Koozali.org: home of the SME Server

Announcement: Howto monitor your sme7 general performance us

Offline hanscees

  • *
  • 267
  • +0/-0
    • nl.linkedin.com/in/hanscees/
Announcement: Howto monitor your sme7 general performance us
« on: October 24, 2005, 11:04:35 PM »
Hi,
I have made this "Howto monitor your sme7 general performance using with RRDtool and collectd" available at:
http://www.hanscees.com/sme7/collectd4sme7howto.html

Here is what you get: take a look!
http://www.hanscees.com/sme7/collection.cgi.html

What this howto will and will not produce:

    * 1. This howto works in that it produces RRD-tool graphics of performance relevant data like cpu (load), memory usage, IO performance and interface statistics (traffic flow).
    * 2. This howto is not yet finished/polished toward sme. Anyone is called upon to improve it. I will be willing to help if needed.
    * 3. This howto should also generally work for centhos4 and redhat 4 enterprise, since sme7 is based upon that.

Special thanks to Florian Forster (of collectd ) for helping me out with some initial troubleshooting. And of course to the volunteers of sme7 (www.contribs.org) for making a great platform.





Hope it is of use to some. I am sure it can be improved. Please let me in on your good ideas.

greetings
Hans-Cees
nl.linkedin.com/in/hanscees/

Offline hanscees

  • *
  • 267
  • +0/-0
    • nl.linkedin.com/in/hanscees/
Announcement: Howto monitor your sme7 general performance us
« Reply #1 on: October 27, 2005, 12:25:33 AM »
Hi
if anyone uses this howto and has an opinion about it I sure would like to know.

cheers

Hans-Cees
nl.linkedin.com/in/hanscees/

Offline Reinhold

  • *
  • 517
  • +0/-0
    • http://127.0.0.1
Announcement: Howto monitor your sme7 general performance us
« Reply #2 on: October 27, 2005, 07:14:00 PM »
Hi Hans,

Just found the HowTo and followed it on my SME7 testbox.
Nice work ... :-D :-D :-D
I do get the results you show on your site with little effort a fe minutes later.

All in all this sure seems to have to potential to replace/follow the defunct sysmon in SME 7.

I have had no time to dig deeper so 2 questions:
- Did you test the lm_sensors part? I could not (yet) get this to display a correct .rrd graph for temps and fan speed ("sensors" itself does work correctly).
- Since I support a lot of Raid5 setups I find myself using hddtemp very often. Surprisingly Version 3.2.2 of collectd does support this. Do you have an updated rpm already .-)

Again: Nice work.

Regards
Reinhold
............

Offline hanscees

  • *
  • 267
  • +0/-0
    • nl.linkedin.com/in/hanscees/
Announcement: Howto monitor your sme7 general performance us
« Reply #3 on: October 28, 2005, 12:03:18 AM »
Quote from: "Reinhold"

I have had no time to dig deeper so 2 questions:
- Did you test the lm_sensors part? I could not (yet) get this to display a correct .rrd graph for temps and fan speed ("sensors" itself does work correctly).
- Since I support a lot of Raid5 setups I find myself using hddtemp very often. Surprisingly Version 3.2.2 of collectd does support this. Do you have an updated rpm already .-)

Again: Nice work.

Regards
Reinhold


Thanks!
I have not done anything to the package itself. I have not tested sensors.
I have produced new rpm's, but did not test them at all. If you want please test them.


http://www.hanscees.com/sme7/collectd-3.2.0-1.i386.rpm
http://www.hanscees.com/sme7/collectd-sensors-3.2.0-1.i386.rpm
http://www.hanscees.com/sme7/collectd-hddtemp-3.2.0-1.i386.rpm


greetings

Hans-cees
nl.linkedin.com/in/hanscees/

Offline Reinhold

  • *
  • 517
  • +0/-0
    • http://127.0.0.1
Announcement: Howto monitor your sme7 general performance us
« Reply #4 on: October 28, 2005, 06:38:39 PM »
Hi Hans,

Had little time so I hope for "a rainy weekend"  ;-)

I did manage a short look into the sensors.c file however and indeed it says:
 /* Only temperature for now.. */
and it seemingly tries the wrong sensor values for this PC.
I will look into this over the weekend ...

