October 2005 - Xymon - Xymon mailman web

[hobbit] Custom graphs on the status page
by rsmrcina＠wi.rr.com 31 Oct '05

31 Oct '05

I have a custom test that I call sp_util (for Spool Utilization, unrelated to the mail spool on Linux/Unix systems). On some hosts, the graph for sp_util is called sp_util.rrd, it contains one data point and that graph show up just fine. On other hosts, there are two data points and to (hopefully) make things easier called the graph pqueue.rrd for those hosts. For these other hosts it is still considered spool utilization, so the column is called sp_util. The intent is to display the sp_util graph under some hosts and the pqueue graph under the other hosts (based on where the graphed data applies). I've tried varying combinations of sp_util and pqueue in the TEST2RRD and GRAPHS settings, with no luck (one or the other shows up under their respective host, never both). Can this be done and if so, how? Thanks. Henrik Storner wrote: > In <43318B96.7040405 at saffron.net> Matti Klock <matti at saffron.net> writes: > > >>Is it possible to add graphs of custom tests to a status page as well as >>the trends page? > > > The current way of configuring what graphs goes with what status is > very ad-hoc and easy to break. > > The way it works now is as follows: > > * You have a status called "foo", which is tracked by the RRD file > "foo.rrd". > > * This is mapped to an RRD file via the TEST2RRD setting. So you must > add "foo" to the list in the TEST2RRD setting. If the RRD filename > is not the same as the status names (e.g. the status column is called > "foo", but the rrd-file is "bar.rrd", then it must be done as "foo=bar"). > > * Then the GRAPHS setting is consulted. This is the list of what graphs > go on the "trends" column, but it is *also* (in this is the messy part) > used to decide what graphs are available at all, and how to split > graphs from multiple RRD files (e.g. disk graphs) into multiple > graphs. So you must have the graph listed here as well. > > To confuse matters further, there's also the TRENDS setting in the bb-hosts > file, which is used to modify what graphs show up on the "trends" page. > > All in all, it's confusing enough that even I get it wrong occasionally. > > > The way I'd like it to work is to get rid of the TEST2RRD and GRAPHS > settings. Instead, I'd like to have a direct mapping from the status > column name to the graph definition in hobbitgraphs.cfg. That's a clean-up > job I need to do sometime soon. > > > Regards, > Henrik > > > To unsubscribe from the hobbit list, send an e-mail to > hobbit-unsubscribe at hswn.dk > > > -- Rich Smrcina VM Assist, Inc. Main: (262)392-2026 Cell: (414)491-6001 Ans Service: (360)715-2467 rich.smrcina at vmassist.com Catch the WAVV! http://www.wavv.org WAVV 2006 - Chattanooga, TN - April 7-11, 2006

1 0

Mailinglist and website offline Wednesday Nov 2nd
by henrik＠hswn.dk 31 Oct '05

31 Oct '05

Due to some change with my DSL connection, the Hobbit mailinglist and website (www.hswn.dk) will probably be off-line for a few hours on Wednesday while the telco people do what has to be done. I don't have any fixed time-schedule for it, but presumably it will be sometime between 08:00 - 16:00 CET (07:00 - 15:00 UTC). Regards, Henrik

1 0

[hobbit] Hobbit sending HTML emails?
by Robert.Taylor＠HendrickAuto.com 29 Oct '05

