SMAX 23.4 low performance due to low NFS disk issue.

in SMAX env. Release Version: 23.4, I have low performance issues so after troubleshooting we found that NFS disk has low performance as the screenshot.

Although we have 4 workers with sufficient memory and disk space (ssd), knowing that we have 4 deployed tenants and one of them is itegrating with UCMDB, another one is integrated with OBM which send a lot of events daily.

what are the procedures to enhance the performance of SMAX and NFS disk? 

  • 0  

    Hi Abo,

    There are certain things you can check around NFS performance. A good starting point is checking the system sizing guides against what you have deployed. 

    docs.microfocus.com/.../SizingForSMA

    Depending on the size of your SMAX system the above link give guidance for the size of the NFS server.

    When you say UCMDB, is this using Native SACM? Is it with UCMDB Containerised or Classic? If Containerised this will have an impact on the number of workers and also the sizing. UCMDB would also expect to have its own NFS and not share the SMAX NFS if using conatinerised UCMDB.

    To help reduce the amount of Disk space used you can also perform a clean up of the Logs held on the NFS Server.

    https://docs.microfocus.com/doc/SMAX/23.4/DiskFullIssue

    https://docs.microfocus.com/doc/SMAX/23.4/archiveDeleteLogs

    Especially if lots of data is being written due to the Integrations you have running.

    Another link worth checking is this:

    https://docs.microfocus.com/doc/SMAX/23.4/ManageStorageSuite

    This provides guidance around Managing the suite Storage and check performance.

    If after doing all that you are still seeing performance issues, might be worth raising a support ticket.

    Thanks,

    Dave

  • 0 in reply to   

    Hello Dave,

    uCMDB is calssic and integrated through CMS Gateway. Sizing recommendation from OT are considered, even more resources are allocated for VMs, tenants are not fully loaded yet, NFS machine are not fully uilized, even all k8s cluster is healthy regarding resources as the screenshot clarifies.

    what can decrease NFS performance to fix ?

  • 0   in reply to 

    Hi Abo,

    With using both the UCMDB and OBM Integrations there is potential that the NFS is getting a certain amount of load. Like you mentioned the OBM is producing a lot of events so this will generate a certain amount of load. I'd still expect this to be handled by the infrastructure that you have provisioned. I suppose it depends on how many events are being produced over a certain duration of time.

    Based upon the fact that there could be external factors at play as well as the software when it comes to NFS performance it might be worth raising a Support ticket in this case. That way you can troubleshoot the problem and make sure everything is working based upon what the logs are reporting along with any tests that need running.

    Thanks,

    Dave