-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
I'm pretty sure this is the trouble. Ladies and gentlemen, i am stumped. Following all the instructions in the docs, manpages, faqs and alien communications, as well as the very helpful samples provided by list members... no new file ever gets created in ~/data/rrd/<hostname>.
I have status, so i know client info is reaching server. I have copied almost verbatim the sample server-side script needed to parse this info and spit it back out to RRD. And i have eschewed NCV and gone in favor of hobbitd_channel --options to make this all work right, or so I thought.
If you make a custom script and the rrd file never happens, where would you look first? The serverside script itself? hobbitlaunch. or
- -graph.cfg options? Somewhere else? Mars? I'd be willing to check Mars at this point.
Rob Munsch Solutions For Progress IT www.solutionsforprogress.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFN8gNBvBcJFK6xYURAlP2AJ49r6WJg4Pg10Nc1JgYDEbmkNvbIgCcD58W lnm6fVJIn7V4PQqvSOktFGQ= =iyfa -----END PGP SIGNATURE-----
Rob, if you implemented the script example I posted, you should be able to look in the script log to see if it is doing anything. Just remember not to refer to NCV anywhere in the configuration.
The log will tell you if the script is being called at all -- you can enable any debug you want -- just make sure you write it to the log, since you will never see STDOUT. You can even cat the incoming file out to the log.
This is what I had to do to get it working -- sort of organically.
GLH
-----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Thursday, October 19, 2006 1:47 PM To: hobbit at hswn.dk Subject: [hobbit] RRD data not being created.
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
I'm pretty sure this is the trouble. Ladies and gentlemen, i am stumped. Following all the instructions in the docs, manpages, faqs and alien communications, as well as the very helpful samples provided by list members... no new file ever gets created in ~/data/rrd/<hostname>.
I have status, so i know client info is reaching server. I have copied almost verbatim the sample server-side script needed to parse this info and spit it back out to RRD. And i have eschewed NCV and gone in favor of hobbitd_channel --options to make this all work right, or so I thought.
If you make a custom script and the rrd file never happens, where would you look first? The serverside script itself? hobbitlaunch. or
- -graph.cfg options? Somewhere else? Mars? I'd be willing to check Mars at this point.
Rob Munsch Solutions For Progress IT www.solutionsforprogress.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFN8gNBvBcJFK6xYURAlP2AJ49r6WJg4Pg10Nc1JgYDEbmkNvbIgCcD58W lnm6fVJIn7V4PQqvSOktFGQ= =iyfa -----END PGP SIGNATURE-----
To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Hubbard, Greg L wrote:
Rob, if you implemented the script example I posted, you should be able to look in the script log to see if it is doing anything. Just remember not to refer to NCV anywhere in the configuration.
The log will tell you if the script is being called at all -- you can enable any debug you want -- just make sure you write it to the log, since you will never see STDOUT. You can even cat the incoming file out to the log.
This is what I had to do to get it working -- sort of organically.
I have .rrd files! huzzah. Now i just need to get the actual graph to appear, instead of a text link.
Also, something Odd is happening such that variables are not being expanded. Frex,
TESTNAME="$2" echo $TESTNAME.rrd
is coming out as just
.rrd
o_O still, it's progress :D
GLH
-----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Thursday, October 19, 2006 1:47 PM To: hobbit at hswn.dk Subject: [hobbit] RRD data not being created.
I'm pretty sure this is the trouble. Ladies and gentlemen, i am stumped. Following all the instructions in the docs, manpages, faqs and alien communications, as well as the very helpful samples provided by list members... no new file ever gets created in ~/data/rrd/<hostname>.
I have status, so i know client info is reaching server. I have copied almost verbatim the sample server-side script needed to parse this info and spit it back out to RRD. And i have eschewed NCV and gone in favor of hobbitd_channel --options to make this all work right, or so I thought.
If you make a custom script and the rrd file never happens, where would you look first? The serverside script itself? hobbitlaunch. or -graph.cfg options? Somewhere else? Mars? I'd be willing to check Mars at this point.
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
Rob Munsch Solutions For Progress IT www.solutionsforprogress.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFN9ZuBvBcJFK6xYURAtk3AJ9PBk9SZedYwNjprD+fBrl3hSAmOgCfWZSO i1Lznw80STQW28ilqmHyuX0= =XMll -----END PGP SIGNATURE-----
Is this when you are running the script by hand? $2 is the second parameter. So to run the script by hand, you have to:
$ ./script node test some.test.file.name
And then you *should* see "test.rrd" written out.
The way I tested this was to capture the output of a client side script (just cut from the browser and paste in a text file), then run it by hand using the above method. The RRD commands are written to the terminal, and any griping is in the log.
Glad you are making progress.
GLH
-----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Thursday, October 19, 2006 2:48 PM To: hobbit at hswn.dk Subject: Re: [hobbit] RRD data not being created.
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Hubbard, Greg L wrote:
Rob, if you implemented the script example I posted, you should be able to look in the script log to see if it is doing anything. Just remember not to refer to NCV anywhere in the configuration.
The log will tell you if the script is being called at all -- you can enable any debug you want -- just make sure you write it to the log, since you will never see STDOUT. You can even cat the incoming file out to the log.
This is what I had to do to get it working -- sort of organically.
I have .rrd files! huzzah. Now i just need to get the actual graph to appear, instead of a text link.
Also, something Odd is happening such that variables are not being expanded. Frex,
TESTNAME="$2" echo $TESTNAME.rrd
is coming out as just
.rrd
o_O still, it's progress :D
GLH
-----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Thursday, October 19, 2006 1:47 PM To: hobbit at hswn.dk Subject: [hobbit] RRD data not being created.
I'm pretty sure this is the trouble. Ladies and gentlemen, i am stumped. Following all the instructions in the docs, manpages, faqs and alien communications, as well as the very helpful samples provided
by list members... no new file ever gets created in ~/data/rrd/<hostname>.
I have status, so i know client info is reaching server. I have copied almost verbatim the sample server-side script needed to parse this info and spit it back out to RRD. And i have eschewed NCV and gone in favor of hobbitd_channel --options
to make this all work right, or so I thought.
If you make a custom script and the rrd file never happens, where would you look first? The serverside script itself? hobbitlaunch. or
-graph.cfg options? Somewhere else? Mars? I'd be willing to check Mars at this point.
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
Rob Munsch Solutions For Progress IT www.solutionsforprogress.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFN9ZuBvBcJFK6xYURAtk3AJ9PBk9SZedYwNjprD+fBrl3hSAmOgCfWZSO i1Lznw80STQW28ilqmHyuX0= =XMll -----END PGP SIGNATURE-----
To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Hubbard, Greg L wrote:
Is this when you are running the script by hand? $2 is the second parameter. So to run the script by hand, you have to:
$ ./script node test some.test.file.name
And then you *should* see "test.rrd" written out.
it looks like i have .rrd files now, but the data points are all NaN. Still no graph, not even a blank one. I'll see what kind of raw output i get (i must have broken parsing while messing with this...).
The way I tested this was to capture the output of a client side script (just cut from the browser and paste in a text file),
Oh cute. Okay, i'll try that now. Thanks!
then run it by
hand using the above method. The RRD commands are written to the terminal, and any griping is in the log.
Glad you are making progress.
GLH
-----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Thursday, October 19, 2006 2:48 PM To: hobbit at hswn.dk Subject: Re: [hobbit] RRD data not being created.
Hubbard, Greg L wrote:
Rob, if you implemented the script example I posted, you should be able to look in the script log to see if it is doing anything. Just remember not to refer to NCV anywhere in the configuration.
The log will tell you if the script is being called at all -- you can enable any debug you want -- just make sure you write it to the log, since you will never see STDOUT. You can even cat the incoming file out to the log.
This is what I had to do to get it working -- sort of organically.
I have .rrd files! huzzah. Now i just need to get the actual graph to appear, instead of a text link.
Also, something Odd is happening such that variables are not being expanded. Frex,
TESTNAME="$2" echo $TESTNAME.rrd
is coming out as just
.rrd
o_O still, it's progress :D
GLH
-----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Thursday, October 19, 2006 1:47 PM To: hobbit at hswn.dk Subject: [hobbit] RRD data not being created.
I'm pretty sure this is the trouble. Ladies and gentlemen, i am stumped. Following all the instructions in the docs, manpages, faqs and alien communications, as well as the very helpful samples provided
by list members... no new file ever gets created in ~/data/rrd/<hostname>. I have status, so i know client info is reaching server. I have copied almost verbatim the sample server-side script needed to parse this info and spit it back out to RRD. And i have eschewed NCV and gone in favor of hobbitd_channel --options
to make this all work right, or so I thought.
If you make a custom script and the rrd file never happens, where would you look first? The serverside script itself? hobbitlaunch. or
-graph.cfg options? Somewhere else? Mars? I'd be willing to check Mars at this point.
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
To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk
Rob Munsch Solutions For Progress IT www.solutionsforprogress.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFOO2GBvBcJFK6xYURAs8+AJ9SATCMb1MNDyuynM9p7ck7DLnDHgCgjp+o GbFPoIYazzxqkSCeXeIOgDo= =1gPl -----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Hubbard, Greg L wrote:
Is this when you are running the script by hand? $2 is the second parameter. So to run the script by hand, you have to:
$ ./script node test some.test.file.name
And then you *should* see "test.rrd" written out.
Weird. running the test by hand gets me this hobbit at randomaccess ~/server/ext $ ./cputemp.sh currant.tbb cputemp currant.tbb.cputemp.out DS:cputemp:GAUGE:600:0:U cputemp.rrd 100
but cputemp.log in /var/log/hobbit shows only
DS:cputemp:GAUGE:600:0:U cputemp.rrd
DS:cputemp:GAUGE:600:0:U cputemp.rrd
DS:cputemp:GAUGE:600:0:U cputemp.rrd
No temp! rrdtool dump on cputemp.rrd shows me a buncha <!-- 2006-10-20 00:00:00 EDT / 1161316800 --> <row><v> NaN </v></row> <!-- 2006-10-20 02:00:00 EDT / 1161324000 --> <row><v> NaN </v></row> <!-- 2006-10-20 04:00:00 EDT / 1161331200 --> <row><v> NaN </v></row> <!-- 2006-10-20 06:00:00 EDT / 1161338400 --> <row><v> NaN </v></row> <!-- 2006-10-20 08:00:00 EDT / 1161345600 --> <row><v> NaN </v></row> <!-- 2006-10-20 10:00:00 EDT / 1161352800 --> <row><v> NaN </v></row> <!-- 2006-10-20 12:00:00 EDT / 1161360000 --> <row><v> NaN </v></row> <!-- 2006-10-20 14:00:00 EDT / 1161367200 --> <row><v> NaN </v></row>
The way I tested this was to capture the output of a client side script (just cut from the browser and paste in a text file), then run it by hand using the above method. The RRD commands are written to the terminal, and any griping is in the log.
Glad you are making progress.
GLH
-----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Thursday, October 19, 2006 2:48 PM To: hobbit at hswn.dk Subject: Re: [hobbit] RRD data not being created.
Hubbard, Greg L wrote:
Rob, if you implemented the script example I posted, you should be able to look in the script log to see if it is doing anything. Just remember not to refer to NCV anywhere in the configuration.
The log will tell you if the script is being called at all -- you can enable any debug you want -- just make sure you write it to the log, since you will never see STDOUT. You can even cat the incoming file out to the log.
This is what I had to do to get it working -- sort of organically.
I have .rrd files! huzzah. Now i just need to get the actual graph to appear, instead of a text link.
Also, something Odd is happening such that variables are not being expanded. Frex,
TESTNAME="$2" echo $TESTNAME.rrd
is coming out as just
.rrd
o_O still, it's progress :D
GLH
-----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Thursday, October 19, 2006 1:47 PM To: hobbit at hswn.dk Subject: [hobbit] RRD data not being created.
I'm pretty sure this is the trouble. Ladies and gentlemen, i am stumped. Following all the instructions in the docs, manpages, faqs and alien communications, as well as the very helpful samples provided
by list members... no new file ever gets created in ~/data/rrd/<hostname>. I have status, so i know client info is reaching server. I have copied almost verbatim the sample server-side script needed to parse this info and spit it back out to RRD. And i have eschewed NCV and gone in favor of hobbitd_channel --options
to make this all work right, or so I thought.
