11 Aug 2021 03:38 PM - last edited on 20 Oct 2022 10:58 AM by MaciejNeumann
All of our Linux machines have both physicals disk and NFS disk. We need to monitor our NFS disks more closely as those are the ones that run out of space more frequently and have autogrow that we must ensure is working reliably.
To monitor these disks we're using the Dynatrace disk monitoring features which do provide more options and details. The problem though is that a mount point across multiple servers could be connected to the same source, so it would be better if we could monitor based on the source name. OR at the least have the sourcename information to help with our analysis of disk usage.
We had such an option in SolarWinds because it captured both the Source and Destination for each disk on linux. If Dynatrace us running the linux 'DF' command to obtain disk name and usage then the source is also returned, why couldn't DT capture that data as well.
Even flagging a disk as an NFS vs Physical would be helpful.
Linux Machine 1:
Source = host:/xxx/yyy/mylocation
Destination = /nfsmnt/xxxxxx
Linux Machine 2:
Source = host:/xxx/yyy/mylocation
Destination = /nfsmnt/yyyyyyy
Has anyone else had this problem or have this need? What creative solutions have others maybe come up with?
11 Aug 2021 09:32 PM
Can you share with us a sample of your Disks on a host? Is the settings set to monitor the NTFS drives?
04 May 2022 04:22 PM
Hi Chad, I am curious about this also. How is Dynatrace scanning the disks? Is it something like running a df command? The setting you reference above - can you elaborate what it does? It attempts to alert based on the nfs source vs the individual mounts points?
Thanks,
Jordan