Thanks for the updated rpms. I'll grab them and report as to the functionality of hddtemp ... asap.
:pint:

Regards
Reinhold
............

Offline Reinhold

  • *
  • 517
  • +0/-0
    • http://127.0.0.1
Announcement: Howto monitor your sme7 general performance us
« Reply #5 on: October 28, 2005, 07:53:02 PM »
Hi Hans,

Short feedback: No hddtemp data :-(

longer version:
Code: [Select]
rpm -Uvh collectd-3.2.0-1.i386.rpm
Error: Failed dependencies:
        perl(XML::Dumper) is needed by collectd-3.2.0-1.i386
        collectd = 3.0.0 is needed by (installed) collectd-sensors-3.0.0-2.i386


... so there are some spec ? problems still.

I decided to install the "official Centos" XML-Dumper

Code: [Select]
wget ftp://ftp.nluug.nl/pub/os/Linux/distr/CentOS/4.2/os/alpha/CentOS/RPMS/perl-XML-Dumper-0.71-2.noarch.rpm
rpm -Uvh perl-XML-Dumper-0.71-2.noarch.rpm


... and (see above) since collectd (falsely ?!) needs still it's predecessor 3.0.0 I went nodeps like you did...

Code: [Select]
rpm -Uvh --nodeps collectd-3.2.0-1.i386.rpm
Preparing...                ########################################### [100%]
   1:collectd               ########################################### [100%]

rpm -Uvh collectd-hddtemp-3.2.0-1.i386.rpm
Preparing...                ########################################### [100%]
   1:collectd-hddtemp       ########################################### [100%]

 rpm -Uvh collectd-sensors-3.2.0-1.i386.rpm
Preparing...                ########################################### [100%]
   1:collectd-sensors       ########################################### [100%]


the new cgi is obviously needed so (with the directories like in your howto)

Code: [Select]
cp /usr/share/doc/collectd-3.2.0/contrib/museum/collection.cgi /home/e-smith/files/ibays/stats/cgi-bin/collection.cgi

... same data as before... collectd does not generate hddtemp-data
...but hddtemp code was installed

Code: [Select]
rpm -ql collectd-hddtemp
/usr/lib/collectd/hddtemp.so
/usr/lib/collectd/hddtemp.so.0
/usr/lib/collectd/hddtemp.so.0.0.0


Note that the "-p www.contribs.org" does give a .rrd but not graph :-(

I am in a bit of a rush now...
just wanted to give the test of the hddtemp rpms (thanks for the work)
...more over the weekend (hopefully)

Let's polish this stuff ;-)

Regards
Reinhold

P.S.: For one reason or another I do have sensor data for TEMP and FANSPEED as .rrd now (3 graphs which is the correct number ... but the data is still invalid)
............

Offline hanscees

  • *
  • 267
  • +0/-0
    • nl.linkedin.com/in/hanscees/
Announcement: Howto monitor your sme7 general performance us
« Reply #6 on: October 28, 2005, 08:33:47 PM »
Quote from: "Reinhold"
Hi Hans,

Short feedback: No hddtemp data :-(

longer version:
Code: [Select]
rpm -Uvh collectd-3.2.0-1.i386.rpm
Error: Failed dependencies:
        perl(XML::Dumper) is needed by collectd-3.2.0-1.i386
        collectd = 3.0.0 is needed by (installed) collectd-sensors-3.0.0-2.i386


... so there are some spec ? problems still.



But you should update collectd-sensors-3.0.0-2.i386 with the newer version of sensors as well I think.?

Quote


I decided to install the "official Centos" XML-Dumper

Code: [Select]
wget ftp://ftp.nluug.nl/pub/os/Linux/distr/CentOS/4.2/os/alpha/CentOS/RPMS/perl-XML-Dumper-0.71-2.noarch.rpm
rpm -Uvh perl-XML-Dumper-0.71-2.noarch.rpm


... and (see above) since collectd (falsely ?!) needs still it's predecessor 3.0.0 I went nodeps like you did...