If you make a custom script and the rrd file never happens, where would you look first? The serverside script itself? hobbitlaunch. or
-graph.cfg options? Somewhere else? Mars? I'd be willing to check Mars at this point.
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
To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk
Rob Munsch Solutions For Progress IT www.solutionsforprogress.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFORPeBvBcJFK6xYURAgrAAJ4g3WBP+uTdrzeoCAwEigHVYboZ0QCeMkTw xAoHPHSGiSqsouVm350RU0w= =yts5 -----END PGP SIGNATURE-----
I am not exactly sure what you are doing, Rob.
It looks like your log file is getting what should be going to STDOUT when the script runs. That won't help.
Can you post the content of cputemp.sh?
BTW, are you planning to ever run more than one custom test? If so, naming the "catcher" script that way will cause some interesting maintenance issues down the road. Or are you doing this simply to get it working...
GLH
-----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Friday, October 20, 2006 1:22 PM To: hobbit at hswn.dk Subject: Re: [hobbit] RRD data not being created.
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Hubbard, Greg L wrote:
Is this when you are running the script by hand? $2 is the second parameter. So to run the script by hand, you have to:
$ ./script node test some.test.file.name
And then you *should* see "test.rrd" written out.
Weird. running the test by hand gets me this hobbit at randomaccess ~/server/ext $ ./cputemp.sh currant.tbb cputemp currant.tbb.cputemp.out DS:cputemp:GAUGE:600:0:U cputemp.rrd 100
but cputemp.log in /var/log/hobbit shows only
DS:cputemp:GAUGE:600:0:U cputemp.rrd
DS:cputemp:GAUGE:600:0:U cputemp.rrd
DS:cputemp:GAUGE:600:0:U cputemp.rrd
No temp! rrdtool dump on cputemp.rrd shows me a buncha <!-- 2006-10-20 00:00:00 EDT / 1161316800 --> <row><v> NaN </v></row> <!-- 2006-10-20 02:00:00 EDT / 1161324000 --> <row><v> NaN </v></row> <!-- 2006-10-20 04:00:00 EDT / 1161331200 --> <row><v> NaN </v></row> <!-- 2006-10-20 06:00:00 EDT / 1161338400 --> <row><v> NaN </v></row> <!-- 2006-10-20 08:00:00 EDT / 1161345600 --> <row><v> NaN </v></row> <!-- 2006-10-20 10:00:00 EDT / 1161352800 --> <row><v> NaN </v></row> <!-- 2006-10-20 12:00:00 EDT / 1161360000 --> <row><v> NaN </v></row> <!-- 2006-10-20 14:00:00 EDT / 1161367200 --> <row><v> NaN </v></row>
The way I tested this was to capture the output of a client side script (just cut from the browser and paste in a text file), then run it by hand using the above method. The RRD commands are written to the terminal, and any griping is in the log.
Glad you are making progress.
GLH
-----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Thursday, October 19, 2006 2:48 PM To: hobbit at hswn.dk Subject: Re: [hobbit] RRD data not being created.
Hubbard, Greg L wrote:
Rob, if you implemented the script example I posted, you should be able to look in the script log to see if it is doing anything. Just remember not to refer to NCV anywhere in the configuration.
The log will tell you if the script is being called at all -- you can
enable any debug you want -- just make sure you write it to the log, since you will never see STDOUT. You can even cat the incoming file out to the log.
This is what I had to do to get it working -- sort of organically.
I have .rrd files! huzzah. Now i just need to get the actual graph to
appear, instead of a text link.
Also, something Odd is happening such that variables are not being expanded. Frex,
TESTNAME="$2" echo $TESTNAME.rrd
is coming out as just
.rrd
o_O still, it's progress :D
GLH
-----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Thursday, October 19, 2006 1:47 PM To: hobbit at hswn.dk Subject: [hobbit] RRD data not being created.
I'm pretty sure this is the trouble. Ladies and gentlemen, i am stumped. Following all the instructions in the docs, manpages, faqs and alien communications, as well as the very helpful samples provided
by list members... no new file ever gets created in ~/data/rrd/<hostname>. I have status, so i know client info is reaching server. I have copied almost verbatim the sample server-side script needed to
parse this info and spit it back out to RRD. And i have eschewed NCV and gone in favor of hobbitd_channel --options
to make this all work right, or so I thought.
If you make a custom script and the rrd file never happens, where would you look first? The serverside script itself? hobbitlaunch. or
-graph.cfg options? Somewhere else? Mars? I'd be willing to check Mars at this point.
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
To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk
Rob Munsch Solutions For Progress IT www.solutionsforprogress.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFORPeBvBcJFK6xYURAgrAAJ4g3WBP+uTdrzeoCAwEigHVYboZ0QCeMkTw xAoHPHSGiSqsouVm350RU0w= =yts5 -----END PGP SIGNATURE-----
To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Hubbard, Greg L wrote:
I am not exactly sure what you are doing, Rob.
That makes two of us!
It looks like your log file is getting what should be going to STDOUT when the script runs. That won't help.
Can you post the content of cputemp.sh? The serverside or client? heck, here's both.
On the server in server/ext:
hobbit at randomaccess ~/server/ext $ cat cputemp.sh #!/bin/sh
Input parameters: Hostname, testname (column), and messagefile
echo date >/tmp/cputemp.runs
TMPLOG="/tmp/$1.$2.out" echo "$1, $2, $3" > $TMPLOG cat $3 >> $TMPLOG
HOSTNAME="$1" TESTNAME="$2" FNAME="$3"
Check the test name so that this script can service multiple
data collection needs
if [ "$TESTNAME" = "cputemp" ]
then
set up the Data Set, testname, and what kind:size:min:max
U means Unknown max.
echo "DS:cputemp:GAUGE:600:0:U"
name the RRD for the host and test
echo cputemp.rrd
Analyze the message we got. If it's only one line, grab it.
The interesting number is the fourth element in this case.
CPU=grep CPU $FNAME | awk '{ print $4 }'
echo $CPU
exit 0
On the client in client/ext:
hobbit at currant:~/client/ext$ cat cputemp.sh #!/bin/sh
COLUMN=cputemp
COLOR=green
MSG="Current CPU Temperature: "
host=hostname
cpu_temp=$( ipmitool sensor | awk -F\| 'NR==2{print $2}' | sed '/\..*$/ s///g' | sed '/^ */ s///g' )
Converting to Fahrenheit.
temp=$((${cpu_temp}*9/5+32))
Ok, we have the temp in degrees as an integer.
Now to see if it's Very Bad.
if ((${temp} > 130)) then COLOR=red MSG="${MSG} ${temp} degrees Fahrenheit!
Hothothot!
"elif ((${temp} > 120)) then COLOR=yellow MSG="${MSG} ${temp} degrees Fahrenheit.
Getting Warmer...
"else MSG="${MSG} ${temp} degrees Fahrenheit.
Chillin' like a Villain.
"fi
Get this to the hobbit server
$BB $BBDISP "status $MACHINE.$COLUMN $COLOR date
${MSG} " exit 0
Or are you doing this simply to get it working...
That was my Cunning Plan, yah.
- -rm
GLH
-----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Friday, October 20, 2006 1:22 PM To: hobbit at hswn.dk Subject: Re: [hobbit] RRD data not being created.
Hubbard, Greg L wrote:
Is this when you are running the script by hand? $2 is the second parameter. So to run the script by hand, you have to:
$ ./script node test some.test.file.name
And then you *should* see "test.rrd" written out.
Weird. running the test by hand gets me this hobbit at randomaccess ~/server/ext $ ./cputemp.sh currant.tbb cputemp currant.tbb.cputemp.out DS:cputemp:GAUGE:600:0:U cputemp.rrd 100
but cputemp.log in /var/log/hobbit shows only
DS:cputemp:GAUGE:600:0:U cputemp.rrd
DS:cputemp:GAUGE:600:0:U cputemp.rrd
DS:cputemp:GAUGE:600:0:U cputemp.rrd
No temp! rrdtool dump on cputemp.rrd shows me a buncha <!-- 2006-10-20 00:00:00 EDT / 1161316800 --> <row><v> NaN </v></row> <!-- 2006-10-20 02:00:00 EDT / 1161324000 --> <row><v> NaN </v></row> <!-- 2006-10-20 04:00:00 EDT / 1161331200 --> <row><v> NaN </v></row> <!-- 2006-10-20 06:00:00 EDT / 1161338400 --> <row><v> NaN </v></row> <!-- 2006-10-20 08:00:00 EDT / 1161345600 --> <row><v> NaN </v></row> <!-- 2006-10-20 10:00:00 EDT / 1161352800 --> <row><v> NaN </v></row> <!-- 2006-10-20 12:00:00 EDT / 1161360000 --> <row><v> NaN </v></row> <!-- 2006-10-20 14:00:00 EDT / 1161367200 --> <row><v> NaN </v></row>
The way I tested this was to capture the output of a client side script (just cut from the browser and paste in a text file), then run it by hand using the above method. The RRD commands are written to the terminal, and any griping is in the log.
Glad you are making progress.
GLH
-----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Thursday, October 19, 2006 2:48 PM To: hobbit at hswn.dk Subject: Re: [hobbit] RRD data not being created.
Hubbard, Greg L wrote:
Rob, if you implemented the script example I posted, you should be able to look in the script log to see if it is doing anything. Just remember not to refer to NCV anywhere in the configuration. The log will tell you if the script is being called at all -- you can
enable any debug you want -- just make sure you write it to the log, since you will never see STDOUT. You can even cat the incoming file out to the log. This is what I had to do to get it working -- sort of organically. I have .rrd files! huzzah. Now i just need to get the actual graph to
appear, instead of a text link.
Also, something Odd is happening such that variables are not being expanded. Frex,
TESTNAME="$2" echo $TESTNAME.rrd
is coming out as just
.rrd
o_O still, it's progress :D
GLH -----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Thursday, October 19, 2006 1:47 PM To: hobbit at hswn.dk Subject: [hobbit] RRD data not being created. I'm pretty sure this is the trouble. Ladies and gentlemen, i am stumped. Following all the instructions in the docs, manpages, faqs and alien communications, as well as the very helpful samples provided by list members... no new file ever gets created in ~/data/rrd/<hostname>. I have status, so i know client info is reaching server. I have copied almost verbatim the sample server-side script needed to
parse this info and spit it back out to RRD. And i have eschewed NCV and gone in favor of hobbitd_channel --options to make this all work right, or so I thought. If you make a custom script and the rrd file never happens, where would you look first? The serverside script itself? hobbitlaunch. or -graph.cfg options? Somewhere else? Mars? I'd be willing to check Mars at this point.
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
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
Rob Munsch Solutions For Progress IT www.solutionsforprogress.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFOSQpBvBcJFK6xYURAvzNAJ9kvb8BKe3DARY8vtTGmJ+aqOBcNgCfQFGf +dGBWtOCbAkyD9gwkOb5R78= =4qGs -----END PGP SIGNATURE-----
I don't think your grep is working in cputemp.sh, so it writes out a blank since you are just echoing the value you get instead of testing it first. You might add some debug, like:
Echo "I found /${CPU}/ value for cputemp" >> $TMPLOG
And see what ya get...
GLH
-----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Friday, October 20, 2006 2:32 PM To: hobbit at hswn.dk Subject: Re: [hobbit] RRD data not being created.
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Hubbard, Greg L wrote:
I am not exactly sure what you are doing, Rob.
That makes two of us!
It looks like your log file is getting what should be going to STDOUT when the script runs. That won't help.
Can you post the content of cputemp.sh? The serverside or client? heck, here's both.
On the server in server/ext:
hobbit at randomaccess ~/server/ext $ cat cputemp.sh #!/bin/sh # Input parameters: Hostname, testname (column), and messagefile #
echo date >/tmp/cputemp.runs
TMPLOG="/tmp/$1.$2.out" echo "$1, $2, $3" > $TMPLOG cat $3 >> $TMPLOG
HOSTNAME="$1" TESTNAME="$2" FNAME="$3"
Check the test name so that this script can service multiple # data
collection needs # if [ "$TESTNAME" = "cputemp" ] # then
set up the Data Set, testname, and what kind:size:min:max
U means Unknown max.
echo "DS:cputemp:GAUGE:600:0:U"
name the RRD for the host and test
echo cputemp.rrd
Analyze the message we got. If it's only one line, grab it.
The interesting number is the fourth element in this case.
