Free SAN and storage performance and capacity monitoring

Hitachi storages installation

Back to storage installation home.

In case of usage of Virtual Appliance
  • Use local account lpar2rrd for hosting of STOR2RRD on the virtual appliance
  • Use /home/stor2rrd/stor2rrd as the product home
  • Use lpar2rrd account on storages as it is configured in /home/stor2rrd/stor2rrd/etc/stor2rrd.cfg (STORAGE_USER)
  • Replace lpar2rrd by stor2rrd user in all examples below.

The program uses Hitachi Storage Navigator Modular 2 CLI interface (HSNM2 CLI).
The CLI is included in the same CD from where you got the SNM GUI. If you dont have the CD you should ask HDS for the SNM CD (usually distributed as an ISO image). It has the CLI in it.

Allow TCP/IP communication from STOR2RRD host to all Hitachi storages on port 2000 (28355 when using secure communication)

Install HSNM2 CLI

  • HSNM2 CLI apparently needs libstdc++ package. Assure that it is installed.
    You have to install it even in case of usage Virtual Appliance 1.10 as it does not contain it (1.20 is ok).
    $ rpm -qa|grep libstdc++
      libstdc++-4.8.3-9.el7.x86_64
    
    Install it if it is not listed:
    # yum install libstdc++     # (RedHat, prp. Debian/Ubuntu )
    # apt-get install libstdc++ # (Debian/Ubuntu )
    
  • Get HSNM2 CLI package for your operating system and install it under root:
    # mkdir /usr/stonavm/
    # cd /usr/stonavm/
    # tar xf /tmp/HSNM2-2810-A-CLI-P01.tar
    # chown stor2rrd /usr/stonavm   # This must be owned by stor2rrd !!!!
    
  • Set environment in your actual shell (just copy&paste to the cmd line)
    LIBPATH=/usr/stonavm/lib:$LIBPATH
    SHLIB_PATH=/usr/stonavm/lib:$SHLIB_PATH
    LD_LIBRARY_PATH=/usr/stonavm/lib:$LD_LIBRARY_PATH
    STONAVM_HOME=/usr/stonavm
    STONAVM_ACT=on
    STONAVM_RSP_PASS=on
    PATH=$PATH:/usr/stonavm
    export  LIBPATH SHLIB_PATH LD_LIBRARY_PATH STONAVM_HOME STONAVM_ACT STONAVM_RSP_PASS PATH
    
  • Register the storage systems to be monitored (adjust IPs in below examples)
    • Automatically
      # auunitaddauto -ip 10.x.x.11 10.x.x.14
      
    • Manuallly
      # auunitadd -unit HUS110 -LAN -ctl0 10.*.*.12 -ctl1 10.*.*.13
      # auunitadd -unit HUS130 -LAN -ctl0 10.*.*.14 -ctl1 10.*.*.15
      
    • Using secure communication (port 28355 on storage):
      # auunitadd -unit HUS110 -LAN -ctl0 10.*.*.12 -ctl1 10.*.*.13 -communicationtyp secure
      
  • Create user stor2rrd (lpar2rrd when you use Appliance) on the storage (storage admin with read role)
  • Register user access for the storage
    # auaccountenv -set -uid stor2rrd -authentication 
    
  • Test connectivity
    # auunitref
    # auunitinfo -unit HUS110
    

STOR2RRD storage configuration

  • Configure storages in etc/storage-list.cfg
    Under stor2rrd account create config entry. Uncomment (remove the hash) example line and adjust it:
    $ vi /home/stor2rrd/stor2rrd/etc/storage-list.cfg
    
    #
    # Hitachi HUS
    #
    # Storage alias:HUS:VOLUME_AGG_DATA_LIM:VOLUME_AGG_IO_LIM:SAMPLE_RATE_MINS
    #
    #HUS110:HUS:256:10:5
    #HUS130:HUS:
    #AMS2000-alias:HUS:
    HUS110:HUS:
    
  • Assure you have enough of disk space on the filesystem where is STOR2RRD installed
    Roughly you might count 2 - 30 GB per a storage (it depends on number of volumes, 30GB for 5000 volumes)
    $ df -g /home   # AIX
    $ df -h /home   # Linux
    
  • Test the storage connectivity:
    $ cd /home/stor2rrd/stor2rrd
    $ ./bin/config_check.sh 
      =========================
      STORAGE: HUS110: HUS 
      =========================
      /usr/stonavm/auunitinfo -unit HUS110
      connection ok
    