Code: [Select]
rpm -Uvh --nodeps collectd-3.2.0-1.i386.rpm
Preparing...                ########################################### [100%]
   1:collectd               ########################################### [100%]

rpm -Uvh collectd-hddtemp-3.2.0-1.i386.rpm
Preparing...                ########################################### [100%]
   1:collectd-hddtemp       ########################################### [100%]

 rpm -Uvh collectd-sensors-3.2.0-1.i386.rpm
Preparing...                ########################################### [100%]
   1:collectd-sensors       ########################################### [100%]



the new cgi is obviously needed so (with the directories like in your howto)



The collection.cgi is fine for a start, but the maintainer shows little interest in beefing it up at this moment I am afraid.

So I am looking for a good perl hacker to change that script. I have been staring at it, but do not understand the code I am afraid.

Quote




Code: [Select]
cp /usr/share/doc/collectd-3.2.0/contrib/museum/collection.cgi /home/e-smith/files/ibays/stats/cgi-bin/collection.cgi

... same data as before... collectd does not generate hddtemp-data
...but hddtemp code was installed

Code: [Select]
rpm -ql collectd-hddtemp
/usr/lib/collectd/hddtemp.so
/usr/lib/collectd/hddtemp.so.0
/usr/lib/collectd/hddtemp.so.0.0.0


Note that the "-p www.contribs.org" does give a .rrd but not graph :-(


another collection.cgi issue.
Quote



I am in a bit of a rush now...
just wanted to give the test of the hddtemp rpms (thanks for the work)
...more over the weekend (hopefully)

Let's polish this stuff ;-)

Regards
Reinhold

P.S.: For one reason or another I do have sensor data for TEMP and FANSPEED as .rrd now (3 graphs which is the correct number ... but the data is still invalid)


Perhaps we should just drop collection.cgi and make a new script, but that seems such a waste.

greetings

Hans-Cees
nl.linkedin.com/in/hanscees/

Offline Reinhold

  • *
  • 517
  • +0/-0
    • http://127.0.0.1
Announcement: Howto monitor your sme7 general performance us
« Reply #7 on: October 29, 2005, 06:47:20 PM »
Hans,

I did a bit of fumbling ...

- A new lm_sensors setup rpm + search a la wellsi.com ... and whatshallIsay ...works now !
I get Case TEMP & CPU TEMP (temp1 and temp2), and CPU FAN Data.
And -(un)happy surprise- I found out that the test-servers cpu-fan has intermittend problems ...
see below... IMO that alone makes a tool like this invaluable.
Note: Fan-scale is Celsius should read RPM -cosmetic-

- A bit of "c"ing into the code and ...Hddtemp does need the daemon mode i.e. you need to start it before "connected"-ing:
Code: [Select]
hddtemp -d /dev/hda
...so it connects now but data is still wrong for the test HD  :cry:

- "Processes" currently gives no .rrd output whatsoever :-?

- Ping logging as in :
Code: [Select]
collected -p www.contribs.org
does work too. See below how "beautifully bad" it pinged last night (while europeans should be sleeping) :-D

- Note: This is an SME7 server-only so no bandwith (yet) on eth1...

AND NOW IT DOES GET "COLORFUL" :roll:
- Although my perl is rather "pedestrian", tonight I cut&paste-ed my way to a output.cgi that does please me enough to continue...

Now here are the graphs in all their "splendor"
(Folks please note ... this is not on contribs so it will only strain any viewer's_&_photobucket's not Ruffdog's bandwitdth)  8-)














Right now the display above is rather special for my test SME7...

(I only display the .rrd files that collectd places in /var/lib/collectd/ with a cut&paste version of collection.cgi)
... but...
with some fiddling I feel able to make this universal enough for display on a server-manager page.

Regards
Reinhold

P.S.: Had a short glimpse on your updated website ... nice going!
............

Offline hanscees

  • *
  • 267
  • +0/-0
    • nl.linkedin.com/in/hanscees/
Announcement: Howto monitor your sme7 general performance us
« Reply #8 on: October 29, 2005, 11:08:50 PM »
Quote from: "Reinhold"
Hans,

I did a bit of fumbling ...