CPU=grep CPU $FNAME | awk '{ print $4 }'
echo $CPU
exit 0
On the client in client/ext:
hobbit at currant:~/client/ext$ cat cputemp.sh #!/bin/sh
COLUMN=cputemp
COLOR=green
MSG="Current CPU Temperature: "
host=hostname
cpu_temp=$( ipmitool sensor | awk -F\| 'NR==2{print $2}' | sed '/\..*$/ s///g' | sed '/^ */ s///g' )
Converting to Fahrenheit.
temp=$((${cpu_temp}*9/5+32))
Ok, we have the temp in degrees as an integer.
Now to see if it's Very Bad.
if ((${temp} > 130)) then COLOR=red MSG="${MSG} ${temp} degrees Fahrenheit!
Hothothot!
"elif ((${temp} > 120)) then COLOR=yellow MSG="${MSG} ${temp} degrees Fahrenheit.
Getting Warmer...
"else MSG="${MSG} ${temp} degrees Fahrenheit.
Chillin' like a Villain.
"fi
Get this to the hobbit server
$BB $BBDISP "status $MACHINE.$COLUMN $COLOR date
${MSG} " exit 0
Or are you doing this simply to get it working...
That was my Cunning Plan, yah.
- -rm
GLH
-----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Friday, October 20, 2006 1:22 PM To: hobbit at hswn.dk Subject: Re: [hobbit] RRD data not being created.
Hubbard, Greg L wrote:
Is this when you are running the script by hand? $2 is the second parameter. So to run the script by hand, you have to:
$ ./script node test some.test.file.name
And then you *should* see "test.rrd" written out.
Weird. running the test by hand gets me this hobbit at randomaccess ~/server/ext $ ./cputemp.sh currant.tbb cputemp currant.tbb.cputemp.out DS:cputemp:GAUGE:600:0:U cputemp.rrd 100
but cputemp.log in /var/log/hobbit shows only
DS:cputemp:GAUGE:600:0:U cputemp.rrd
DS:cputemp:GAUGE:600:0:U cputemp.rrd
DS:cputemp:GAUGE:600:0:U cputemp.rrd
No temp! rrdtool dump on cputemp.rrd shows me a buncha <!-- 2006-10-20 00:00:00 EDT / 1161316800 --> <row><v> NaN </v></row> <!-- 2006-10-20 02:00:00 EDT / 1161324000 --> <row><v> NaN </v></row> <!-- 2006-10-20 04:00:00 EDT / 1161331200 --> <row><v> NaN </v></row> <!-- 2006-10-20 06:00:00 EDT / 1161338400 --> <row><v> NaN </v></row> <!-- 2006-10-20 08:00:00 EDT / 1161345600 --> <row><v> NaN </v></row> <!-- 2006-10-20 10:00:00 EDT / 1161352800 --> <row><v> NaN </v></row> <!-- 2006-10-20 12:00:00 EDT / 1161360000 --> <row><v> NaN </v></row> <!-- 2006-10-20 14:00:00 EDT / 1161367200 --> <row><v> NaN </v></row>
The way I tested this was to capture the output of a client side script (just cut from the browser and paste in a text file), then run
it by hand using the above method. The RRD commands are written to the terminal, and any griping is in the log.
Glad you are making progress.
GLH
-----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Thursday, October 19, 2006 2:48 PM To: hobbit at hswn.dk Subject: Re: [hobbit] RRD data not being created.
Hubbard, Greg L wrote:
Rob, if you implemented the script example I posted, you should be able to look in the script log to see if it is doing anything. Just
remember not to refer to NCV anywhere in the configuration. The log will tell you if the script is being called at all -- you can
enable any debug you want -- just make sure you write it to the log,
since you will never see STDOUT. You can even cat the incoming file
out to the log. This is what I had to do to get it working -- sort of organically. I have .rrd files! huzzah. Now i just need to get the actual graph to
appear, instead of a text link.
Also, something Odd is happening such that variables are not being expanded. Frex,
TESTNAME="$2" echo $TESTNAME.rrd
is coming out as just
.rrd
o_O still, it's progress :D
GLH -----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Thursday, October 19, 2006 1:47 PM To: hobbit at hswn.dk Subject: [hobbit] RRD data not being created. I'm pretty sure this is the trouble. Ladies and gentlemen, i am stumped. Following all the instructions in the docs, manpages, faqs
and alien communications, as well as the very helpful samples provided by list members... no new file ever gets created in ~/data/rrd/<hostname>. I have status, so i know client info is reaching server. I have copied almost verbatim the sample server-side script needed to
parse this info and spit it back out to RRD. And i have eschewed NCV and gone in favor of hobbitd_channel --options to make this all work right, or so I thought. If you make a custom script and the rrd file never happens, where would you look first? The serverside script itself? hobbitlaunch. or -graph.cfg options? Somewhere else? Mars? I'd be willing to check
Mars at this point.
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
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
Rob Munsch Solutions For Progress IT www.solutionsforprogress.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFOSQpBvBcJFK6xYURAvzNAJ9kvb8BKe3DARY8vtTGmJ+aqOBcNgCfQFGf +dGBWtOCbAkyD9gwkOb5R78= =4qGs -----END PGP SIGNATURE-----
To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Hubbard, Greg L wrote:
I don't think your grep is working in cputemp.sh, so it writes out a blank since you are just echoing the value you get instead of testing it first. You might add some debug, like:
Echo "I found /${CPU}/ value for cputemp" >> $TMPLOG
And see what ya get...
I found /102/ value for cputemp
Okay. There was a leading space, but i trimmed it off - that line now reads
CPU=grep CPU $FNAME | awk '{ print $4 }' | sed '/^ / s///'
but i am still not getting any kind of graph. "hobbit graph cputemp" remains a simple text link. Thanks for your efforts.
Fri Oct 20 16:46:43 EDT 2006
Current CPU Temperature: 105 degrees Fahrenheit.
Chillin' like a Villain.Status unchanged in 1 days,5 hours, 54 minutes Status message received from 10.10.10.12 Client data available hobbit graph cputemp Zoom graph
GLH
-----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Friday, October 20, 2006 2:32 PM To: hobbit at hswn.dk Subject: Re: [hobbit] RRD data not being created.
Hubbard, Greg L wrote:
I am not exactly sure what you are doing, Rob.
That makes two of us!
It looks like your log file is getting what should be going to STDOUT when the script runs. That won't help.
Can you post the content of cputemp.sh? The serverside or client? heck, here's both.
On the server in server/ext:
hobbit at randomaccess ~/server/ext $ cat cputemp.sh #!/bin/sh # Input parameters: Hostname, testname (column), and messagefile #
echo
date>/tmp/cputemp.runsTMPLOG="/tmp/$1.$2.out" echo "$1, $2, $3" > $TMPLOG cat $3 >> $TMPLOG
HOSTNAME="$1" TESTNAME="$2" FNAME="$3"
Check the test name so that this script can service multiple # data
collection needs # if [ "$TESTNAME" = "cputemp" ] # then
set up the Data Set, testname, and what kind:size:min:max
U means Unknown max.
echo "DS:cputemp:GAUGE:600:0:U"
name the RRD for the host and test
echo cputemp.rrd
Analyze the message we got. If it's only one line, grab it.
The interesting number is the fourth element in this case.
CPU=
grep CPU $FNAME | awk '{ print $4 }'echo $CPUexit 0
On the client in client/ext:
hobbit at currant:~/client/ext$ cat cputemp.sh #!/bin/sh
COLUMN=cputemp COLOR=green MSG="Current CPU Temperature: " host=
hostnamecpu_temp=$( ipmitool sensor | awk -F\| 'NR==2{print $2}' | sed '/\..*$/ s///g' | sed '/^ */ s///g' )
Converting to Fahrenheit.
temp=$((${cpu_temp}*9/5+32))
Ok, we have the temp in degrees as an integer.
Now to see if it's Very Bad.
if ((${temp} > 130)) then COLOR=red MSG="${MSG} ${temp} degrees Fahrenheit!
Hothothot! "elif ((${temp} > 120)) then COLOR=yellow MSG="${MSG} ${temp} degrees Fahrenheit.
Getting Warmer... "else MSG="${MSG} ${temp} degrees Fahrenheit.
Chillin' like a Villain. "fi
Get this to the hobbit server
$BB $BBDISP "status $MACHINE.$COLUMN $COLOR
date${MSG} " exit 0
Or are you doing this simply to get it working...
That was my Cunning Plan, yah.
-rm
GLH
-----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Friday, October 20, 2006 1:22 PM To: hobbit at hswn.dk Subject: Re: [hobbit] RRD data not being created.
Hubbard, Greg L wrote:
Is this when you are running the script by hand? $2 is the second parameter. So to run the script by hand, you have to: $ ./script node test some.test.file.name And then you *should* see "test.rrd" written out. Weird. running the test by hand gets me this hobbit at randomaccess ~/server/ext $ ./cputemp.sh currant.tbb cputemp currant.tbb.cputemp.out DS:cputemp:GAUGE:600:0:U cputemp.rrd 100
but cputemp.log in /var/log/hobbit shows only
DS:cputemp:GAUGE:600:0:U cputemp.rrd
DS:cputemp:GAUGE:600:0:U cputemp.rrd
DS:cputemp:GAUGE:600:0:U cputemp.rrd
No temp! rrdtool dump on cputemp.rrd shows me a buncha <!-- 2006-10-20 00:00:00 EDT / 1161316800 --> <row><v> NaN </v></row> <!-- 2006-10-20 02:00:00 EDT / 1161324000 --> <row><v> NaN </v></row> <!-- 2006-10-20 04:00:00 EDT / 1161331200 --> <row><v> NaN </v></row> <!-- 2006-10-20 06:00:00 EDT / 1161338400 --> <row><v> NaN </v></row> <!-- 2006-10-20 08:00:00 EDT / 1161345600 --> <row><v> NaN </v></row> <!-- 2006-10-20 10:00:00 EDT / 1161352800 --> <row><v> NaN </v></row> <!-- 2006-10-20 12:00:00 EDT / 1161360000 --> <row><v> NaN </v></row> <!-- 2006-10-20 14:00:00 EDT / 1161367200 --> <row><v> NaN </v></row>
The way I tested this was to capture the output of a client side script (just cut from the browser and paste in a text file), then run
it by hand using the above method. The RRD commands are written to the terminal, and any griping is in the log. Glad you are making progress. GLH -----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Thursday, October 19, 2006 2:48 PM To: hobbit at hswn.dk Subject: Re: [hobbit] RRD data not being created. Hubbard, Greg L wrote:
Rob, if you implemented the script example I posted, you should be able to look in the script log to see if it is doing anything. Just
remember not to refer to NCV anywhere in the configuration. The log will tell you if the script is being called at all -- you can enable any debug you want -- just make sure you write it to the log,
since you will never see STDOUT. You can even cat the incoming file
out to the log. This is what I had to do to get it working -- sort of organically. I have .rrd files! huzzah. Now i just need to get the actual graph to appear, instead of a text link. Also, something Odd is happening such that variables are not being expanded. Frex, TESTNAME="$2" echo $TESTNAME.rrd is coming out as just .rrd o_O still, it's progress :D GLH -----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Thursday, October 19, 2006 1:47 PM To: hobbit at hswn.dk Subject: [hobbit] RRD data not being created. I'm pretty sure this is the trouble. Ladies and gentlemen, i am stumped. Following all the instructions in the docs, manpages, faqs
and alien communications, as well as the very helpful samples provided by list members... no new file ever gets created in ~/data/rrd/<hostname>. I have status, so i know client info is reaching server. I have copied almost verbatim the sample server-side script needed to parse this info and spit it back out to RRD. And i have eschewed NCV and gone in favor of hobbitd_channel --options to make this all work right, or so I thought. If you make a custom script and the rrd file never happens, where would you look first? The serverside script itself? hobbitlaunch. or -graph.cfg options? Somewhere else? Mars? I'd be willing to check
Mars at this point. 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
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
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
Rob Munsch Solutions For Progress IT www.solutionsforprogress.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFOTYLBvBcJFK6xYURAh/vAJ946E/6VtDuteetUWbDuYV/1jjzTwCdFnVN S24iLKkZ/uBHzKPZBtoBKjw= =CZKd -----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
just to confirm: i found an old thread message from Henrik that says
- --extra-scripts
whereas the FAQ says
- --extra-script and --extra-tests
which makes sense, since you can have multiple extra-tesTS but only one extra scripT; but given my situation, i thought i'd better check...