  • Schedule to run storage agent from stor2rrd crontab (lpar2rrd on Virtual Appliance)
    $ crontab -e
    
    # Hitachi HUS && AMS 2000 storage agent 
    0,5,10,15,20,25,30,35,40,45,50,55 * * * * /home/stor2rrd/stor2rrd/load_husperf.sh > /home/stor2rrd/stor2rrd/load_husperf.out
    
    Assure there is already an entry with the GUI creation running once an hour in crontab
    $ crontab -e
    
    # STOR2RRD GUI
    0 * * * * /home/stor2rrd/stor2rrd/load.sh > /home/stor2rrd/stor2rrd/load.out 2>&1
    
  • Let run the storage agent for 15 - 20 minutes to get data, then:
    $ cd /home/stor2rrd/stor2rrd
    $ ./load.sh
    
  • Go to the web GUI: http://<your web server>/stor2rrd/
    Use Ctrl-F5 to refresh the web browser cache.

Allow TCP/IP communication from STOR2RRD host to all Hitachi storages on port 1099,31001 and 51100

The program uses two Hitachi APIs. You have to install both of them.
  • Command Control Interface (CCI)
  • Hitachi Export tool

Installation of Hitachi CCI

Obtain CCI installation package from your Hitachi representatives.
  1. Insert the installation media into the proper I/O device. Move to the desired directory for CCI (under root)
    # mkdir /opt
    # cpio -idmu < /dev/XXXX    # where XXXX = I/O device with install media
    # ln -s /opt/HORCM /HORCM
    
  2. Execute the CCI installation command:
    # /HORCM/horcminstall.sh
    
  3. Verify installation of the proper version using the raidqry command:
    # raidqry -h
      Model: RAID-Manager/HP-UX
      Ver&Rev: 01-29-03/05
      Usage: raidqry [options]
    
  4. Assure that everything is executable and writeable by stor2rrd user (it will be started under stor2rrd).
    This is a must! Under root identification execute this:
    # touch /HORCM/etc/USE_OLD_IOCT 
    # chown -R stor2rrd /HORCM/log* /HORCM/.uds /HORCM/usr
    # chmod 755 /HORCM/usr/bin /HORCM/usr/bin/* /HORCM/log /HORCM/etc/horcmgr /HORCM/etc/*conf /HORCM/.uds/ /HORCM/.uds/*
    

Configuration of CCI

  • CCI communication with storage can be done either via LAN (it is something described below) or via command device (SAN attached volume from the storage). When you have many storages in place 40+ then use rather command device as LAN communication might not be reliable enough
  • Each storage must have its own config file /etc/horcmXX.conf
  • Check if local ports 10001 and 10002 are not used (nothing is listening there)
    # netstat -an|grep -i listen| egrep "11001|11002"
    
  • storage with IP 192.168.1.1, conf file /etc/horcm1.conf will use local port 11001 (UDP)
    Use storage node IP. SVP IP must be used in etc/storage-list.cfg further.
    # vi /etc/horcm1.conf
    
    HORCM_MON
    # ip_address service poll(10ms) timeout(10ms)
    localhost    11001   1000       3000
    HORCM_CMD
    # dev_name dev_name dev_name
    \\.\IPCMD-192.168.1.1-31001
    
  • storage with IP 192.168.1.10, conf file /etc/horcm2.conf
    change localhost port to 11002 (from 11001 which is used above)
    # vi /etc/horcm2.conf
    
    HORCM_MON
    # ip_address service poll(10ms) timeout(10ms)
    localhost    11002   1000       3000
    HORCM_CMD
    # dev_name dev_name dev_name
    \\.\IPCMD-192.168.1.10-31001
    
  • Start it under stor2rrd account (definitelly not under root!). Use lpar2rrd account on the Virtual Appliance
    This starts HORM instance 1 (/etc/horcm1.conf)
    # su - stor2rrd
    $ /HORCM/usr/bin/horcmstart.sh 1
    
  • Start HORM instance 1 & 2 (/etc/horcm1.conf & /etc/horcm2.conf)
    # su - stor2rrd
    $ /HORCM/usr/bin/horcmstart.sh 1 2
    
  • Check if they are running
    $ ps -ef | grep horcm
      stor2rrd 19660912 1 0 Feb 26 - 0:03 horcmd_02
      stor2rrd 27590770 1 0 Feb 26 - 0:09 horcmd_01
    
  • Place it into operating system start/stop scripts
    # su - stor2rrd -c "/HORCM/usr/bin/horcmstart.sh 1 2"
    # su - stor2rrd -c "/HORCM/usr/bin/horcmshutdown.sh 1 2"
    

Installation of Hitachi Export Tool

    It is typically located on a CD that comes packaged with the Service Processor on the HDS USP Array. The Export Tool can also be obtained by contacting HDS support. (CD location: /ToolsPack/ExportTool)
    Hitachi produces a new Export Tool for each release of the firmware. So unless you make sure all of the storages are running on the same firmware version then you will need to obtain the appropriate version of the Export Tool to meet the firmware version you are running at the site.