- A new lm_sensors setup rpm + search a la wellsi.com ... and whatshallIsay ...works now !


a la welsii.com??

Quote



I get Case TEMP & CPU TEMP (temp1 and temp2), and CPU FAN Data.
And -(un)happy surprise- I found out that the test-servers cpu-fan has intermittend problems ...
see below... IMO that alone makes a tool like this invaluable.
Note: Fan-scale is Celsius should read RPM -cosmetic-

- A bit of "c"ing into the code and ...Hddtemp does need the daemon mode i.e. you need to start it before "connected"-ing:
Code: [Select]
hddtemp -d /dev/hda



Could you state that a little more explicit? Do you need to give the hddtmp command before running collectd?
Doesn't collectd start the hddtmp up?

Quote



...so it connects now but data is still wrong for the test HD  :cry:

- "Processes" currently gives no .rrd output whatsoever :-?

- Ping logging as in :
Code: [Select]
collected -p www.contribs.org
does work too. See below how "beautifully bad" it pinged last night (while europeans should be sleeping) :-D

- Note: This is an SME7 server-only so no bandwith (yet) on eth1...

AND NOW IT DOES GET "COLORFUL" :roll:
- Although my perl is rather "pedestrian", tonight I cut&paste-ed my way to a output.cgi that does please me enough to continue...


could you post it or somewhere? Or mail it to me? My perl is probably worse than yours.
I asked Floria Foster from collectd to "upgrade " the scirpt. See the collectd mailing list at
http://mailman.verplant.org/listinfo/collectd

I would like a script that gives you the choice of a time frame, so you can see both day/week/month data. That is good to spot trends. I know howto do that with  RRD,
see
http://www.macsat.com/rrdtool.php
and the script that leads to.



Quote

(I only display the .rrd files that collectd places in /var/lib/collectd/ with a cut&paste version of collection.cgi)
... but...
with some fiddling I feel able to make this universal enough for display on a server-manager page.


that would be great indeed. I feel comfertable to make a howto, but hacking server-manager is beyond what I can do now. I just don't understand how it works yet.

Quote



Regards
Reinhold

P.S.: Had a short glimpse on your updated website ... nice going!


Thanks, it was a bit late when I finished it:-)

Hans-Cees
nl.linkedin.com/in/hanscees/

matsk

Announcement: Howto monitor your sme7 general performance us
« Reply #9 on: October 30, 2005, 12:47:04 AM »
To simplify the installation of cpan modules you could use
Code: [Select]
# install perl-xml-dumper
cpan install XML::Dumper


instead of the code :
Code: [Select]
wget http://ftp.debian.org/debian/pool/main/libx/libxml-dumper-perl/libxml-dumper-perl_0.79.orig.tar.gz

# install perl-xml-dumper
tar -zxvf libxml-dumper-perl_0.79.orig.tar.gz
cd XML-Dumper-0.79
perl Makefile.PL
make
make test
make install


The only thing that could be confusing is that the first time cpan is executed it has to be configured.
The advantage with cpan is that it takes care of dependencies in the same matter that yum does and will keep your cpan modules updated and at the latest version.

/Mats
PS. And hanscees: could you update your webpage "script" with the latest rpm's that you have buildt ?

matsk

Announcement: Howto monitor your sme7 general performance us
« Reply #10 on: October 30, 2005, 01:06:52 AM »
If you would like to have a "short URL" you could create a index.html
Code: [Select]
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<html>
<head>
<title>SME statistics</title>
<meta http-equiv="REFRESH" content="0;url=http://the ip or hostname/stats/cgi-bin/collection.cgi"></HEAD>
<BODY>

</BODY>
</HTML>


This will redirect you to the cgi script and you dont have to remember it, just surf to "http://the ip or hostname/stats/"

Offline Reinhold

  • *
  • 517
  • +0/-0
    • http://127.0.0.1
Announcement: Howto monitor your sme7 general performance us
« Reply #11 on: October 30, 2005, 10:37:15 AM »
Hans,

The damned CPU-Fan really died on me... so I lost access to my test SME7 ... at least for the weekend :cry:

As to your other questions:

- Ian Wells, very active contribs member (Thanks Ian!), has done a lot on lm_sensors see here
This will tell you how to find and activate the sensors in your PC as used by collected.
Note: SME7 does include a working version of lm_sensors - a new rpm install should not be needed ...