Rob Munsch wrote:
Hubbard, Greg L wrote:
I don't think your grep is working in cputemp.sh, so it writes out a blank since you are just echoing the value you get instead of testing it first. You might add some debug, like:
Echo "I found /${CPU}/ value for cputemp" >> $TMPLOG
And see what ya get...
I found /102/ value for cputemp
Okay. There was a leading space, but i trimmed it off - that line now reads
CPU=
grep CPU $FNAME | awk '{ print $4 }' | sed '/^ / s///'but i am still not getting any kind of graph. "hobbit graph cputemp" remains a simple text link. Thanks for your efforts.
Fri Oct 20 16:46:43 EDT 2006
Current CPU Temperature: 105 degrees Fahrenheit.
Chillin' like a Villain.
Status unchanged in 1 days,5 hours, 54 minutes Status message received from 10.10.10.12 Client data available hobbit graph cputemp Zoom graph
GLH
-----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Friday, October 20, 2006 2:32 PM To: hobbit at hswn.dk Subject: Re: [hobbit] RRD data not being created.
Hubbard, Greg L wrote:
I am not exactly sure what you are doing, Rob. That makes two of us!
It looks like your log file is getting what should be going to STDOUT when the script runs. That won't help. Can you post the content of cputemp.sh? The serverside or client? heck, here's both.
On the server in server/ext:
hobbit at randomaccess ~/server/ext $ cat cputemp.sh #!/bin/sh # Input parameters: Hostname, testname (column), and messagefile #
echo
date>/tmp/cputemp.runsTMPLOG="/tmp/$1.$2.out" echo "$1, $2, $3" > $TMPLOG cat $3 >> $TMPLOG
HOSTNAME="$1" TESTNAME="$2" FNAME="$3"
Check the test name so that this script can service multiple # data
collection needs # if [ "$TESTNAME" = "cputemp" ] # then
set up the Data Set, testname, and what kind:size:min:max
U means Unknown max.
echo "DS:cputemp:GAUGE:600:0:U"
name the RRD for the host and test
echo cputemp.rrd
Analyze the message we got. If it's only one line, grab it.
The interesting number is the fourth element in this case.
CPU=
grep CPU $FNAME | awk '{ print $4 }'echo $CPUexit 0
On the client in client/ext:
hobbit at currant:~/client/ext$ cat cputemp.sh #!/bin/sh
COLUMN=cputemp COLOR=green MSG="Current CPU Temperature: " host=
hostnamecpu_temp=$( ipmitool sensor | awk -F\| 'NR==2{print $2}' | sed '/\..*$/ s///g' | sed '/^ */ s///g' )
Converting to Fahrenheit.
temp=$((${cpu_temp}*9/5+32))
Ok, we have the temp in degrees as an integer.
Now to see if it's Very Bad.
if ((${temp} > 130)) then COLOR=red MSG="${MSG} ${temp} degrees Fahrenheit!
Hothothot! "elif ((${temp} > 120)) then COLOR=yellow MSG="${MSG} ${temp} degrees Fahrenheit.
Getting Warmer... "else MSG="${MSG} ${temp} degrees Fahrenheit.
Chillin' like a Villain. "fi
Get this to the hobbit server
$BB $BBDISP "status $MACHINE.$COLUMN $COLOR
date${MSG} " exit 0
Or are you doing this simply to get it working... That was my Cunning Plan, yah.
-rm
GLH
-----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Friday, October 20, 2006 1:22 PM To: hobbit at hswn.dk Subject: Re: [hobbit] RRD data not being created. Hubbard, Greg L wrote:
Is this when you are running the script by hand? $2 is the second parameter. So to run the script by hand, you have to: $ ./script node test some.test.file.name And then you *should* see "test.rrd" written out. Weird. running the test by hand gets me this hobbit at randomaccess ~/server/ext $ ./cputemp.sh currant.tbb cputemp currant.tbb.cputemp.out DS:cputemp:GAUGE:600:0:U cputemp.rrd 100 but cputemp.log in /var/log/hobbit shows only DS:cputemp:GAUGE:600:0:U cputemp.rrd DS:cputemp:GAUGE:600:0:U cputemp.rrd DS:cputemp:GAUGE:600:0:U cputemp.rrd
No temp! rrdtool dump on cputemp.rrd shows me a buncha <!-- 2006-10-20 00:00:00 EDT / 1161316800 --> <row><v> NaN </v></row> <!-- 2006-10-20 02:00:00 EDT / 1161324000 --> <row><v> NaN </v></row> <!-- 2006-10-20 04:00:00 EDT / 1161331200 --> <row><v> NaN </v></row> <!-- 2006-10-20 06:00:00 EDT / 1161338400 --> <row><v> NaN </v></row> <!-- 2006-10-20 08:00:00 EDT / 1161345600 --> <row><v> NaN </v></row> <!-- 2006-10-20 10:00:00 EDT / 1161352800 --> <row><v> NaN </v></row> <!-- 2006-10-20 12:00:00 EDT / 1161360000 --> <row><v> NaN </v></row> <!-- 2006-10-20 14:00:00 EDT / 1161367200 --> <row><v> NaN </v></row>
The way I tested this was to capture the output of a client side script (just cut from the browser and paste in a text file), then run it by hand using the above method. The RRD commands are written to the terminal, and any griping is in the log. Glad you are making progress. GLH -----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Thursday, October 19, 2006 2:48 PM To: hobbit at hswn.dk Subject: Re: [hobbit] RRD data not being created. Hubbard, Greg L wrote:
Rob, if you implemented the script example I posted, you should be able to look in the script log to see if it is doing anything. Just remember not to refer to NCV anywhere in the configuration. The log will tell you if the script is being called at all -- you can enable any debug you want -- just make sure you write it to the log, since you will never see STDOUT. You can even cat the incoming file out to the log. This is what I had to do to get it working -- sort of organically. I have .rrd files! huzzah. Now i just need to get the actual graph to appear, instead of a text link. Also, something Odd is happening such that variables are not being expanded. Frex, TESTNAME="$2" echo $TESTNAME.rrd is coming out as just .rrd o_O still, it's progress :D GLH -----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Thursday, October 19, 2006 1:47 PM To: hobbit at hswn.dk Subject: [hobbit] RRD data not being created. I'm pretty sure this is the trouble. Ladies and gentlemen, i am stumped. Following all the instructions in the docs, manpages, faqs and alien communications, as well as the very helpful samples provided by list members... no new file ever gets created in ~/data/rrd/<hostname>. I have status, so i know client info is reaching server. I have copied almost verbatim the sample server-side script needed to parse this info and spit it back out to RRD. And i have eschewed NCV and gone in favor of hobbitd_channel --options to make this all work right, or so I thought. If you make a custom script and the rrd file never happens, where would you look first? The serverside script itself? hobbitlaunch. or -graph.cfg options? Somewhere else? Mars? I'd be willing to check Mars at this point. 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
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
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
Rob Munsch Solutions For Progress IT www.solutionsforprogress.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFOTz0BvBcJFK6xYURAun5AJ9IKG809D+rOXwHoFzWPoXe4FwX4ACfZuuZ yI51hfSXhApJ9i/0JnQSIBg= =c3DA -----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Have a nice weekend, everyone. Every test passes, but nothing actually HAPPENS.
<!-- 2006-10-20 16:25:00 EDT / 1161375900 --> <row><v> NaN </v></row> <!-- 2006-10-20 16:30:00 EDT / 1161376200 --> <row><v> NaN </v></row> <!-- 2006-10-20 16:35:00 EDT / 1161376500 --> <row><v> NaN </v></row> <!-- 2006-10-20 16:40:00 EDT / 1161376800 --> <row><v> NaN </v></row> <!-- 2006-10-20 16:45:00 EDT / 1161377100 --> <row><v> NaN </v></row> <!-- 2006-10-20 16:50:00 EDT / 1161377400 --> <row><v> NaN </v></row> <!-- 2006-10-20 16:55:00 EDT / 1161377700 --> <row><v> NaN </v></row> <!-- 2006-10-20 17:00:00 EDT / 1161378000 --> <row><v> NaN </v></row> <!-- 2006-10-20 17:05:00 EDT / 1161378300 --> <row><v> NaN </v></row> <!-- 2006-10-20 17:10:00 EDT / 1161378600 --> <row><v> NaN </v></row> <!-- 2006-10-20 17:15:00 EDT / 1161378900 --> <row><v> NaN </v></row> <!-- 2006-10-20 17:20:00 EDT / 1161379200 --> <row><v> NaN </v></row>
Every indication is that, ultimately, a valid DS: gets exported to a proper .rrd file and so on and so forth and blah blah blah. It just never draws the damn graph. I'm going home. Thank you all for your patience and support.
Rob Munsch wrote:
just to confirm: i found an old thread message from Henrik that says --extra-scripts
whereas the FAQ says --extra-script and --extra-tests
which makes sense, since you can have multiple extra-tesTS but only one extra scripT; but given my situation, i thought i'd better check...
Rob Munsch wrote:
Hubbard, Greg L wrote:
I don't think your grep is working in cputemp.sh, so it writes out a blank since you are just echoing the value you get instead of testing it first. You might add some debug, like: Echo "I found /${CPU}/ value for cputemp" >> $TMPLOG And see what ya get... I found /102/ value for cputemp
Okay. There was a leading space, but i trimmed it off - that line now reads
CPU=
grep CPU $FNAME | awk '{ print $4 }' | sed '/^ / s///'but i am still not getting any kind of graph. "hobbit graph cputemp" remains a simple text link. Thanks for your efforts.
Fri Oct 20 16:46:43 EDT 2006
Current CPU Temperature: 105 degrees Fahrenheit.
Chillin' like a Villain.
Status unchanged in 1 days,5 hours, 54 minutes Status message received from 10.10.10.12 Client data available hobbit graph cputemp Zoom graph
GLH
-----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Friday, October 20, 2006 2:32 PM To: hobbit at hswn.dk Subject: Re: [hobbit] RRD data not being created. Hubbard, Greg L wrote:
I am not exactly sure what you are doing, Rob. That makes two of us! It looks like your log file is getting what should be going to STDOUT when the script runs. That won't help. Can you post the content of cputemp.sh? The serverside or client? heck, here's both. On the server in server/ext:
hobbit at randomaccess ~/server/ext $ cat cputemp.sh #!/bin/sh # Input parameters: Hostname, testname (column), and messagefile # echo
date>/tmp/cputemp.runs TMPLOG="/tmp/$1.$2.out" echo "$1, $2, $3" > $TMPLOG cat $3 >> $TMPLOG HOSTNAME="$1" TESTNAME="$2" FNAME="$3"Check the test name so that this script can service multiple # data
collection needs # if [ "$TESTNAME" = "cputemp" ] # then
set up the Data Set, testname, and what kind:size:min:max
U means Unknown max.
echo "DS:cputemp:GAUGE:600:0:U"
name the RRD for the host and test
echo cputemp.rrd
Analyze the message we got. If it's only one line, grab it.
The interesting number is the fourth element in this case.
CPU=
grep CPU $FNAME | awk '{ print $4 }'echo $CPU exit 0On the client in client/ext:
hobbit at currant:~/client/ext$ cat cputemp.sh #!/bin/sh COLUMN=cputemp COLOR=green MSG="Current CPU Temperature: " host=
hostnamecpu_temp=$( ipmitool sensor | awk -F\| 'NR==2{print $2}' | sed '/\..*$/ s///g' | sed '/^ */ s///g' )
Converting to Fahrenheit.
temp=$((${cpu_temp}*9/5+32))
Ok, we have the temp in degrees as an integer.
