Problem with hybernate on Synology

[marc_al]

Hello,

I have a small problem with SABNZBD on a Synology DS210J : it will always wake up the hard drive.
I have seen that is is related to the feature to scan a directory every XXX seconds to see if there is a nzb file to download.

I would like to disable this feature (I use the add-on for firefox so this feature is useless for me.
Can you please tell me how I can do it?
I have put the directory as Empty but it still does something (and wake up the hard drive) and I can not choose a greater value than one hour for the scan interval.
Perhaps the problem is also there on the RSS feed (as there is no option to remove it) but I have no idea.
It would be great to be able to choose a bigger interval (1 billion for example)

Also, I use the last firmware for the Synology.

Thank you
Marc

[shypike]

Possibly it's the logging.
To what level did you set it?

[marc_al]

Hello,

The login is set at the value +Info. I will put it to the value Error / Warning to see if it solves the problem.

The logging is set to go on a USB key so it shouldn't wake the hard drive (unless the synology makes something)

Thank you
Marc

[shypike]

The RSS feed will only write to disk if you have feeds enabled
for automatic scanning.

[marc_al]

Hello,

My problem is not solved.
I have made a test and when the program checks for nzb, it also checks the incomplete directory (that wakes up the hard drive).
So perhaps the problem is not with the nzb, but with the scan of the incmplete directory (even when the queue is empty) . Also, I use the version 0.5.2

EDIT : When the NZB scan directory is completely empty (instead of the USB key), every hour (something like one hour and 1 minute), sabnzbd does something (I have no idea what) but It is sabnzbd that does it (when sabnzbd is stopped (shut down) nothing happens on the NAS.

Marc

[shypike]

I'll have a look, but don't hold your breadth.
Not exactly an easy thing to find out.

[marc_al]

Hello,

Thank you for your time



EDIT : I have made some tests with a timer at 300 seconds and here is what I have seen
From what I can see, every 5 minutes  (the time I have put) it does something (the last time it was the maximum allowed so one hour). So perhaps you could simply enable a longer value for the scan interval (for example 100 000 or 1 000 000 seconds) (so SabNzbd would not wake up the computer again)?

Also, I don't understand why there is a scan on history1.db so many times, but it stops after a moment.
Here is the log
Jul  4 06:46:26 kernel: [/var/mail/root] opened by pid 11936 [u:(-ash), comm:(ash)]
Jul  4 06:48:55 kernel: [/var/packages/sab2/target/sab/admin/history1.db] opened by pid 11910 [u:(/var/packages/sab2/target/utils), comm:(python)]
Jul  4 06:53:55 kernel: [/var/packages/sab2/target/sab/admin/history1.db] opened by pid 11911 [u:(/var/packages/sab2/target/utils), comm:(python)]
Jul  4 06:56:59 kernel: [/var/packages/sab2/target/sab/admin/history1.db] opened by pid 11913 [u:(/var/packages/sab2/target/utils), comm:(python)]
Jul  4 07:01:59 kernel: [/var/packages/sab2/target/sab/admin/history1.db] opened by pid 11914 [u:(/var/packages/sab2/target/utils), comm:(python)]
Jul  4 07:06:59 kernel: [/var/packages/sab2/target/sab/admin/history1.db] opened by pid 11915 [u:(/var/packages/sab2/target/utils), comm:(python)]
Jul  4 07:33:00 kernel: [/etc/localtime] opened by pid 2712 [u:(/usr/syno/bin/scemd), comm:(scemd)]
Jul  4 07:36:57 kernel: [/var/packages/sab2/target/sab/admin/history1.db] opened by pid 11906 [u:(/var/packages/sab2/target/utils), comm:(python)]
Jul  4 07:41:57 kernel: [/volume1/Sys/SABnzbd/incomplete] opened by pid 11908 [u:(/var/packages/sab2/target/utils), comm:(python)]
Jul  4 07:46:57 kernel: [/volume1/Sys/SABnzbd/incomplete] opened by pid 11909 [u:(/var/packages/sab2/target/utils), comm:(python)]
Jul  4 07:51:57 kernel: [/volume1/Sys/SABnzbd/incomplete] opened by pid 11910 [u:(/var/packages/sab2/target/utils), comm:(python)]
Jul  4 07:56:57 kernel: [/volume1/Sys/SABnzbd/incomplete] opened by pid 11911 [u:(/var/packages/sab2/target/utils), comm:(python)]
Jul  4 08:01:57 kernel: [/volume1/Sys/SABnzbd/incomplete] opened by pid 11913 [u:(/var/packages/sab2/target/utils), comm:(python)]
Jul  4 08:06:57 kernel: [/volume1/Sys/SABnzbd/incomplete] opened by pid 11912 [u:(/var/packages/sab2/target/utils), comm:(python)]
Jul  4 08:11:57 kernel: [/volume1/Sys/SABnzbd/incomplete] opened by pid 11914 [u:(/var/packages/sab2/target/utils), comm:(python)]