- hddtemp needs to be installed separately.
Best get it from the DAG Centos repo - or a mirror since it seems down/busy a lot recently. See here
After installation you can start hddtemp in two different modes.
First mode gives the temp data directly, second mode goes into daemon state (background)
and you can access the data via a port on the pc locally, try (ssh'd into SME7)
Code: [Select]
hddtemp -d /dev/hd[abcd]
telnet localhost 7634

man hddtemp will tell you in detail what this means and a look into hddtemp.c from collectd shows  what it should do :-x  


Last not least:
sysmon ...on SME 6.0 does give more and better organized data in the fashion you (too) would like to see, namely day/month/year values.  
Have a look. I do have the rpm source available locally or here (?)and this is not on the SME7 test server :-)  
(Note: original was from eneo/ShadLord so the above may be incorrectly googled.)
- Maybe best for us to look into the code there before cutting&pasting more
Perl... which I (still) sort of dislike for being so ... "unforgiving" :-(

Regards
Reinhold

matsk:
Perl pro? Please have a look into sysmon src as above!
(lazy like I tend to be) I already did forward my own smestats/index.html (like you proposed here) but
IMHO collectd primarily has to go into a server-manager-link to be accessible... and it has to refresh there too ;-)
............

Offline morpheusx

  • *
  • 35
  • +0/-0
Announcement: Howto monitor your sme7 general performance us
« Reply #12 on: October 30, 2005, 08:12:04 PM »
Thanks for the FAQ Hans. It worked fine for me on 7.0b4.

Couple of questions...

I added the new 3.2 collectd rpm's and hddtemp as you posted above, but no new graphs appeared on my ibay. Do I have to manually add hddtemp into the stats somehow? I do have a set of 3 hddtemp.so files in /usr/lib/collectd and hddtemp is running as a daemon.

telnet localhost 7634
Trying 127.0.0.1...
Connected to localhost.
Escape character is '^]'.
|/dev/hda|Maxtor 6B200P0|29|C|
Connection closed by foreign host.


Running 'cpan install XML::Dumper' as matsk suggested caused a huge amount of errors. I don't know why but it seems to be a gcc path issue?

[root@host cpustats]# cpan install XML::Dumper
CPAN: Storable loaded ok
Going to read /root/.cpan/Metadata
  Database was generated on Sun, 30 Oct 2005 08:13:39 GMT
Running install for module install
Running make for J/JU/JUNOS/junoscript-perl-6.4I0.tar.gz
CPAN: Digest::MD5 loaded ok
CPAN: Compress::Zlib loaded ok
Checksum for /root/.cpan/sources/authors/id/J/JU/JUNOS/junoscript-perl-6.4I0.tar.gz ok
Scanning cache /root/.cpan/build for sizes
junoscript-perl-6.4I0/
[snip]
Removing previously used /root/.cpan/build/junoscript-perl-6.4I0

  CPAN.pm: Going to build J/JU/JUNOS/junoscript-perl-6.4I0.tar.gz

Checking if your kit is complete...
Looks good
Warning: prerequisite IO::Tty 0 not found.
Writing Makefile for junoscript-perl
---- Unsatisfied dependencies detected during [J/JU/JUNOS/junoscript-perl-6.4I0.tar.gz] -----
    IO::Tty
Shall I follow them and prepend them to the queue
of modules we are processing right now? [yes]
Running make test
  Delayed until after prerequisites
Running make install
  Delayed until after prerequisites
Running install for module IO::Tty
Running make for R/RG/RGIERSIG/IO-Tty-1.02.tar.gz
Checksum for /root/.cpan/sources/authors/id/R/RG/RGIERSIG/IO-Tty-1.02.tar.gz ok
IO-Tty-1.02/
IO-Tty-1.02/Pty.pm
IO-Tty-1.02/Makefile.PL
IO-Tty-1.02/ChangeLog
IO-Tty-1.02/test.pl
IO-Tty-1.02/Tty.xs
IO-Tty-1.02/Tty.pm
IO-Tty-1.02/README
IO-Tty-1.02/try
IO-Tty-1.02/MANIFEST
Removing previously used /root/.cpan/build/IO-Tty-1.02