29 Oct '05

Another option is to write custom alert script and set that as the action for the alert. I have written just such a script so that my MRTG and UPS alerts come in as HTML and include the appropriate graphs. I would be happy to email it to you if you would like, just be forwarned that I am not a developer so run at your own risk ;-). Robert -----Original Message----- From: Scheblein, Adam [mailto:adam.scheblein at marquette.edu] Sent: Fri 10/28/2005 4:56 PM To: hobbit at hswn.dk Cc: Subject: RE: [hobbit] Hobbit sending HTML emails? There is a way to make your mail program send html e-mails, however, if you check your version to make sure it has the -a, then under the mail variable, set it to be (and it shouldn't matter if you use mail or mailx): mailx -a "Content-type: text/html;" Adam -----Original Message----- From: Henrik Stoerner [mailto:henrik at hswn.dk] Sent: Friday, October 28, 2005 2:53 PM To: hobbit at hswn.dk Subject: Re: [hobbit] Hobbit sending HTML emails? On Fri, Oct 28, 2005 at 12:46:58PM -0700, Charles Jones (charljon) wrote: > > I've setup a Hobbit 4.1.2 instance and have been getting emails from it > that have HTML tags in them, but aren't recognized by either MS Outlook > or Mozilla Thunderbird as an HTML-email. So, either hobbit isn't > properly setting the header on the mails it sends, or something about > the alert output is causing the problem. Unless you changed something special in your setup, Hobbit handles all alerts as plain text, and just feeds them into your local "mail" utility. > The alert below is an example of how the emails I'm getting look, with > bare HTML tags in the message. I'm 99% sure this is an external add-on test that generates a status in HTML format. Regards, Henrik 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

Benoit Vanderbeck/vlafo is niet op kantoor.
by benoit.vanderbeck＠vlafo.be 29 Oct '05

29 Oct '05

I will be out of the office starting 29/10/2005 and will not return until 07/11/2005. Ik ben niet op het kantoor. Voor dringende zaken ivm het systeembeheer van het Fonds kan u een mail sturen naar exploitatie at vlafo.be. Benoit

1 0

clientlaunch.cfg vs. hobbittasks.cfg
by rmunsch＠solutionsforprogress.com 28 Oct '05

28 Oct '05

Hello! Old bb user, new to setting up a set of clients with a hobbit server in a new environment. There have been, of course, some oddities... The first line in clientlaunch.cfg cheerfully maintains: # The hobbittasks.cfg file is loaded by "hobbitlaunch". I gather this is a typo/leftover, and should now start "The clientlaunch.cfg..."? If not, I am powerfully confused XD Speaking of confused... i am using a Debian system and think this is a Debian/bash conspiracy on my end, but my hobbitclient.log shows nothing but scads of shell-init: error retrieving current directory: getcwd: cannot access parent directories: No such file or directory when i attempt to start the client, and nothing else. The archives don't seem to yield a hit on this. Kernel is 2.6.11, smp (on the client in question. the server install and function appear fine). Thanks! -- Rob Munsch Systems Analyst, Solutions for Progress http://www.solutionsforprogress.com

2 3

[hobbit] Hobbit sending HTML emails?
by adam.scheblein＠marquette.edu 28 Oct '05

28 Oct '05

There is a way to make your mail program send html e-mails, however, if you check your version to make sure it has the -a, then under the mail variable, set it to be (and it shouldn't matter if you use mail or mailx): mailx -a "Content-type: text/html;" Adam -----Original Message----- From: Henrik Stoerner [mailto:henrik at hswn.dk] Sent: Friday, October 28, 2005 2:53 PM To: hobbit at hswn.dk Subject: Re: [hobbit] Hobbit sending HTML emails? On Fri, Oct 28, 2005 at 12:46:58PM -0700, Charles Jones (charljon) wrote: > > I've setup a Hobbit 4.1.2 instance and have been getting emails from it > that have HTML tags in them, but aren't recognized by either MS Outlook > or Mozilla Thunderbird as an HTML-email. So, either hobbit isn't > properly setting the header on the mails it sends, or something about > the alert output is causing the problem. Unless you changed something special in your setup, Hobbit handles all alerts as plain text, and just feeds them into your local "mail" utility. > The alert below is an example of how the emails I'm getting look, with > bare HTML tags in the message. I'm 99% sure this is an external add-on test that generates a status in HTML format. Regards, Henrik To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk

1 0

Hobbit sending HTML emails?
by charljon＠cisco.com 28 Oct '05

28 Oct '05

I've setup a Hobbit 4.1.2 instance and have been getting emails from it that have HTML tags in them, but aren't recognized by either MS Outlook or Mozilla Thunderbird as an HTML-email. So, either hobbit isn't properly setting the header on the mails it sends, or something about the alert output is causing the problem. The alert below is an example of how the emails I'm getting look, with bare HTML tags in the message. The alert text itself is from the "bb-roracle.ksh" script which checks the health of an Oracle database. I have sanitized the alert text to remove any private information but otherwise is the same text that I received. -----Original Message----- From: hobbit at mydomain.com [mailto:hobbit at mydomain.com] Sent: Friday, October 28, 2005 1:11 PM To: jonescr at cisco.com Subject: Hobbit [70835] oradb5.mydomain.com:oradb CRITICAL (RED) red Fri Oct 28 12:03:27 MST 2005 </pre> Process Check (MYDB)<hr> &green Database MYDB UP processes: ckpt dbw0 lgwr pmon reco smon Database Checks (MYDB)<hr> &green Database check: MYDB is up User Count Check (MYDB)<hr> Oracle Users in MYDB: 6 User Check (MREDW)<hr> <pre> username osuser machine term oracle oradb5 UNKNOWN oracle oradb5 UNKNOWN oracle oradb5 UNKNOWN oracle oradb5 UNKNOWN oracle oradb5 UNKNOWN oracle oradb5 UNKNOWN DB1 oracle oradb5.mydomain unknown DB1 oracle oradb5.mydomain unknown BB bb oradb5 DB1 oracle oradb5.mydomain unknown DB1 oracle oradb5.mydomain unknown 11 rows selected. </pre> Tablespace Check (MYDB)<hr> &green Tablespace MYDB:TEST totals 3000.0Mb and is 26% used. &green Tablespace MYDB:UNDO totals 1000.0Mb and is 52% used. &green Tablespace MYDB:TABLE1 totals 5000.0Mb and is 44% used. &green Tablespace MYDB:TEMP totals 13000.0Mb and is 0% used. &green Tablespace MYDB:SYSTEM totals 1000.0Mb and is 14% used. &green Tablespace MYDB:TABLE2 totals 2000.0Mb and is 9% used. &green Tablespace MYDB:TABLE3 totals 11218.0Mb and is 81% used. &green Tablespace MYDB:TABLE4 totals 4000.0Mb and is 29% used. &green Tablespace MYDB:TABLE5 totals 1000.0Mb and is 39% used. &green Tablespace MYDB:TABLE6 totals 10000.0Mb and is 0% used. &green Tablespace MYDB:TABLE7 totals 12825.0Mb and is 89% used. &green Tablespace MYDB:TABLE8 totals 5230.0Mb and is 72% used. &green Total Tablespace including Temp: 69273Mb Extent Check (MYDB)<hr> &green No objects are exceeding 60%. Extent test ok. &clear Shadow check: Disabled &clear Pin Hit Ratio check: Disabled &clear Sql Area Hit Ratio check: Disabled &clear Block Buffer Hit Ratio check: Disabled Shared Memory Check (MYDB)<hr> &green Database MYDB: No Request Failures. Free Space is 1311K, AVG Free Space is 328K, Used Space is 0K, AVG Used Space is 0K. &clear RollBack check: Disabled &clear Invalid Object check: Disabled Deadlock Check (MYDB)<hr> &red SID 0 Blocker 1 Requester 2. &red SID 3 Blocker 4 Requester 5. Process Check (MYDB2)<hr> &green Database MYDB2 UP processes: ckpt dbw0 lgwr pmon reco smon Database Checks (MYDB2)<hr> &green Database check: MYDB2 is up User Count Check (MYDB2)<hr> Oracle Users in MYDB2: 0 User Check (MYDB2)<hr> <pre> username osuser machine term oracle oradb5 UNKNOWN oracle oradb5 UNKNOWN oracle oradb5 UNKNOWN oracle oradb5 UNKNOWN oracle oradb5 UNKNOWN oracle oradb5 UNKNOWN BB bb oradb5 7 rows selected. </pre> &clear Tablespace check: Disabled Extent Check (MYDB2)<hr> &green No objects are exceeding 60%. Extent test ok. &clear Shadow check: Disabled &clear Pin Hit Ratio check: Disabled &clear Sql Area Hit Ratio check: Disabled &clear Block Buffer Hit Ratio check: Disabled Shared Memory Check (MYDB2)<hr> &green Database MYDB2: No Request Failures. Free Space is 1054K, AVG Free Space is 88K, Used Space is 0K, AVG Used Space is 0K. &clear RollBack check: Disabled &clear Invalid Object check: Disabled Deadlock Check (MYDB2)<hr> &green Database MYDB2: No deadlocks detected. Process Check (MYDB3)<hr> &green Database MYDB3 UP processes: ckpt dbw0 lgwr pmon reco smon Database Checks (MYDB3)<hr> &green Database check: MYDB3 is up User Count Check (MYDB3)<hr> Oracle Users in MYDB3: 0 User Check (MYDB3)<hr> <pre> username osuser machine term oracle oradb5 UNKNOWN oracle oradb5 UNKNOWN oracle oradb5 UNKNOWN oracle oradb5 UNKNOWN oracle oradb5 UNKNOWN oracle oradb5 UNKNOWN SYS bb oradb5 7 rows selected. </pre> &clear Tablespace check: Disabled Extent Check (MYDB3)<hr> &green No objects are exceeding 60%. Extent test ok. &clear Shadow check: Disabled &clear Pin Hit Ratio check: Disabled &clear Sql Area Hit Ratio check: Disabled &clear Block Buffer Hit Ratio check: Disabled Shared Memory Check (MYDB3)<hr> &green Database MYDB3: No Request Failures. Free Space is 1311K, AVG Free Space is 328K, Used Space is 0K, AVG Used Space is 0K. &clear RollBack check: Disabled &clear Invalid Object check: Disabled Deadlock Check (MYDB3)<hr> &green Database MYDB3: No deadlocks detected. <center><hr>bb-roracle.ksh Version: 2.06</center> See http://hobbit.mydomain.com/hobbit-cgi/bb-hostsvc.sh?HOSTSVC=oradb5,mydom ain,com.oradb

