1 Reply Latest reply: Apr 5, 2012 5:53 AM by sgrant RSS

Netapp Management Console 3.0: misleading no_atime_update recommendation and network graph problem

cambiasso
Currently Being Moderated

Hello,

I recently upgraded Operations manager to 4.0 and installed Netapp Management Console 3.0.The new features are very useful, especially client statistics, I'd like to be able to collect them at the volume level too.

I have noticed only 2 problems:

 

1 - in the NAS Issues health check section of Diagnosis tool I get this recommendation

Enable the no_atime_update option for volumes. This is recommended if you can tolerate losing a few seconds of access time information when a storage system crashes.

 

As far as I know if you enable no_atime_update the access time information is NEVER updated, while the recommendation seems to mean that it isn't updated only in case of storage crash. Do you have any idea of the rationale behind it?

 

2 - network throughput graphs are no longer updated, because the corresponding counters (load_inbound_mbps, load_outbound_mbps...) are missing. In the Data collection configuration wizard I can't find these counters. Is there a workaround other than create my custom graphs with the available counters?

 

Thanks.

 

 

  • Re: Netapp Management Console 3.0: misleading no_atime_update recommendation and network graph problem
    sgrant
    Currently Being Moderated

    Had the same problem as 2 above, however from the manual:

     

    Top Storage Systems by Network Throughput

    This panel displays the load_total_mbps counter, which is calculated using the following formula:

    load_total_mbps = load_inbound_mbps + load_outbound_mbps

    load_inbound_mbps = system:system:net_data_recv + fcp:fcp:fcp_write_data +

    iscsi:iscsi:iscsi_write_data

    load_outbound_mbps = system:system:net_data_sent + fcp:fcp:fcp_read_data +

    iscsi:iscsi:iscsi_read_data

     

    We had stopped collecting FCP stats since it was not in use, therefore the "load" counters could not be calculated. Once we started collecting these again, the Network Throughput graph started to update again.

More Like This

  • Retrieving data ...