Now to see if it's Very Bad.
if ((${temp} > 130)) then COLOR=red MSG="${MSG} ${temp} degrees Fahrenheit! Hothothot! " elif ((${temp} > 120)) then COLOR=yellow MSG="${MSG} ${temp} degrees Fahrenheit. Getting Warmer... " else MSG="${MSG} ${temp} degrees Fahrenheit. Chillin' like a Villain. " fi
Get this to the hobbit server
$BB $BBDISP "status $MACHINE.$COLUMN $COLOR
date${MSG} " exit 0Or are you doing this simply to get it working... That was my Cunning Plan, yah. -rm GLH -----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Friday, October 20, 2006 1:22 PM To: hobbit at hswn.dk Subject: Re: [hobbit] RRD data not being created. Hubbard, Greg L wrote:
Is this when you are running the script by hand? $2 is the second parameter. So to run the script by hand, you have to: $ ./script node test some.test.file.name And then you *should* see "test.rrd" written out. Weird. running the test by hand gets me this hobbit at randomaccess ~/server/ext $ ./cputemp.sh currant.tbb cputemp currant.tbb.cputemp.out DS:cputemp:GAUGE:600:0:U cputemp.rrd 100 but cputemp.log in /var/log/hobbit shows only DS:cputemp:GAUGE:600:0:U cputemp.rrd DS:cputemp:GAUGE:600:0:U cputemp.rrd DS:cputemp:GAUGE:600:0:U cputemp.rrd No temp! rrdtool dump on cputemp.rrd shows me a buncha <!-- 2006-10-20 00:00:00 EDT / 1161316800 --> <row><v> NaN </v></row> <!-- 2006-10-20 02:00:00 EDT / 1161324000 --> <row><v> NaN </v></row> <!-- 2006-10-20 04:00:00 EDT / 1161331200 --> <row><v> NaN </v></row> <!-- 2006-10-20 06:00:00 EDT / 1161338400 --> <row><v> NaN </v></row> <!-- 2006-10-20 08:00:00 EDT / 1161345600 --> <row><v> NaN </v></row> <!-- 2006-10-20 10:00:00 EDT / 1161352800 --> <row><v> NaN </v></row> <!-- 2006-10-20 12:00:00 EDT / 1161360000 --> <row><v> NaN </v></row> <!-- 2006-10-20 14:00:00 EDT / 1161367200 --> <row><v> NaN </v></row> The way I tested this was to capture the output of a client side script (just cut from the browser and paste in a text file), then run it by hand using the above method. The RRD commands are written to the terminal, and any griping is in the log. Glad you are making progress. GLH -----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Thursday, October 19, 2006 2:48 PM To: hobbit at hswn.dk Subject: Re: [hobbit] RRD data not being created. Hubbard, Greg L wrote:
Rob, if you implemented the script example I posted, you should be able to look in the script log to see if it is doing anything. Just remember not to refer to NCV anywhere in the configuration. The log will tell you if the script is being called at all -- you can enable any debug you want -- just make sure you write it to the log, since you will never see STDOUT. You can even cat the incoming file out to the log. This is what I had to do to get it working -- sort of organically. I have .rrd files! huzzah. Now i just need to get the actual graph to appear, instead of a text link. Also, something Odd is happening such that variables are not being expanded. Frex, TESTNAME="$2" echo $TESTNAME.rrd is coming out as just .rrd o_O still, it's progress :D GLH -----Original Message----- From: Rob Munsch [mailto:rmunsch at solutionsforprogress.com] Sent: Thursday, October 19, 2006 1:47 PM To: hobbit at hswn.dk Subject: [hobbit] RRD data not being created. I'm pretty sure this is the trouble. Ladies and gentlemen, i am stumped. Following all the instructions in the docs, manpages, faqs and alien communications, as well as the very helpful samples provided by list members... no new file ever gets created in ~/data/rrd/<hostname>. I have status, so i know client info is reaching server. I have copied almost verbatim the sample server-side script needed to parse this info and spit it back out to RRD. And i have eschewed NCV and gone in favor of hobbitd_channel --options to make this all work right, or so I thought. If you make a custom script and the rrd file never happens, where would you look first? The serverside script itself? hobbitlaunch. or -graph.cfg options? Somewhere else? Mars? I'd be willing to check Mars at this point. 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
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
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
To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk
Rob Munsch Solutions For Progress IT www.solutionsforprogress.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFOT69BvBcJFK6xYURArhnAKCJ+ywB4upqaGVhdho2p/91trVdvQCdGgMg 0c8kHM+9Jc+c1lBlHb3r2mY= =22Ux -----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
A weekend off seemed to help. I see actual numbers in the .rrd now. Maybe this will help. The text link where my graph should be is this:
http://hobbit/hobbit-cgi/hobbitgraph.sh?host=currant.tbb&service=cputemp&gra...
anyone see anything obviously wrong there?
I am currently mucking about with hobbitlaunch. and hobbitserver.cfg as i think the basic setup of a new graph is what is lacking. Hopefully, when this is done, i will be able to contribute the missing bits to documentation. Assuming, of course, i am not at that time living in an asylum ,-)
Rob Munsch Solutions For Progress IT www.solutionsforprogress.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFPQ4wBvBcJFK6xYURAplWAJ9i9cxddls8DEmZ7BLpNvKmIBN9LQCbB7oH oovgKKHaZmgs2e9nZuWXkbI= =tZP7 -----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
A last-ditch plea for help. For those not already skipping any thread starting RRD, if anyone has the time i will buy you a 6-pack of the beer of your choice. Here's everything i got.
The issue: cputemp.rrd is created. It shows values in the XML. The custom test shows green, the page shows the MSG i crafted. But, There is no graph. Not even a blank one with "nan," there's nothing but a text link!
The Script: server-side only; since we are capturing the desired number, no need to bother with the client now.
#!/bin/sh
Input parameters: Hostname, testname (column), and messagefile
echo date >/tmp/cputemp.runs
TMPLOG="/tmp/$1.$2.out" echo "$1, $2, $3" > $TMPLOG cat $3 >> $TMPLOG
HOSTNAME="$1" TESTNAME="$2" FNAME="$3"
set up the Data Set, testname, and what kind:time:min:max
U means Unknown.
echo "DS:cputemp:GAUGE:600:0:U"
name the RRD for the host and test
echo "cputemp.rrd"
Analyze the message we got. If it's only one line, grab it.
The interesting number is the fourth element in this case.
We're getting a leading space: kill it.
CPU=grep CPU $FNAME | awk '{ print $4 }' | sed '/^ / s///'
echo "I found ${CPU} value for cputemp" >> $TMPLOG
echo "${CPU}"
exit 0
TMPLOG shows:
currant.tbb, cputemp, /home/hobbit/server/tmp/rrd_msg_27289 status currant,tbb.cputemp green Mon Oct 23 15:37:50 EDT 2006
Current CPU Temperature: 109 degrees Fahrenheit. I found 109 value for cputemp
So we're getting the number. A dump of the rrd file seems to back this up
<!-- 2006-10-23 14:35:00 EDT / 1161628500 --> <row><v> 1.0037272727e+02 </v></row> <!-- 2006-10-23 14:40:00 EDT / 1161628800 --> <row><v> 1.0093333333e+02 </v></row> <!-- 2006-10-23 14:45:00 EDT / 1161629100 --> <row><v> 1.0085666667e+02 </v></row>
Ok, great. So data is fine. On to the graph commands! The fault must be in something i did, or didn't, in the server configs.
In hobbitgraph.cfg:
[cputemp] TITLE CPU Temperature YAXIS DegreesF DEF:cputemp.rrd:cputemp:AVERAGE LINE2:cputemp#0000FF:CPU Temperature GPRINT:cputemp:LAST: \: %5.1lf (cur) GPRINT:cputemp:MAX: \: %5.1lf (max) GPRINT:cputemp:MIN: \: %5.1lf (min) GPRINT:cputemp:AVERAGE: \: %5.1lf (avg)\n
very boring and plain, yes? nothing fancy.
hobbitlaunch.cfg:
[rrdstatus] ENVFILE /home/hobbit/server/etc/hobbitserver.cfg NEEDS hobbitd CMD hobbitd_channel --channel=status
- --log=$BBSERVERLOGS/rrd-status.log hobbitd_rrd --rrddir=$BBVAR/rrd
- --extra-script=$BBHOME/ext/cputemp.sh --extra-tests=cputemp
"rrddata" updates RRD files with information that arrives as "data"
messages.
If you want RRD graphs of your monitoring BB data, then you want to
run this.
[rrddata] ENVFILE /home/hobbit/server/etc/hobbitserver.cfg NEEDS hobbitd CMD hobbitd_channel --channel=data
- --log=$BBSERVERLOGS/rrd-data.log hobbitd_rrd --rrddir=$BBVAR/rrd
- --extra-script=$BBHOME/ext/cputemp.sh --extra-tests=cputemp
And lastly, hobbitserver.cfg:
TEST2RRD="cputemp,cpu=la,..." GRAPHS="cputemp,la,..."
note that i have tried every combo there i could think of. just test2rrd, just graphs, cputemp=cputemp on one, on the other, on both...
nothing causes a graph to appear. ever. i'm going nuts. I have definitely covered everything documented; many many thanks to Greg and others for pointing out that i was also following the NCV instructions simultaneously for a while there. :)
But... i don't see what could possibly be left that isn't happening. If anyone's gotten non-ncv graphs to appear on custom scripts, the aforementioned 6-pack (bottles not cans) is awaiting you if you could check your server .cfgs and see what you have in there that i don't. The script itself is simple enough: text link instead of picture tells me that RRD isn't even trying to draw anything.
Rob Munsch Solutions For Progress IT www.solutionsforprogress.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFPR4eBvBcJFK6xYURAgjzAJ0fT6bISR3Av5C1oLK23w3rENimqACfZ4NR QNSSU9RE/j+BTTFK+8aodsE= =3f7H -----END PGP SIGNATURE-----
On 10/23/06, Rob Munsch <rmunsch at solutionsforprogress.com> wrote:
In hobbitgraph.cfg:
[cputemp] TITLE CPU Temperature YAXIS DegreesF DEF:cputemp.rrd:cputemp:AVERAGE LINE2:cputemp#0000FF:CPU Temperature GPRINT:cputemp:LAST: \: %5.1lf (cur) GPRINT:cputemp:MAX: \: %5.1lf (max) GPRINT:cputemp:MIN: \: %5.1lf (min) GPRINT:cputemp:AVERAGE: \: %5.1lf (avg)\n
wild guess #1 - shouldn't that be
DEF:cputemp=cputemp.rrd:cputemp:AVERAGEthe manpage (man rrdgraph) says:
DEF:vname=rrdfile:ds-name:CF[:step=step][:start=time][:end=time]vname is the variable to graph rrdfile is the rrd it comes from ds-name is the data-set within the rrd
Ralph
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Ralph Mitchell wrote:
On 10/23/06, Rob Munsch <rmunsch at solutionsforprogress.com> wrote:
In hobbitgraph.cfg:
[cputemp] TITLE CPU Temperature YAXIS DegreesF DEF:cputemp.rrd:cputemp:AVERAGE LINE2:cputemp#0000FF:CPU Temperature GPRINT:cputemp:LAST: \: %5.1lf (cur) GPRINT:cputemp:MAX: \: %5.1lf (max) GPRINT:cputemp:MIN: \: %5.1lf (min) GPRINT:cputemp:AVERAGE: \: %5.1lf (avg)\n
wild guess #1 - shouldn't that be
DEF:cputemp=cputemp.rrd:cputemp:AVERAGE
i thought i'd had that at some point: i'll try now. I'd feel better if i had blank broken graph, i don't know which variable to adjust :D
the manpage (man rrdgraph) says:
DEF:vname=rrdfile:ds-name:CF[:step=step][:start=time][:end=time]vname is the variable to graph rrdfile is the rrd it comes from ds-name is the data-set within the rrd
Think i misunderstood how that worked from the other [sections] in the file i was stea^H^H^H getting ideas from.
Ralph
To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk
Rob Munsch Solutions For Progress IT www.solutionsforprogress.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFPSC4BvBcJFK6xYURAskNAJ4tcwcJdqeG+d3MsL5FbzihjcRDJgCfYEXB ZHmvZxae9hQb3BY+Ek69xic= =7jDw -----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
sonofa...
So what kind of beer you like, Ralph?
Rob Munsch wrote:
Ralph Mitchell wrote:
On 10/23/06, Rob Munsch <rmunsch at solutionsforprogress.com> wrote:
In hobbitgraph.cfg:
[cputemp] TITLE CPU Temperature YAXIS DegreesF DEF:cputemp.rrd:cputemp:AVERAGE LINE2:cputemp#0000FF:CPU Temperature GPRINT:cputemp:LAST: \: %5.1lf (cur) GPRINT:cputemp:MAX: \: %5.1lf (max) GPRINT:cputemp:MIN: \: %5.1lf (min) GPRINT:cputemp:AVERAGE: \: %5.1lf (avg)\n
wild guess #1 - shouldn't that be
DEF:cputemp=cputemp.rrd:cputemp:AVERAGEi thought i'd had that at some point: i'll try now. I'd feel better if i had blank broken graph, i don't know which variable to adjust :D
the manpage (man rrdgraph) says:
DEF:vname=rrdfile:ds-name:CF[:step=step][:start=time][:end=time]vname is the variable to graph rrdfile is the rrd it comes from ds-name is the data-set within the rrd
Think i misunderstood how that worked from the other [sections] in the file i was stea^H^H^H getting ideas from.
