = multi # multi single or both > > > singlesendashost = no > > > bbfulltestcolumn = DBCheck > > > bbmysqlperfcolumn = MySQLPerf > > > bbconncolumn = ChkConn > > > bbauditcolumn = Audit > > > bbtablespacecolumn = TblSpace > > > bbextentcolumn = Extent > > > bbhitcachecolumn = HitCache > > > bbmemreqcolumn = MemReq > > > bbrollbackcolumn = RollBack > > > bbinvobjcolumn = InvObj > > > bblockscolumn = Locks > > > bbsessioncolumn = Session > > > bbsessionuserlist = yes > > > username = XXXXXXXXXXXXXXXXX > > > password = XXXXXXXXXXXXXXXXX > > > tblspcheck = 90:95 > > > tblspreportautoext = 90:95 > > > pinlibcheck = 95:90 > > > blbuffcheck = 95:90 > > > hitcachecheck = 95:90 > > > rowcachecheck = 95:90 > > > extentcheck = 90:95 > > > rollbackcheck = 90:95 > > > sessioncheck = 90:95 > > > processcheck = 90:95 > > > rb.SYSTEM = 101:101 # don't report SYSTEM > > > rollback problem > > > for all databases > > > invobjwarn = ::: # Report all as yellow > instead of red > > > invobjignore = > > > dbtype = mysql > > > > > > > > > [XXXXXX.XXXXXXXXX.XXXXXXXXXX.XXX] > > > dbtype = mysql > > > hostname = localhost > > > port = 3306 > > > username = XXXXXXXXXXXXXX > > > password = XXXXXXXXXXXXXX > > > > > > [CUSTOM testcheck] > > > > > > [CUSTOM unumber] > > > > > > The results I get back are: > > > > > > Test Time = 00:00:00 > > > Host Checked = 0 > > > Status Msg = 0 > > > > > > No Colors > > > > > > No Events > > > > > > No Types > > > > > > No Hosts > > > > > > No Hosts > > > > > > Errors > > > No General Errors > > > No Hosts Errors > > > > > > Warnings > > > No General Warnings > > > No Hosts Warnings > > > > > > > > > dbcheck.pl version 1.07 - column dbcheck_fast lifetime > 30m, tested > > > in ~ 00:00:00 (max 00:01:00) > > > > > > Can someone show me what I am missing here please? > > > > > > Cheers > > > > > > Iain > > > > > > > > > > > > > > > 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

Acknowledge Purple Alerts over acknowledgment webpage not possible
by roland.graeub＠rtc.ch 20 Nov '06

20 Nov '06

Hello All Its not possible to acknowledge an purple alert over the acknowledgment webpage. In my acknowledge webpage overview only yellow and red status are displayed, purple is missing.. Anybody else have the same issue ? Acking a purple status with mail works fine. Using 4.2.0. Regards Roland

1 0

dbcheck
by iain＠shihad.org 20 Nov '06

20 Nov '06

Morning, I have having a little issue getting the dbcheck script to check our mysql database. I am running the script from the client side. Below is the config for dbcheck.ini. Local config replaced with XXXXXXX [default] warningout =send conn = conn include = defaultgroup = timeoutdbconn = 20s timeoudbtselect = 1h fastcheck = 30m mediumcheck = 2h slowcheck = 6h bbdisplay = XXXXXX.XXXXXXX.XXXXXXXXXXXX.XXXXXX bbsendtype = multi # multi single or both singlesendashost = no bbfulltestcolumn = DBCheck bbmysqlperfcolumn = MySQLPerf bbconncolumn = ChkConn bbauditcolumn = Audit bbtablespacecolumn = TblSpace bbextentcolumn = Extent bbhitcachecolumn = HitCache bbmemreqcolumn = MemReq bbrollbackcolumn = RollBack bbinvobjcolumn = InvObj bblockscolumn = Locks bbsessioncolumn = Session bbsessionuserlist = yes username = XXXXXXXXXXXXXXXXX password = XXXXXXXXXXXXXXXXX tblspcheck = 90:95 tblspreportautoext = 90:95 pinlibcheck = 95:90 blbuffcheck = 95:90 hitcachecheck = 95:90 rowcachecheck = 95:90 extentcheck = 90:95 rollbackcheck = 90:95 sessioncheck = 90:95 processcheck = 90:95 rb.SYSTEM = 101:101 # don't report SYSTEM rollback problem for all databases invobjwarn = ::: # Report all as yellow instead of red invobjignore = dbtype = mysql [XXXXXX.XXXXXXXXX.XXXXXXXXXX.XXX] dbtype = mysql hostname = localhost port = 3306 username = XXXXXXXXXXXXXX password = XXXXXXXXXXXXXX [CUSTOM testcheck] [CUSTOM unumber] The results I get back are: Test Time = 00:00:00 Host Checked = 0 Status Msg = 0 No Colors No Events No Types No Hosts No Hosts Errors No General Errors No Hosts Errors Warnings No General Warnings No Hosts Warnings dbcheck.pl version 1.07 - column dbcheck_fast lifetime 30m, tested in ~ 00:00:00 (max 00:01:00) Can someone show me what I am missing here please? Cheers Iain

2 2

client-local.cfg question
by cdanger2k＠yahoo.com 20 Nov '06

20 Nov '06

Is there a way in client-local.cfg to group hosts by "like" hostnames? For example, if I have: web01 web02 web03 web04 web05 And I want to setup file monitoring in /var/log/httpd, is there some way to group them so that I don't have to define each one? I'd like to do something like: [web*] log:`find /var/log/httpd -type f -maxdepth 1` size<10000000 instead of having to do: [web01] ... [web02] ... [web03] ... Thanks- Jay

3 2