  CPAN.pm: Going to build R/RG/RGIERSIG/IO-Tty-1.02.tar.gz

Now let's see what we can find out about your system
(logfiles of failing tests are available in the conf/ dir)...

ERROR: cannot run the configured compiler 'gcc'
(see conf/compilerok.log).  Please fix this by adjusting
perls Config.pm, creating a symlink to the right location
or maybe by just adding it to your PATH.  You could also build
perl yourself.

Running make test
  Make had some problems, maybe interrupted? Won't test
Running make install
  Make had some problems, maybe interrupted? Won't install
Running make for J/JU/JUNOS/junoscript-perl-6.4I0.tar.gz
  Is already unwrapped into directory /root/.cpan/build/junoscript-perl-6.4I0

  CPAN.pm: Going to build J/JU/JUNOS/junoscript-perl-6.4I0.tar.gz

cp lib/JUNOS/Access/clear_text.pm blib/lib/JUNOS/Access/clear_text.pm
cp lib/JUNOS/Access/stubs.pm blib/lib/JUNOS/Access/stubs.pm
cp lib/JUNOS/Methods.pm blib/lib/JUNOS/Methods.pm
cp lib/JUNOS/Access.pm blib/lib/JUNOS/Access.pm
cp lib/JUNOS/Response.pm blib/lib/JUNOS/Response.pm
cp lib/JUNOS/Access/ssh.pm blib/lib/JUNOS/Access/ssh.pm
cp lib/JUNOS/Device.pm blib/lib/JUNOS/Device.pm
cp lib/JUNOS/6.4I0/jkernel_methods.pl blib/lib/JUNOS/6.4I0/jkernel_methods.pl
cp lib/JUNOS/version.pm blib/lib/JUNOS/version.pm
cp lib/JUNOS/6.4I0/jcrypto_methods.pl blib/lib/JUNOS/6.4I0/jcrypto_methods.pl
cp lib/JUNOS/Access/xnm.pm blib/lib/JUNOS/Access/xnm.pm
cp lib/JUNOS/Trace.pm blib/lib/JUNOS/Trace.pm
cp lib/JUNOS/6.4I0/jroute_methods.pl blib/lib/JUNOS/6.4I0/jroute_methods.pl
cp lib/JUNOS/Access/rsh.pm blib/lib/JUNOS/Access/rsh.pm
cp lib/JUNOS/Access/ssl.pm blib/lib/JUNOS/Access/ssl.pm
cp lib/JUNOS/Access/telnet.pm blib/lib/JUNOS/Access/telnet.pm
cp lib/JUNOS/DOM/Parser.pm blib/lib/JUNOS/DOM/Parser.pm
Manifying blib/man3/JUNOS::Access::clear_text.3pm
Manifying blib/man3/JUNOS::Access::stubs.3pm
Manifying blib/man3/JUNOS::Methods.3pm
Manifying blib/man3/JUNOS::Access.3pm
Manifying blib/man3/JUNOS::Access::ssh.3pm
Manifying blib/man3/JUNOS::Response.3pm
Manifying blib/man3/JUNOS::Access::rsh.3pm
Manifying blib/man3/JUNOS::Access::ssl.3pm
Manifying blib/man3/JUNOS::Device.3pm
Manifying blib/man3/JUNOS::Access::telnet.3pm
Manifying blib/man3/JUNOS::Trace.3pm
Manifying blib/man3/JUNOS::Access::xnm.3pm
Manifying blib/man3/JUNOS::DOM::Parser.3pm
  /usr/bin/make  -- OK
Running make test
PERL_DL_NONLAZY=1 /usr/bin/perl "-MExtUtils::Command::MM" "-e" "test_harness(0, 'blib/lib', 'blib/arch')" t/*.t
t/base-1....Can't locate IO/Pty.pm in @INC (@INC contains: /root/.cpan/build/junoscript-perl-6.4I0/blib/lib/5.8.5/i386-linux-thread-multi
[snip]

