Hi,
I have 2 devices with linux-based firmware. They support SSH and basic Unix commands (uname, ps, df, etc.), so they can be monitored under /Server/SSH/Linux. There are hundreds of mount points on them, which are just bind mounts. Most of them are read-only and 100% full. If I'll model this device with default plugins I'll get a lot of unneeded components under Components/File Systems.
I've disabled "df" modeler plugin, modeled the device and then added file system manually using mount point as ID.
Now Zenoss shows that Total = Used = Free Bytes = 0, Util% = "Unknown", Block Utilization graph fails to render, Inode Utilization graph is rendered just fine.
Is disk space monitoring possible without modeling a file system with default plugin? Can I manually set Total Bytes on manually added file system? Or maybe I've got a bug?
Thanks in advance for any suggestions!