2 1

Minor graphing quandry
by rsmrcina＠wi.rr.com 28 Oct '05

28 Oct '05

I have external tests running for two different operating systems. They are both reporting 'procs' back to Hobbit. One one system it is really the number of jobs running and on the other system it is the number of users (virtual machines) running. I purposely used procs since it has essentially the same meaning. For the status display that works out great. Now I want to graph the results. Ultimately I would like to have one graph text read 'Running jobs' and for the other 'Logged on machines'. I could certainly leave them both as procs and be done with it (and that works just fine), but the extra touch would be nice. In the extra script I am creating an rrd called procs.rrd for each host. I just can't seem to get the actual graph created using one graph definition. Is there an example of something like this currently available? Or does anyone have any insight as to how I might do this? Thanks. -- Rich Smrcina VM Assist, Inc. Main: (262)392-2026 Cell: (414)491-6001 Ans Service: (360)715-2467 rich.smrcina at vmassist.com Catch the WAVV! http://www.wavv.org WAVV 2006 - Chattanooga, TN - April 7-11, 2006

1 0

Test a host at a different interval
by Jason.Chambers＠geosoft.com 28 Oct '05

28 Oct '05

Hey Guys, The manual suggests that I can run test on a different time interval than the rest, but I cannot seem to find what the tag is. IE. I want ping a few connections every 30 minutes instead of every 5. Help? Jason Chambers IT Helpdesk Support eosoft Inc. 85 Richmond St. West - 8th Floor Toronto, Ontario, Canada M5H 2C9 Tel: 416-369-0111 x344 Fax: 416-369-9599 www.geosoft.com

1 0

badntp
by tlp-hobbit＠holme-pedersen.dk 28 Oct '05

28 Oct '05

Is this option still not recognised by hobbit ? /Thoams

1 0