Ralph
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
Rob Munsch Solutions For Progress IT www.solutionsforprogress.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFPSEXBvBcJFK6xYURAggGAJ9NE1tg+SqLhO2ttbaiuBUxSoMPiACeMOG/ P/xlX3bDbMZPKLjunQFUHsw= =SOX4 -----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Rob Munsch wrote:
sonofa...
So what kind of beer you like, Ralph?
Of note: instead of
- --extra-tests=<blah>
as it says in the docs, i have the singular, --extra-test=, and that works. Do both work, or was this also a gotcha? (for a while i had cputemp=, but when that didn't seem to work, i started ripping everything apart and starting over).
Rob Munsch wrote:
Ralph Mitchell wrote:
On 10/23/06, Rob Munsch <rmunsch at solutionsforprogress.com> wrote:
In hobbitgraph.cfg:
[cputemp] TITLE CPU Temperature YAXIS DegreesF DEF:cputemp.rrd:cputemp:AVERAGE LINE2:cputemp#0000FF:CPU Temperature GPRINT:cputemp:LAST: \: %5.1lf (cur) GPRINT:cputemp:MAX: \: %5.1lf (max) GPRINT:cputemp:MIN: \: %5.1lf (min) GPRINT:cputemp:AVERAGE: \: %5.1lf (avg)\n
wild guess #1 - shouldn't that be DEF:cputemp=cputemp.rrd:cputemp:AVERAGE i thought i'd had that at some point: i'll try now. I'd feel better if i had blank broken graph, i don't know which variable to adjust :D
the manpage (man rrdgraph) says: DEF:vname=rrdfile:ds-name:CF[:step=step][:start=time][:end=time] vname is the variable to graph rrdfile is the rrd it comes from ds-name is the data-set within the rrd Think i misunderstood how that worked from the other [sections] in the file i was stea^H^H^H getting ideas from.
Ralph 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
Rob Munsch Solutions For Progress IT www.solutionsforprogress.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFPSG5BvBcJFK6xYURAo8wAJ9DKdDfV+CMwtySWqDeBERFzIkfIgCgicJo lZxmxJR/nswfD855dRidA5I= =pG7P -----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
randomaccess - hobbitd_rrd Thu Oct 19 15:43:50 2006
- Program crashed Fatal signal caught!
aaaargh
Rob Munsch Solutions For Progress IT www.solutionsforprogress.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFN9yVBvBcJFK6xYURApZaAJ9NwkzKknGBt6QjxgS/JHeiecxBMgCfSj6j fZryEKHk4j2/lVK0+tgw/fA= =Flde -----END PGP SIGNATURE-----
On Thu, Oct 19, 2006 at 04:14:13PM -0400, Rob Munsch wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
randomaccess - hobbitd_rrd Thu Oct 19 15:43:50 2006
- Program crashed Fatal signal caught!
aaaargh
Most likely, your script returned bogus data to Hobbit. Could you try dumping the data into a file ? I.e. instead of
echo "myfile.rrd" echo "DS:....."
you would do
echo "myfile.rrd" >>/tmp/logfile echo "DS:..." >>/tmp/logfile
Send the data we stored to Hobbit
cat /tmp/logfile
Henrik
Hello,
My hobbit-alert.cfg file has the following line in it:
HOST=* MAIL mprice at hst.nasa.gov
I can run the command like this with the output below:
[hobbit at hobbithost bin]$ ./hobbitd_alert --test v43mgt conn 00003217 2006-10-23 11:59:07 send_alert v43mgt:conn state Paging 00003217 2006-10-23 11:59:07 Matching host:service:page 'v43mgt:conn:' against rule line 129 00003217 2006-10-23 11:59:07 *** Match with 'HOST=*' *** 00003217 2006-10-23 11:59:07 Matching host:service:page 'v43mgt:conn:' against rule line 129 00003217 2006-10-23 11:59:07 *** Match with 'HOST=*' *** 00003217 2006-10-23 11:59:07 Mail alert with command '/var/spool/mail/hobbit "Hobbit [12345] v43mgt:conn CRITICAL (RED)" mprice at hst.nasa.gov'
It seems to work but I never get the test email, I do get regular alarms when a host is down though. What is wrong my testing??? I just configured the hobbit server to alert me via email, so i am trying to test this.
thanx in advance, michael a. price
Michael A. Price
Från: Michael A. Price [mailto:mprice at hst.nasa.gov] Skickat: den 23 oktober 2006 18:02 Till: hobbit at hswn.dk Ämne: [hobbit] hobbitd_alert command is not working ( maybe )
Hello,
My hobbit-alert.cfg file has the following line in it:
HOST=* MAIL mprice at hst.nasa.gov Hi
00003217 2006-10-23 11:59:07 *** Match with 'HOST=*' *** 00003217 2006-10-23 11:59:07 Mail alert with command '/var/spool/mail/hobbit "Hobbit [12345] v43mgt:conn CRITICAL (RED)" mprice at hst.nasa.gov'
The command to send the mail look very odd. What have you as MAILC=?? in hobbitserver.cfg? Should be like
For the hobbitd_alert module
MAILC="mailx" # Command used to send an e-mail with no subject
MAIL="$MAILC -s" # Command used to send an e-mail with a subject
Lars
Lars,
Mine is defined as the following.
MAILC="mail" # Command used to send an e-mail with no subject
MAIL="$MAILC -s" # Command used to send an e-mail with a subject
If that was messed up, wouldnt it prevent my alerts not working at all, because they work now. Its just I cant get them to work when I use the hobbitd_alert command.
thanx, michael
Lars Ebeling wrote:
Från: Michael A. Price [mailto:mprice at hst.nasa.gov] Skickat: den 23 oktober 2006 18:02 Till: hobbit at hswn.dk Ämne: [hobbit] hobbitd_alert command is not working ( maybe )
Hello,
My hobbit-alert.cfg file has the following line in it:
HOST=* MAIL mprice at hst.nasa.gov
Hi
00003217 2006-10-23 11:59:07 *** Match with 'HOST=*' *** 00003217 2006-10-23 11:59:07 Mail alert with command '/var/spool/mail/hobbit "Hobbit [12345] v43mgt:conn CRITICAL (RED)" mprice at hst.nasa.gov'
The command to send the mail look very odd. What have you as MAILC=?? in hobbitserver.cfg? Should be like
For the hobbitd_alert module
MAILC="mailx" # Command used to send an e-mail with no subject
MAIL="$MAILC -s" # Command used to send an e-mail with a subject
Lars
To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk
When I run hobbitd_alert I get this:
$ ./hobbitd_alert --test leopg9 msgs
00016117 2006-10-23 19:00:20 send_alert leopg9:msgs state Paging
00016117 2006-10-23 19:00:20 Matching host:service:page 'leopg9:msgs:' against r ule line 115
00016117 2006-10-23 19:00:20 *** Match with 'HOST=leopg9' ***
00016117 2006-10-23 19:00:20 Matching host:service:page 'leopg9:msgs:' against r ule line 116
00016117 2006-10-23 19:00:20 *** Match with 'MAIL hobbit REPEAT=1h COLOR=red REC OVERED' ***
00016117 2006-10-23 19:00:20 Mail alert with command 'mail -s "Hobbit [12345] le opg9:msgs CRITICAL (RED)" hobbit'
-----Ursprungligt meddelande----- Från: Michael A. Price [mailto:mprice at hst.nasa.gov] Skickat: den 23 oktober 2006 18:53 Till: hobbit at hswn.dk Ämne: Re: SV: [hobbit] hobbitd_alert command is not working ( maybe )
Lars,
Mine is defined as the following.
MAILC="mail" # Command used to send an e-mail with no subject
MAIL="$MAILC -s" # Command used to send an e-mail with a subject
If that was messed up, wouldnt it prevent my alerts not working at all, because they work now. Its just I cant get them to work when I use the hobbitd_alert command.
thanx, michael
Lars Ebeling wrote:
Från: Michael A. Price [mailto:mprice at hst.nasa.gov] Skickat: den 23 oktober 2006 18:02 Till: hobbit at hswn.dk Ämne: [hobbit] hobbitd_alert command is not working ( maybe )
Hello,
My hobbit-alert.cfg file has the following line in it:
HOST=* MAIL mprice at hst.nasa.gov
Hi
00003217 2006-10-23 11:59:07 *** Match with 'HOST=*' *** 00003217 2006-10-23 11:59:07 Mail alert with command '/var/spool/mail/hobbit "Hobbit [12345] v43mgt:conn
CRITICAL (RED)"
mprice at hst.nasa.gov'
The command to send the mail look very odd. What have you as MAILC=?? in hobbitserver.cfg? Should be like # For the hobbitd_alert module
MAILC="mailx" # Command used to send an e-mail with no subject
MAIL="$MAILC -s" # Command used to send an e-mail with a subject
Lars
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
I think your on to something here, I am looking around for the problem now :-)
thanx, michael
Lars Ebeling wrote:
When I run hobbitd_alert I get this:
$ ./hobbitd_alert --test leopg9 msgs
00016117 2006-10-23 19:00:20 send_alert leopg9:msgs state Paging
00016117 2006-10-23 19:00:20 Matching host:service:page 'leopg9:msgs:' against r ule line 115
00016117 2006-10-23 19:00:20 *** Match with 'HOST=leopg9' ***
00016117 2006-10-23 19:00:20 Matching host:service:page 'leopg9:msgs:' against r ule line 116
00016117 2006-10-23 19:00:20 *** Match with 'MAIL hobbit REPEAT=1h COLOR=red REC OVERED' ***
00016117 2006-10-23 19:00:20 Mail alert with command 'mail -s "Hobbit [12345] le opg9:msgs CRITICAL (RED)" hobbit'
-----Ursprungligt meddelande----- Från: Michael A. Price [mailto:mprice at hst.nasa.gov] Skickat: den 23 oktober 2006 18:53 Till: hobbit at hswn.dk Ämne: Re: SV: [hobbit] hobbitd_alert command is not working ( maybe )
Lars,
Mine is defined as the following.
MAILC="mail" # Command used to send an e-mail with no subject
MAIL="$MAILC -s" # Command used to send an e-mail with a subject
If that was messed up, wouldnt it prevent my alerts not working at all, because they work now. Its just I cant get them to work when I use the hobbitd_alert command.
thanx, michael
Lars Ebeling wrote:
Från: Michael A. Price [mailto:mprice at hst.nasa.gov] Skickat: den 23 oktober 2006 18:02 Till: hobbit at hswn.dk Ämne: [hobbit] hobbitd_alert command is not working ( maybe )
Hello,
My hobbit-alert.cfg file has the following line in it:
HOST=* MAIL mprice at hst.nasa.gov
Hi
00003217 2006-10-23 11:59:07 *** Match with 'HOST=*' *** 00003217 2006-10-23 11:59:07 Mail alert with command '/var/spool/mail/hobbit "Hobbit [12345] v43mgt:conn
CRITICAL (RED)"
mprice at hst.nasa.gov'
The command to send the mail look very odd. What have you
as MAILC=??
in hobbitserver.cfg? Should be like # For the hobbitd_alert module
MAILC="mailx" # Command
used to send an
e-mail with no subject
MAIL="$MAILC -s" # Command
used to send an
e-mail with a subject
Lars
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
When I changed user (to root) I got the same as you:
./hobbitd_alert --test leopg9 msgs
00017560 2006-10-23 19:22:00 send_alert leopg9:msgs state Paging
00017560 2006-10-23 19:22:00 Matching host:service:page 'leopg9:msgs:' against r ule line 115
00017560 2006-10-23 19:22:00 *** Match with 'HOST=leopg9' ***
00017560 2006-10-23 19:22:00 Matching host:service:page 'leopg9:msgs:' against r ule line 116
00017560 2006-10-23 19:22:00 *** Match with 'MAIL hobbit REPEAT=1h COLOR=red REC OVERED' ***
00017560 2006-10-23 19:22:00 Mail alert with command '/var/mail/root "Hobbit [12 345] leopg9:msgs CRITICAL (RED)" hobbit'
-----Ursprungligt meddelande----- Från: Michael A. Price [mailto:mprice at hst.nasa.gov] Skickat: den 23 oktober 2006 19:05 Till: hobbit at hswn.dk Ämne: Re: SV: SV: [hobbit] hobbitd_alert command is not working ( maybe )
I think your on to something here, I am looking around for the problem now :-)
thanx, michael
Lars Ebeling wrote:
When I run hobbitd_alert I get this:
$ ./hobbitd_alert --test leopg9 msgs
00016117 2006-10-23 19:00:20 send_alert leopg9:msgs state Paging
00016117 2006-10-23 19:00:20 Matching host:service:page 'leopg9:msgs:' against r ule line 115
00016117 2006-10-23 19:00:20 *** Match with 'HOST=leopg9' ***
00016117 2006-10-23 19:00:20 Matching host:service:page 'leopg9:msgs:' against r ule line 116
00016117 2006-10-23 19:00:20 *** Match with 'MAIL hobbit REPEAT=1h COLOR=red REC OVERED' ***
00016117 2006-10-23 19:00:20 Mail alert with command 'mail -s "Hobbit [12345] le opg9:msgs CRITICAL (RED)" hobbit'
-----Ursprungligt meddelande----- Från: Michael A. Price [mailto:mprice at hst.nasa.gov] Skickat: den 23 oktober 2006 18:53 Till: hobbit at hswn.dk Ämne: Re: SV: [hobbit] hobbitd_alert command is not working ( maybe )
Lars,
Mine is defined as the following.
