UNSOLVED

bonnu06

updated

13 years ago

B

bonnu06

2 Intern

•

258 Posts

0

1313

September 9th, 2013 05:00

Regarding HBA monitoring tool on a server or SAn fabric

Hello,

Since past 2 weeks we have 2003 R2 server that deops of the paths, i mean one hba goes down automatically and comes back on it's it own. I know it has a bad HBA for now, in future we want to have some monitoring software to monitor if any hba goes down that would generate an email and alert us. All our server windows 2003 and 2008 r2. We have 9506 core switch as well. Can someone suggest me a good monitoring software on the server side and fabric side if any ?

  • bonnu06

    2 Intern

    •

    258 Posts

    453

    0

    Posted September 9th, 2013 06:00

    We are still at system center 2007... i believe it won't work...

  • dynamox

    11 Legend

    •

    20419 Posts

    •

    87439 Points

    453

    0

    Posted September 9th, 2013 06:00

    PowerPath will generate alerts when it loses paths, if your systems already monitor syslog or windows event viewer ..you can instruct them to look for specific errors and generate alerts.

  • dynamox

    11 Legend

    •

    20419 Posts

    •

    87439 Points

    453

    0

    Posted September 9th, 2013 06:00

    i don't know what steps you are looking for ? You need to talk to your system admin, they need to figure out how to trap PowerPath/HBA specific errors (and trust me OS spits out tons of errors whenever it loses paths) and then configure their tools to send alarms/emails or whatever they do. For example we use System Center 2012 to monitor Windows Servers, whenever PowerPath related messages are detected in the log, it automatically connects to our ticketing system and generates a ticket.

  • bonnu06

    2 Intern

    •

    258 Posts

    453

    0

    Posted September 9th, 2013 06:00

    Hello dynamox thanks for the reply... i don't know how to set it up.. can you guide me throgught the steps.. if you have 5 minuts?

  • dynamox

    11 Legend

    •

    20419 Posts

    •

    87439 Points

    453

    0

    Posted September 9th, 2013 07:00

    we had it working with 2007 too, we just recently migrated to 2012.

  • umichklewis

    4 Apprentice

    •

    1190 Posts

    453

    0

    Posted September 9th, 2013 12:00

    On the fabric side, most any modern FC switch can generate alerts (SNMP traps or email) on port status.  Our Cisco switches used to send SNMP traps to my collection station.  When we installed ProSphere, it collects the port up/down errors from DCNM and sends email to the storage team and server team.  If you can't get PowerPath monitoring working, take a look at the native SNMP traps on your switch - Cisco, Brocade and QLogic will all generates traps/and or email.

    Thanks!

    Karl

  • dynamox

    11 Legend

    •

    20419 Posts

    •

    87439 Points

    376

    0

    Posted September 9th, 2013 12:00

    issues that we had with Karl's approach was that alerts went to the storage team only and we had no clue if they were legitimate issues with HBAs/cables/ports or the box was being rebooted.

  • bonnu06

    2 Intern

    •

    258 Posts

    453

    0

    Posted September 9th, 2013 12:00

    Hello Karl,

    Thanks for the reply.. yes we do have Cisco running on DCNM.. do you have any setup document that explains setting up the alerts..

  • umichklewis

    4 Apprentice

    •

    1190 Posts

    288

    0

    Posted September 9th, 2013 12:00

    http://www.cisco.com/en/US/products/ps9369/products_installation_and_configuration_guides_list.html

    You need to find the version that matches the version of DCNM you're on - 5.1 or 6.1.  Thanks!

  • dynamox

    11 Legend

    •

    20419 Posts

    •

    87439 Points

    288

    0

    Posted September 9th, 2013 12:00

    another option for all PowerPath environment, it works for Linux and Windows, not sure about other platforms.

    What is PowerPath Viewer?
    EMC SW: PowerPath 5.3
    EMC SW: PowerPath for Windows

    EMC SW: PowerPath Viewer

    PowerPath Viewer is a tool that enables users to monitor multiple PowerPath hosts from a single monitoring station. Versions 1.0 supports PowerPath for Windows version 5.3 and above.

    Note the following features of the PowerPath Viewer:

    • The viewer console can monitor up to 500 PowerPath hosts. Each discovered host monitors PowerPath driver events and reports to the viewer console.
    • The viewer console consolidates the information and presents it in a GUI.
    • The user can separate hosts into groups and see a snapshot of the status of all hosts. The user can also get information on paths and bus alerts.
    • The viewer can be configured to send email alerts when a error is detected.
    • Port Number:TCP 9083 is where the management component will listen for connection requests from the PowerPath viewer console. It is important that this port is open in the firewall.
    • Heartbeat Interval: The heartbeat interval is set to 10 minutes by default, but can be set to any value between 1 and 60 minutes.
    • Single Path Gatekeepers: The management component will not flag gatekeeper (GK) devices with a single path if the "Single Path Gatekeeper OK" bit is set. This is for users who intentionally configure only one path to their GK devices, and do not want to see these alerts. This is the default.
    • These 3 values (Port Number, Heartbeat Interval and Single Path Gatekeepers) are kept in the Registry and are read at startup (HKEY_LOCAL_MACHINE\Software\EMC\PowerPath\ManagementComponent\). If any of the values are modified you must restart the Windows PowerPath service for the change to take effect.
    • The configuration data for PowerPath viewer is located in the "users\(username)\emc\powerpath" folder. This file contains a list of hosts and host groups and email settings.
    • CSV files can be created to set up multiple PowerPath Viewer hosts.