Hard Disk Monitoring

First of all, thank you very much for the useful feature of providing the status of NAS drives in the Pro version.

After finally getting the thing running on my Synology 4-bay NAS (had to fight a corrupted schedule database, ran into the solution by accident), I immediately came up with an idea to schedule reading the report file. As my NAS is used only for media files such as movies or TV series, the drives are asleep for most of the time and are usually active only during a rather short period of time during the weekend. The report file is scheduled to be created during such time to avoid unnecessary spin-ups as much as possible.

Currently I need to copy the report file on my desktop PC at the end the script so the NAS can be idling most of the time. If I could schedule reading the NAS drive status file(s), I could leave it on the NAS and read it right after it's been created.

Do you think this kind of thing could be easily added?

Thanks for your message and the information, good to hear you had success in configuring NAS monitoring.

Yes, I completely understand the situation about sleeping disks.

Generally it is possible to make scheduling as you wrote: to access disk status of NAS drives (for example) only between 12:00 and 23:00 on specific days or so, but to be honest, not sure if would be useful.

Generally NAS devices should have some kind of cache: if the file we read (the status source file in this case) not changed/not updated, then the NAS should provide it from its cache - without waking the drives for this purpose.
As you wrote, you already configured creation of the status sources in a schedule to prevent wake up when the drives are not in use.

So did you experience that reading the status source file(s) wake up the drive(s), even if the contents of the file is not changed, not updated?

Thanks!

I didn't test the caching yet, but will do that later. What I'm wondering about caching is that wouldn't the file be purged a long time ago from the cache if the report file is produced and I watch movies or TV series after that, which means that the NAS is reading from the HDDs and caching more recent data?

Yes, this is exactly how cache should work - and exactly this is what we want: when the NAS is active, hard disks are spinning due to read/write operations (so their status may be changed, updated) then we exactly want to have new, fresh reports to monitor the updated status, possible changes.
Then yes, we expect that status sources should be read and updated (which is not a problem - as the hard disk is actually spinning).

hdsentinel wrote:Yes, this is exactly how cache should work - and exactly this is what we want: when the NAS is active, hard disks are spinning due to read/write operations (so their status may be changed, updated) then we exactly want to have new, fresh reports to monitor the updated status, possible changes.
Then yes, we expect that status sources should be read and updated (which is not a problem - as the hard disk is actually spinning).

But the problem is that I'm reading the status weekly on a set date and time, regardless of the disks spinning or not. If there was a way in the Synology task scheduler to set it to launch the script whenever the disks are woken up, I would use it.

So the problem is that I'm reading the status and writing the report file at a fixed point and there may be so much I/O traffic after that that the cache does not contain the file anymore when HD Sentinel periodically requests it. Of course, I could get around the issue by copying it over the network to my desktop PC which is usually on 24/7.

I finally did some tests and it seems that the HDDs do not spin down if I make HDS read the report file off the NAS. The NAS log file shows a CIFS connection every 30 minutes or so as I have set the poll interval to 30 minutes in HDS.

Yes, I completely expected that there is connection (based on the configured detection frequency in Hard Disk Sentinel).

I just expected that if the NAS "knows" that
- the drives are sleeping
- the file requested (the status source) is small enough to fit in its cache (as it may be some KBytes only)
- the file is not changed since last polling (as it is not updated, if I understand correctly you configured with cron to do not update in inactive periods)

then it provides the cached version without waking up the drives.

But you gave good idea about scheduling - just then it would be good to make it flexible to cover not only NAS devices, but also internal/external drives too.

I'm still collecting experiences, thoughts, ideas - and in a future version the scheduling option will be surely added - probably for all possible drives.
Thanks for the tip!