MAILC="mail" # Command used to send an e-mail with no subject
MAIL="$MAILC -s" # Command used to send an e-mail with a subject
If that was messed up, wouldnt it prevent my alerts not working at all, because they work now. Its just I cant get them to work when I use the hobbitd_alert command.
thanx, michael
Lars Ebeling wrote:
Från: Michael A. Price [mailto:mprice at hst.nasa.gov] Skickat: den 23 oktober 2006 18:02 Till: hobbit at hswn.dk Ämne: [hobbit] hobbitd_alert command is not working ( maybe )
Hello,
My hobbit-alert.cfg file has the following line in it:
HOST=* MAIL mprice at hst.nasa.gov
Hi
00003217 2006-10-23 11:59:07 *** Match with 'HOST=*' ***
00003217
2006-10-23 11:59:07 Mail alert with command '/var/spool/mail/hobbit "Hobbit [12345] v43mgt:conn
CRITICAL (RED)"
mprice at hst.nasa.gov'
The command to send the mail look very odd. What have you
as MAILC=??
in hobbitserver.cfg? Should be like # For the hobbitd_alert module
MAILC="mailx" # Command
used to send an
e-mail with no subject
MAIL="$MAILC -s" # Command
used to send an
e-mail with a subject
Lars
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
To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk
My hobbit-user (hobbit) does not have any environmentvariabel MAIL set. But the root user does.
Lars
I figured out the problem.
I unset my environmental variable for the hobbit user.
But now, when I run the hobbitd_alert command it looks good:
[hobbit at neuron hobbit]$ hobbitd_alert --test v43mgt cpu 00014397 2006-10-23 13:55:56 send_alert v43mgt:cpu state Paging 00014397 2006-10-23 13:55:56 Matching host:service:page 'v43mgt:cpu:' against rule line 129 00014397 2006-10-23 13:55:56 *** Match with 'HOST=*' *** 00014397 2006-10-23 13:55:56 Matching host:service:page 'v43mgt:cpu:' against rule line 129 00014397 2006-10-23 13:55:56 *** Match with 'HOST=*' *** 00014397 2006-10-23 13:55:56 Mail alert with command 'mail -s "Hobbit [12345] v43mgt:cpu CRITICAL (RED)" mprice at hst.nasa.gov'
It still doesnt send the email because the command line is wrong, it doesnt have a message body so it hangs in the process. The mail command expects a message body and the hobbitd_alert command is not giving it, so I hangs and never gets sent.
Does anyone know a way to get around this??
thanx for the help Lars and anyone else who knows what to do,
Michael A. Price
Lars Ebeling wrote:
My hobbit-user (hobbit) does not have any environmentvariabel MAIL set. But the root user does.
Lars
To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk
The problem is the my environmental variable $MAIL is set to /var/spool/mail/hobbit and I think this is trumping the programs setting of MAIL="$MAILC -s".
Is there a band aid for this???
thanx, michael
Lars Ebeling wrote:
When I run hobbitd_alert I get this:
$ ./hobbitd_alert --test leopg9 msgs
00016117 2006-10-23 19:00:20 send_alert leopg9:msgs state Paging
00016117 2006-10-23 19:00:20 Matching host:service:page 'leopg9:msgs:' against r ule line 115
00016117 2006-10-23 19:00:20 *** Match with 'HOST=leopg9' ***
00016117 2006-10-23 19:00:20 Matching host:service:page 'leopg9:msgs:' against r ule line 116
00016117 2006-10-23 19:00:20 *** Match with 'MAIL hobbit REPEAT=1h COLOR=red REC OVERED' ***
00016117 2006-10-23 19:00:20 Mail alert with command 'mail -s "Hobbit [12345] le opg9:msgs CRITICAL (RED)" hobbit'
-----Ursprungligt meddelande----- Från: Michael A. Price [mailto:mprice at hst.nasa.gov] Skickat: den 23 oktober 2006 18:53 Till: hobbit at hswn.dk Ämne: Re: SV: [hobbit] hobbitd_alert command is not working ( maybe )
Lars,
Mine is defined as the following.
MAILC="mail" # Command used to send an e-mail with no subject
MAIL="$MAILC -s" # Command used to send an e-mail with a subject
If that was messed up, wouldnt it prevent my alerts not working at all, because they work now. Its just I cant get them to work when I use the hobbitd_alert command.
thanx, michael
Lars Ebeling wrote:
Från: Michael A. Price [mailto:mprice at hst.nasa.gov] Skickat: den 23 oktober 2006 18:02 Till: hobbit at hswn.dk Ämne: [hobbit] hobbitd_alert command is not working ( maybe )
Hello,
My hobbit-alert.cfg file has the following line in it:
HOST=* MAIL mprice at hst.nasa.gov
Hi
00003217 2006-10-23 11:59:07 *** Match with 'HOST=*' *** 00003217 2006-10-23 11:59:07 Mail alert with command '/var/spool/mail/hobbit "Hobbit [12345] v43mgt:conn
CRITICAL (RED)"
mprice at hst.nasa.gov'
The command to send the mail look very odd. What have you
as MAILC=??
in hobbitserver.cfg? Should be like # For the hobbitd_alert module
MAILC="mailx" # Command
used to send an
e-mail with no subject
MAIL="$MAILC -s" # Command
used to send an
e-mail with a subject
Lars
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
On Mon, Oct 23, 2006 at 01:20:31PM -0400, Michael A. Price wrote:
The problem is the my environmental variable $MAIL is set to /var/spool/mail/hobbit and I think this is trumping the programs setting of MAIL="$MAILC -s".
Is there a band aid for this???
Run hobbitd_alert via the "bbcmd" utility. When the alert module is started by hobbitlaunch, it takes care of loading the necessary settings.
Regards, Henrik
On Mon, Oct 23, 2006 at 07:02:07PM +0200, Lars Ebeling wrote:
When I run hobbitd_alert I get this:
$ ./hobbitd_alert --test leopg9 msgs
When running hobbit commands by hand, ALWAYS run them through the "bbcmd" utility. This makes sure all of the settings from hobbitserver.cfg are loaded. So:
$ bbcmd hobbitd_alert --test ......
If you don't do that, then hobbitd_alert will pick up whatever value your MAIL environment has, instead of the one from hobbitserver.cfg.
Regards, Henrik
1 Step Closer, thanx guys...
I am still not getting emails alerts when I run hobbitd_alert by hand. Although I do get alerts from the hobbit server when things break.
here is the output below, I think it is because the alert is not creating text in the body so the 'mail ' command is hanging. Any work arounds for this???
[hobbit at chickeen hobbit]$ bbcmd hobbitd_alert --test v43mgt conn 2006-10-26 09:24:01 Using default environment file /home/hobbit/server/etc/hobbitserver.cfg 00024942 2006-10-26 09:24:01 send_alert v43mgt:conn state Paging 00024942 2006-10-26 09:24:01 Matching host:service:page 'v43mgt:conn:' against rule line 129 00024942 2006-10-26 09:24:01 *** Match with 'HOST=*' *** 00024942 2006-10-26 09:24:01 Matching host:service:page 'v43mgt:conn:' against rule line 129 00024942 2006-10-26 09:24:01 *** Match with 'HOST=*' *** 00024942 2006-10-26 09:24:01 Mail alert with command 'mail -s "Hobbit [12345] v43mgt:conn CRITICAL (RED)" mprice at hst.nasa.gov' 00024942 2006-10-26 09:24:01 Matching host:service:page 'v43mgt:conn:' against rule line 131 00024942 2006-10-26 09:24:01 Failed 'HOST=maina' (hostname not in include list)
thanx, michael
Michael A. Price Performance Network Engineering NASA/GSFC Code 440.8/LMB Greenbelt, Maryland 20770 Phone: 240-684-1356 Cell: 410-507-7476 e-mail: mprice at hst.nasa.gov
Henrik Stoerner wrote:
On Mon, Oct 23, 2006 at 07:02:07PM +0200, Lars Ebeling wrote:
When I run hobbitd_alert I get this:
$ ./hobbitd_alert --test leopg9 msgs
When running hobbit commands by hand, ALWAYS run them through the "bbcmd" utility. This makes sure all of the settings from hobbitserver.cfg are loaded. So:
$ bbcmd hobbitd_alert --test ......
If you don't do that, then hobbitd_alert will pick up whatever value your MAIL environment has, instead of the one from hobbitserver.cfg.
Regards, Henrik
To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Michael A. Price wrote:
1 Step Closer, thanx guys...
I am still not getting emails alerts when I run hobbitd_alert by hand. Although I do get alerts from the hobbit server when things break.
You're running --test. It shows you:
00024942 2006-10-26 09:24:01 Mail alert with command 'mail -s "Hobbit [12345] v43mgt:conn CRITICAL (RED)" mprice at hst.nasa.gov'
You have a winner! This will only test that a rule "fires" and triggers your mail command, which it did.
I would simply type in, as the hobbit user,
$ mail -s "Hobbit Blah Test Text (RED)" mprice at hst.nasa.gov
and see if you get a blank email with that subject. If so, it's something else. But if not, maybe the hobbit user can't access mail, or the mail command is broked, or something like that.
here is the output below, I think it is because the alert is not creating text in the body so the 'mail ' command is hanging. Any work arounds for this???
[hobbit at chickeen hobbit]$ bbcmd hobbitd_alert --test v43mgt conn 2006-10-26 09:24:01 Using default environment file /home/hobbit/server/etc/hobbitserver.cfg 00024942 2006-10-26 09:24:01 send_alert v43mgt:conn state Paging 00024942 2006-10-26 09:24:01 Matching host:service:page 'v43mgt:conn:' against rule line 129 00024942 2006-10-26 09:24:01 *** Match with 'HOST=*' *** 00024942 2006-10-26 09:24:01 Matching host:service:page 'v43mgt:conn:' against rule line 129 00024942 2006-10-26 09:24:01 *** Match with 'HOST=*' *** 00024942 2006-10-26 09:24:01 Mail alert with command 'mail -s "Hobbit [12345] v43mgt:conn CRITICAL (RED)" mprice at hst.nasa.gov' 00024942 2006-10-26 09:24:01 Matching host:service:page 'v43mgt:conn:' against rule line 131 00024942 2006-10-26 09:24:01 Failed 'HOST=maina' (hostname not in include list)
thanx, michael
Michael A. Price Performance Network Engineering NASA/GSFC Code 440.8/LMB Greenbelt, Maryland 20770 Phone: 240-684-1356 Cell: 410-507-7476 e-mail: mprice at hst.nasa.gov
Henrik Stoerner wrote:
On Mon, Oct 23, 2006 at 07:02:07PM +0200, Lars Ebeling wrote:
When I run hobbitd_alert I get this:
$ ./hobbitd_alert --test leopg9 msgs
When running hobbit commands by hand, ALWAYS run them through the "bbcmd" utility. This makes sure all of the settings from hobbitserver.cfg are loaded. So:
$ bbcmd hobbitd_alert --test ......