Marc

[shypike]

Using the web UI is enough to cause disk access.
Lots of other disk access is caused by looking at the queue and history pages.
For the smpl and Plush skins disk access cannot be prevented.
You'll need to avoid all UI access to stop disk access, this may include third party utilities that communicate with SABnzbd.
Why exactly "incomplete" is touched (despite an empty queue), I don't know yet.
The "incomplete" folder is scanned to get the free disk space of the drive.
I think it's a problem in the NAS. Why does it power-up the drive just to find out the free space?

[marc_al]

Hello,

When I did these logs, I didn't access the UI (I let the NAS running doing nothing).
I suppose that the history1.db are done because of the extension for firefox (nzbdstatus) when I start firefox.
Is it necessary to scan the incomplete folder every time the program scans for nzb files? Perhaps you should only check for it when there is something in the queue. (or enable greater values for the scan )

For the NAS, I asked for them the question and they told me that even scaning a directory on A USB key will wake up the hard drive (I guess it is a problem with every Synology).

Thank you again for your time
Marc

[shypike]

The incomplete folder is not scanned when looking for NZB files.
It must be the FF extension.

[marc_al]

Hello,

I have made a test without the extension, but there is still something happening every hour (between one hour and one hour and 1 minute).
When I stop SabNZBD, nothing happens (no event every hour).

Do you have any idea that I could test?

Thank you
Marc

[shypike]

Is there anything in the scheduler?
Do you have active RSS feeds, newzbin readouts?

[marc_al]

Hello,

The sheduler is empty (I don't need it) and the RSS feed is emty too (I don't use it)

Marc

[marc_al]

Hello,

I have contacted Synology and I have made a lot of tests with them and I don't know where the problem lies :
For example, When SabNZBD is shut downed, the NAS can hibernate, but when SabNzbd is started and the Synology is disconnected from the network , it can hibernate, so it seems to be a problem with something on my network. I have no idea what, but I think that you can consider the problem as solved.
They have also told me that any access to any file / directory on the NAS (USB key) will wake up the hard drive

Thank you again for your time and sorry for the problem.
Marc

[sentenza34]

Hello,

The sheduler is empty (I don't need it) and the RSS feed is emty too (I don't use it)

Marc

Hi all,

I had the same problem and it was because of the RSS feed polling. I simply changed "RSS Checking Interval" in the General options tab to the maximum accepted value which is 14400 (24 hours) and my NAS doesn't wake up every hour anymore...

For me, I had traces every hour like:
Feb 10 14:23:36 kernel: [85989.320000] [/etc/localtime] opened by pid 3039 [u:(/var/packages/sab2/target/utils), comm:(python)]

So even with empty RSS feed SABnzdb still wakes up the NAS every hour by accessing /etc/localtime. There should be an option to completely deactivate RSS polling or not to poll when RSS feed is empty...

Thierry