load.sh issue with CPU-CORE rrd creation

jjsk1
jjsk1
in NetApp

I'm trying to run load.sh for the first time and getting this error w/ stor2rrd 2.01

RRDp::create read error : /usr/local/home/lpar2rrd/stor2rrd/data/xx/CPU-CORE/netapp1clus1-01 -processor0.rrd : ERROR: unknown option 'p' at bin/Xorux_lib.pm line 61. : No such file or directory

create_rrd err : unable to create /usr/local/home/lpar2rrd/stor2rrd/data/xx/CPU-CORE/"netapp1clus1-01 "-processor0.rrd (filesystem is full?) at bin/LoadDataModule.pm: line 1273


looks like it's trying to add an extra space in the middle of the filename

in the config.html there is a space after cluster name like this

Hostadapter Level Configuration
------------------------
name,node_name,node_uuid
0a,"netapp1clus1-01 ",
0b,"netapp1clus1-01 ",

...

CPU-core Level Configuration
------------------------
cpu,node_name,node_uuid
processor0,"netapp1clus1-01 ",
processor1,"netapp1clus1-01 ",

any idea how to fix this?

Comments

  • Pavel
    Pavel
    Hi,

    ok, I see the problem, your node name contains a space. This would not be a problem, problem is double quotes.

    Fix:
    1. install the latest 2.01
    2. get this fix:
    http://www.stor2rrd.com/download/LoadDataModule.pm.gz

    Gunzip it and copy to /home/stor2rrd/stor2rrd/bin (755, stor2rrd owner)
    -rwxr-xr-x 1 stor2rrd stor2rrd 801704 Jun 15 09:53 bin/LoadDataModule.pm
    If your web browser gunzips it automatically then just rename it: mv LoadDataModule.pm.gz LoadDataModule.pm

    let it work, the error should disappear.

    Let us know, we have not simulated it but it should work.



  • jjsk1
    jjsk1
    edited June 2017

    thanks but no luck yet.  we have checked and there does not appear to be a space at the end of the node name on the netapp.

    however some commands for example "system node show" do add a space in the output.  I think the part of stor2rrd that initially reads into %netapp hash needs to trim the node name and remove any quotes or spaces, although I could be wrong.

    I have checked some other netapp arrays and they all produce this extra space after node name even though it's not actually part of the name..

    from /usr/local/home/lpar2rrd/stor2rrd/load_netappperf.sh:

    Argument ""a8f2d8ca-eaa5-11e4-a465-459d56f1691 "" isn't numeric in division (/) at /usr/local/home/lpar2rrd/stor2rrd/bin/naperf.pl line 837.
    Argument ""d5ff1992-eaa5-11e4-8e7-d71363c1ed30 "" isn't numeric in division (/) at /usr/local/home/lpar2rrd/stor2rrd/bin/naperf.pl line 837.
    Use of uninitialized value $RRDp::error_mode in string eq at /usr/lib64/perl5/vendor_perl/RRDp.pm line 168, <FH> line 431.
    ERROR: /usr/local/home/lpar2rrd/stor2rrd/data/xx/CPU-CORE/netapp1clus1-01 -processor0.rrd: conversion of '- ' to float not complete: tail '- ' at /usr/local/home/lpar2rrd/stor2rrd/bin/LoadDataModule.pm line 16740.

    and load.sh:

    Use of uninitialized value $RRDp::error_mode in string eq at /usr/lib64/perl5/vendor_perl/RRDp.pm line 168, <FH> line 431.
    xx : load_host failed: ERROR: /usr/local/home/lpar2rrd/stor2rrd/data/xx/CPU-CORE/netapp1clus1-01 -processor0.rrd: conversion of '- ' to float not complete: tail '- ' at bin/LoadDataModule.pm line 16740.
     ./bin/storage.pl:186

  • jjsk1
    jjsk1
    edited June 2017

    if this helps here is how node_name strings are stored in the cache file:

     'netapp1clus1-01:clus2' => {
                                                      'node_name' => '"netapp1clus1-01 "',
                                                      'vserver_name' => '"netapp1clus1-01 "',

    they have double quotes and a space as part of the string.  many occurrences of the same bad string.  this must be how it's returned from the netapp  

    I got the above info from this code

    #!/usr/bin/perl -w

    use strict;
    use Storable;
    use Data::Dumper;

    my $netapp=retrieve('xx.cache.file');

    print Dumper($netapp);

  • Pavel
    Pavel
    Hi,

    thanks, we will provide a fix soon. Double quotes causing a problem, looks like they are even in UUID string.
  • Jirka
    Jirka
    edited June 2017
    @jjsk1:

    could you please provide us with some degug data? Please follow these steps:
    Activate data collection:
    su - stor2rrd
cd /home/stor2rrd/stor2rrd
echo "export KEEP_OUT_FILES=1" > etc/.magic
echo "export DEBUG_FULL=2" >> etc/.magic
        
    Wait for 15 minutes, then send us data for analysis (via https://upload.stor2rrd.com):
    cd /home/stor2rrd/stor2rrd
tar cvf dump.tar data/YOUR_FILER_NAME/*perf* \ 
                 data/YOUR_FILER_NAME/*conf*  \
                 data/YOUR_FILER_NAME/*.dump 
gzip -9 dump.tar
rm etc/.magic

    dump.tar now should contain files like YOUR_FILER_NAME-*.cli.dump and your_filer_name-*.api.dump , that's what we need. 

    Don't worry, there will be no private content included.


  • ferpaschoal
    ferpaschoal
    Hi, any fix for this issue? I'm getting the same problem...

    Thanks!!
  • Pavel
    Pavel
    Hi,

    upgrade to this version  till there is released new one
     http://www.stor2rrd.com/download/stor2rrd-2.01-1.tar
    There will be a new version 2.10 release in 2 weeks where it will be also fixed.
    http://www.stor2rrd.com/note210.htm?1.8

  • Pavel
    Pavel
    just to finish that thread to help others ...


    use this code:
    http://www.stor2rrd.com/download-static/naperf.pl.gz
    Gunzip it and copy to /home/stor2rrd/stor2rrd/bin (755, stor2rrd owner)
    -rwxrwxr-x 1 stor2rrd stor2rrd 99075 Apr 12 13:50 bin/naperf.pl

    If your web browser gunzips it automatically then just rename it: mv naperf.pl.gz naperf.pl


    wait 10 minutes and then remove this:
    cd /home/stor2rrd/stor2rrd
    rm data/<storage name>/<storage name>*

    After that it will start working


Sign In or Register to comment.

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In Register

Categories

In this Discussion