If you don't do that, then hobbitd_alert will pick up whatever value your MAIL environment has, instead of the one from hobbitserver.cfg.
Regards, Henrik
To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk
Rob Munsch Solutions For Progress IT www.solutionsforprogress.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFQMKjBvBcJFK6xYURAvr5AKCDXHuOTiE4Iu9SgeGQG5+7VU/UVACghLsQ 6T3Exn1Ozxjl5mQbXWZdUEY= =leiW -----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Michael A. Price wrote:
I am still not getting emails alerts when I run hobbitd_alert by hand. Although I do get alerts from the hobbit server when things break. I didn't read this well enough.
When you run the --test you get the block below showing you what WOULD fire if this were an actual emergency. *No email will actually be sent in test mode.*
here is the output below, I think it is because the alert is not creating text in the body so the 'mail ' command is hanging. Any work arounds for this???
[hobbit at chickeen hobbit]$ bbcmd hobbitd_alert --test v43mgt conn 2006-10-26 09:24:01 Using default environment file /home/hobbit/server/etc/hobbitserver.cfg 00024942 2006-10-26 09:24:01 send_alert v43mgt:conn state Paging 00024942 2006-10-26 09:24:01 Matching host:service:page 'v43mgt:conn:' against rule line 129 00024942 2006-10-26 09:24:01 *** Match with 'HOST=*' *** 00024942 2006-10-26 09:24:01 Matching host:service:page 'v43mgt:conn:' against rule line 129 00024942 2006-10-26 09:24:01 *** Match with 'HOST=*' *** 00024942 2006-10-26 09:24:01 Mail alert with command 'mail -s "Hobbit [12345] v43mgt:conn CRITICAL (RED)" mprice at hst.nasa.gov' 00024942 2006-10-26 09:24:01 Matching host:service:page 'v43mgt:conn:' against rule line 131 00024942 2006-10-26 09:24:01 Failed 'HOST=maina' (hostname not in include list)
thanx, michael
Michael A. Price Performance Network Engineering NASA/GSFC Code 440.8/LMB Greenbelt, Maryland 20770 Phone: 240-684-1356 Cell: 410-507-7476 e-mail: mprice at hst.nasa.gov
Henrik Stoerner wrote:
On Mon, Oct 23, 2006 at 07:02:07PM +0200, Lars Ebeling wrote:
When I run hobbitd_alert I get this:
$ ./hobbitd_alert --test leopg9 msgs
When running hobbit commands by hand, ALWAYS run them through the "bbcmd" utility. This makes sure all of the settings from hobbitserver.cfg are loaded. So:
$ bbcmd hobbitd_alert --test ......
If you don't do that, then hobbitd_alert will pick up whatever value your MAIL environment has, instead of the one from hobbitserver.cfg.
Regards, Henrik
To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk
Rob Munsch Solutions For Progress IT www.solutionsforprogress.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFQMLqBvBcJFK6xYURAnXVAJ9UQUIjibuICHG3kjXDAtqWRljgLACfZmAP fMeRgWb+jGPTRDTHphq2c8M= =V1cc -----END PGP SIGNATURE-----
Makes perfect sense...
Thanks Rob ;-)
Michael A. Price Performance Network Engineering NASA/GSFC Code 440.8/LMB Greenbelt, Maryland 20770 Phone: 240-684-1356 Cell: 410-507-7476 e-mail: mprice at hst.nasa.gov
Rob Munsch wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Michael A. Price wrote:
I am still not getting emails alerts when I run hobbitd_alert by hand. Although I do get alerts from the hobbit server when things break.
I didn't read this well enough.
When you run the --test you get the block below showing you what WOULD fire if this were an actual emergency. *No email will actually be sent in test mode.*
here is the output below, I think it is because the alert is not creating text in the body so the 'mail ' command is hanging. Any work arounds for this???
[hobbit at chickeen hobbit]$ bbcmd hobbitd_alert --test v43mgt conn 2006-10-26 09:24:01 Using default environment file /home/hobbit/server/etc/hobbitserver.cfg 00024942 2006-10-26 09:24:01 send_alert v43mgt:conn state Paging 00024942 2006-10-26 09:24:01 Matching host:service:page 'v43mgt:conn:' against rule line 129 00024942 2006-10-26 09:24:01 *** Match with 'HOST=*' *** 00024942 2006-10-26 09:24:01 Matching host:service:page 'v43mgt:conn:' against rule line 129 00024942 2006-10-26 09:24:01 *** Match with 'HOST=*' *** 00024942 2006-10-26 09:24:01 Mail alert with command 'mail -s "Hobbit [12345] v43mgt:conn CRITICAL (RED)" mprice at hst.nasa.gov' 00024942 2006-10-26 09:24:01 Matching host:service:page 'v43mgt:conn:' against rule line 131 00024942 2006-10-26 09:24:01 Failed 'HOST=maina' (hostname not in include list)
thanx, michael
Michael A. Price Performance Network Engineering NASA/GSFC Code 440.8/LMB Greenbelt, Maryland 20770 Phone: 240-684-1356 Cell: 410-507-7476 e-mail: mprice at hst.nasa.gov
Henrik Stoerner wrote:
On Mon, Oct 23, 2006 at 07:02:07PM +0200, Lars Ebeling wrote:
When I run hobbitd_alert I get this:
$ ./hobbitd_alert --test leopg9 msgs
When running hobbit commands by hand, ALWAYS run them through the "bbcmd" utility. This makes sure all of the settings from hobbitserver.cfg are loaded. So:
$ bbcmd hobbitd_alert --test ......
If you don't do that, then hobbitd_alert will pick up whatever value your MAIL environment has, instead of the one from hobbitserver.cfg.
Regards, Henrik
To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk
Rob Munsch Solutions For Progress IT www.solutionsforprogress.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFQMLqBvBcJFK6xYURAnXVAJ9UQUIjibuICHG3kjXDAtqWRljgLACfZmAP fMeRgWb+jGPTRDTHphq2c8M= =V1cc -----END PGP SIGNATURE-----
To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk
On Thu, Oct 19, 2006 at 02:46:37PM -0400, Rob Munsch wrote:
I'm pretty sure this is the trouble. Ladies and gentlemen, i am stumped. Following all the instructions in the docs, manpages, faqs and alien communications, as well as the very helpful samples provided by list members... no new file ever gets created in ~/data/rrd/<hostname>.
I have status, so i know client info is reaching server. I have copied almost verbatim the sample server-side script needed to parse this info and spit it back out to RRD. And i have eschewed NCV and gone in favor of hobbitd_channel --options to make this all work right, or so I thought.
Lets see your hobbitlaunch.cfg setup for the [rrdstatus] and [rrddata] tasks. Also, could you run this command for 10 minutes (as the hobbit user) and let us see what it reports:
bbcmd hobbitd_channel --channel=status
hobbitd_capture --tests=YOUR-COLUMN-NAME
That will tell us what kind of data arrives from your test script.
Does your server-side script run at all ? Put a line like echo "This is my script" >/tmp/itruns and see if the /tmp/itruns file appears when the status is updated.
Have you checked the Hobbit log-files in ~hobbit/server/logs/ ? The hobbitlaunch.log file and rrd-status.log file.
Regards, Henrik
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Henrik Stoerner wrote:
On Thu, Oct 19, 2006 at 02:46:37PM -0400, Rob Munsch wrote:
Lets see your hobbitlaunch.cfg setup for the [rrdstatus] and [rrddata] tasks.
here y'go: [rrdstatus] ENVFILE /home/hobbit/server/etc/hobbitserver.cfg NEEDS hobbitd CMD hobbitd_channel --channel=status - --log=$BBSERVERLOGS/rrd-status.log hobbitd_rrd --rrddir=$BBVAR/rrd - --extra-script=/home/hobbit/server/ext/cputemp.sh --extra-tests=cputemp # "rrddata" updates RRD files with information that arrives as "data" messages. # If you want RRD graphs of your monitoring BB data, then you want to run this. [rrddata] ENVFILE /home/hobbit/server/etc/hobbitserver.cfg NEEDS hobbitd CMD hobbitd_channel --channel=data - --log=$BBSERVERLOGS/rrd-data.log hobbitd_rrd --rrddir=$BBVAR/rrd - --extra-script=/home/hobbit/server/ext/cputemp.sh --extra-tests=cputemp
Also, could you run this command for 10 minutes (as the hobbit user) and let us see what it reports:
bbcmd hobbitd_channel --channel=status \ hobbitd_capture --tests=YOUR-COLUMN-NAME
hobbit at randomaccess ~/server/bin $ ./bbcmd hobbitd_channel - --channel=status hobbitd_capture --tests=cputemp 2006-10-20 11:48:54 Using default environment file /home/hobbit/server/etc/hobbitserver.cfg ## @@status#148 1161359362.326942 10.10.10.12 currant.tbb cputemp 1161361162 green green 1161269556 0 0 1161359311 status currant,tbb.cputemp green Fri Oct 20 11:49:22 EDT 2006 Current CPU Temperature: 100 degrees Fahrenheit. Chillin' like a Villain. ## @@status#156 1161359370.741441 10.10.10.12 currant.tbb cputemp 1161361170 green green 1161269556 0 0 1161359363 status currant,tbb.cputemp green Fri Oct 20 11:49:30 EDT 2006 Current CPU Temperature: 100 degrees Fahrenheit. Chillin' like a Villain. ## @@status#171 1161359422.240126 10.10.10.12 currant.tbb cputemp 1161361222 green green 1161269556 0 0 1161359372 status currant,tbb.cputemp green Fri Oct 20 11:50:22 EDT 2006 Current CPU Temperature: 100 degrees Fahrenheit. Chillin' like a Villain. etc.
That will tell us what kind of data arrives from your test script.
Does your server-side script run at all ? Put a line like echo "This is my script" >/tmp/itruns and see if the /tmp/itruns file appears when the status is updated.
hobbit at randomaccess /tmp $ ls -lah cp* - -rw-r--r-- 1 hobbit nogroup 29 Oct 20 11:52 cputemp.runs
Have you checked the Hobbit log-files in ~hobbit/server/logs/ ? The hobbitlaunch.log file and rrd-status.log file.
I have, found a type or two and corrected them. beyond that just the ubiquitous mysterious 2006-10-20 12:02:55 Tried to down BOARDBUSY: Invalid argument seems amiss. Everything else is pretending to work ,-) I think i'm out of actual "mechanical" errors; i think i'm just feeding Hobbit garbage data. And i still only have a text link instead of a graph, sigh. Anyone recommend any good RRD books..? :) thanks Henrik and everyone, - -- Rob Munsch Solutions For Progress IT www.solutionsforprogress.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFFOPOVBvBcJFK6xYURAp+RAJ9X5hMvKUCQAIUtArf+UE+kSCq42wCfU/jA aSL/wWJ+dCd9CZJXJsS34DQ= =zvFt -----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
This seemed the right way to do it. Are there any obvious "no, dumbass"es in these definitions..? With no good data, i should see a blank graph if these were right, yesno?
[cputemp] TITLE CPU Temperature YAXIS DegreesF DEF:cputemp=cputemp.rrd:cputemp:AVERAGE AREA:cputemp#00CC00:CPU Temp Avg GPRINT:cputemp:LAST: \: %5.11f (cur) GPRINT:cputemp:MAX: \: %5.1lf (max) GPRINT:cputemp:MIN: \: %5.1lf (min) GPRINT:cputemp:AVERAGE: \: %5.1lf (avg)\n
in cputemp.sh:
echo "DS:cputemp:GAUGE:600:0:U"
name the RRD for the host and test
echo cputemp.rrd
Analyze the message we got. If it's only one line, grab it.
The interesting number is the fourth element in this case.
CPU=grep CPU $FNAME | awk '{ print $4 }'
echo $CPU
Rob Munsch Solutions For Progress IT www.solutionsforprogress.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFORKmBvBcJFK6xYURAppwAJwPNDmGEibGhuRVuMF2N8Cac1eyKQCeJVzH yy/Ln3g2XlFWMDVF7xIGl0o= =6TiM -----END PGP SIGNATURE-----
participants (6)
-
greg.hubbard＠eds.com
-
henrik＠hswn.dk
-
lars.ebeling＠leopg9.no-ip.org
-
mprice＠hst.nasa.gov
-
ralphmitchell＠gmail.com
-
rmunsch＠solutionsforprogress.com