    Find our firmware release of your storage (83-01-28/00) identified by /etc/horcm1.conf (-I1):
    Under stor2rrd account!
    # su - stor2rrd
    $ raidcom -login $login $paswd -I1
    $ raidqry -l -I1 
      No  Group    Hostname     HORCM_ver   Uid   Serial#   Micro_ver     Cache(MB)
       1    ---   localhost   01-35-03-08     0   471234    83-01-28/00      320000
    $ raidcom -logout -I1
    
    Install each version of the Export Tool into separate directory named as firmware of your storage (just 6 numbers like in this example firmware 83-01-28) under root user:
    # mkdir /opt/hds
    # mkdir /opt/hds/83-01-28
    # cd /opt/hds/83-01-28
    # tar xvf export-tool.tar
    # chown -R stor2rrd /opt/hds
    
    Directory /opt/hds is optional, it is configurable in /home/stor2rrd/stor2rrd/etc/sto2rrd.cfg : VSP_CLIDIR=/opt/hds
    The HDS Performance Monitor License must exist for each array and monitoring must be enabled.
    Storage configuration example

STOR2RRD storage configuration

  • Configure Export Tool installation directory in /home/stor2rrd/stor2rrd/etc/stor2rrd.cfg
     VSP_CLIDIR=/opt/hds
    
  • Create user stor2rrd on the storage, read only access
  • Encrypt password used for storage access (stor2rrd)
    $ perl ./bin/spasswd.pl
      Encode password for storage authentication:
      -------------------------------------------
      Enter password:
      Re-enter password:
      Copy the following string to the password field of the corresponding line in etc/storage-list.cfg:
        IT5gYGAK
    
  • Configure storages in etc/storage-list.cfg
    Under stor2rrd account create config entry. Uncomment (remove the hash) example line and adjust it.
    Use storage SVP IP here (CCI above uses storage node IP).
    $ vi /home/stor2rrd/stor2rrd/etc/storage-list.cfg
    
    #
    # Hitachi VSP-G
    #
    # Storage Alias:VSPG:IP address/hostname:storage user:password:/etc/hormcm.conf:VOLUME_AGG_DATA_LIM:VOLUME_AGG_IO_LIM:SAMPLE_RATE_MINS
    # to encrypt password use: perl ./bin/spasswd.pl
    VSPG-600:VSPG:192.168.1.1:stor2rrd:IT5gYGAK:/etc/hormcm1.conf:1024:10:0
    VSPG-400:VSPG:192.168.1.10:stor2rrd:IT5gYGAK:/etc/hormcm2.conf
    
  • Assure you have enough of disk space on the filesystem where is STOR2RRD installed
    Roughly you might count 2 - 30 GB per a storage (it depends on number of volumes, 30GB for 5000 volumes)
    $ df -g /home   # AIX
    $ df -h /home   # Linux
    
  • Test the storage connectivity under stor2rrd user:
    $ cd /home/stor2rrd/stor2rrd
    $ ./bin/config_check.sh 
      =========================
      STORAGE: VSPG-600 : VSPG
      =========================
      connection ok
    
  • Schedule to run storage agent from stor2rrd crontab (lpar2rrd on Virtual Appliance)
    $ crontab -e
    
    # Hitachi VSP-G
    0,5,10,15,20,25,30,35,40,45,50,55 * * * * /home/stor2rrd/stor2rrd/load_vspgperf.sh > /home/stor2rrd/stor2rrd/load_vspgperf.out
    
    Assure there is already an entry with the GUI creation running once an hour in crontab
    $ crontab -e
    
    # STOR2RRD GUI
    0 * * * * /home/stor2rrd/stor2rrd/load.sh > /home/stor2rrd/stor2rrd/load.out 2>&1
    
  • Let run the storage agent for 15 - 20 minutes to get data, then:
    $ cd /home/stor2rrd/stor2rrd
    $ ./load.sh
    
  • Go to the web GUI: http://<your web server>/stor2rrd/
    Use Ctrl-F5 to refresh the web browser cache.

 

Links

Download

Others

Contact