/usr/lib/perl5/vendor_perl/5.8.0 /usr/lib/perl5/vendor_perl .) at /root/.cpan/build/junoscript-perl-6.4I0/blib/lib/JUNOS/Access.pm line 41.
BEGIN failed--compilation aborted at /root/.cpan/build/junoscript-perl-6.4I0/blib/lib/JUNOS/Access.pm line 41.
Compilation failed in require at /root/.cpan/build/junoscript-perl-6.4I0/blib/lib/JUNOS/Device.pm line 45.
BEGIN failed--compilation aborted at /root/.cpan/build/junoscript-perl-6.4I0/blib/lib/JUNOS/Device.pm line 45.
Compilation failed in require at t/base-2.t line 12.
BEGIN failed--compilation aborted at t/base-2.t line 12.
t/base-2....dubious                                                          
        Test returned status 2 (wstat 512, 0x200)
FAILED--2 test scripts could be run, alas--no output ever seen
make: *** [test_dynamic] Error 2
  /usr/bin/make test -- NOT OK
Running make install
  make test had returned bad status, won't install without force
XML::Dumper is up to date.

Offline hanscees

  • *
  • 267
  • +0/-0
    • nl.linkedin.com/in/hanscees/
Announcement: Howto monitor your sme7 general performance us
« Reply #13 on: October 30, 2005, 09:12:38 PM »
Quote from: "matsk"
To simplify the installation of cpan modules you could use
Code: [Select]
# install perl-xml-dumper
cpan install XML::Dumper


instead of the code :
Code: [Select]
wget http://ftp.debian.org/debian/pool/main/libx/libxml-dumper-perl/libxml-dumper-perl_0.79.orig.tar.gz

# install perl-xml-dumper
tar -zxvf libxml-dumper-perl_0.79.orig.tar.gz
cd XML-Dumper-0.79
perl Makefile.PL
make
make test
make install


The only thing that could be confusing is that the first time cpan is executed it has to be configured.
The advantage with cpan is that it takes care of dependencies in the same matter that yum does and will keep your cpan modules updated and at the latest version.

/Mats
PS. And hanscees: could you update your webpage "script" with the latest rpm's that you have buildt ?


That might work very nicely. But it could stab you in the back badly too. cpan will get you a new perl version if some dependency asks it too. I would not recommend this in a production environment at all. I would choose the rpm, in which case you do not need to install I think. Just do rpm -ihv ... I think

I am updating my howto, but not with everything. I want to keep it stable. If we have something better that is tested and works I will gladly post it there.

Hans-Cees
nl.linkedin.com/in/hanscees/

Offline hanscees

  • *
  • 267
  • +0/-0
    • nl.linkedin.com/in/hanscees/
Announcement: Howto monitor your sme7 general performance us
« Reply #14 on: October 30, 2005, 10:14:35 PM »
Quote from: "Reinhold"
Hans,

- Ian Wells, very active contribs member (Thanks Ian!), has done a lot on lm_sensors see here
This will tell you how to find and activate the sensors in your PC as used by collected.
Note: SME7 does include a working version of lm_sensors - a new rpm install should not be needed ...


I have followed this lm_sensors stuff.
In sme7 lm_sensors is installed. But it does not seem to have all kinds of via modules.
Can anynbody else check if their sme7 has lm_sensor modules? I cannot find vt1211 nor vt8235. I have a via -board (via pd) and should copy a sensor module. But I do not know where to copy it from .

This site:
http://seanm.ca/eden/lmsensors.html

says:
Copy the vt8231 file to /lib/modules/uname -r/kernel/drivers/sensors.

But I do not know where to get that module. It should be in lm_sensors already if I understand it right, since it is supported bu lm_sensors.

Hans-Cees
nl.linkedin.com/in/hanscees/