Unfortunately no one can be told what fun_plug is - you have to see it for yourself.
You are not logged in.
Hi,
i am using the DNS-323 since several years.
Running 1.09 on it with ffp installed and the HDD are hibernating fine.
Now a bought a second one.
Running 1.09 as well with ffp.
This device the HDD are not hibernating at all.
I was even renaming fun_plug, so that no ffp is starting. The HDD keep running.
Only difference between the 2 DNS are the HDD. Old one has 2 WD 1 TB in RAID-1. New one has 2 WD 2 TB in Raid-1.
itunes, UPNP, etc. disabled.
Any idea, what i could check for?
Offline
No answers yet? I think i've seen some other posts about this problem and i'm going through the same problem as well. Start by disconnecting the network cable to make sure the problem isn't coming from outside the dns. I did that but the HDD is still waking up every 3-5 minutes.
Do you have any other apps running in the background through ffp? Like transmission, twonky, any other? You may get more responses if you give more details. I'm going to create a new post for my problem if it helps.
Offline
Here's a discussion on the same issue:
http://dns323.kood.org/forum/viewtopic.php?id=6175
For my part, my setup is as follows :
-i'm running ffp 0.5 off a usb stick
-firmware 1.09 (2x 1TB drives; individual disks)
-installed transmission
-have 5 or 6 rsynch jobs scheduled to run once a month
-no bittorrent add-on; no twonky installed either
I just turned off transmission, and all server services (ftp, unpnp,itunes, dhcp)- i'm going to disconnect the network cable and see if the baby sleeps(hibernates) well.
Offline
I have 3 units with an assortment of WD and Seagate drives, 1TB, 1.tB and 2TB, all non-RAID. Running FFP 0.5 where anything that logs and writes to the drive (other than saved torrent downloads) is setup to use the USB stick. Running F/W 1.09 and the drives hibernate without any problems.
Use 'ps' and 'top" to see what running, check to see what writes to the drives (logs primarily) and teh drives should power down (assuming you set that in the config as well). Wouldn't hurt to reset to defaults and manually reconfigure too to ensure a clean configuration file.
Offline
Well after all that my unit is still waking up every 3-5 minutes. I'll try looking into logs. There's obviously a running process that's making it do this... Other suggestions?
I lost a hard drive recently after only 4 years of use. Perhaps i'm being paranoid but i suspect it was caused by the drive constantly waking up- i've never left any torrents seeding.
Offline
esebag wrote:
Well after all that my unit is still waking up every 3-5 minutes.
All what? What did 'ps -ef' output?
Offline
PID USER COMMAND
1 root init
2 root [ksoftirqd/0]
3 root [events/0]
4 root [khelper]
5 root [kthread]
11 root [kblockd/0]
14 root [khubd]
49 root [pdflush]
50 root [pdflush]
52 root [aio/0]
51 root [kswapd0]
190 root [scsi_eh_0]
191 root [scsi_eh_1]
192 root [scsi_eh_2]
193 root [scsi_eh_3]
204 root [mtdblockd]
218 root [kcryptd/0]
219 root [kmirrord/0]
232 root [loop0]
1322 root smbd -D
1324 root smbd -D
1325 root nmbd -D
1387 root xmldb -n config
1719 root chkbutton
1745 root /web/webs
1763 root fancontrol 0
1771 root op_server 3 3 3
1787 root -sh
1894 root atd
1924 root lpd Waiting
1958 root mserver
1959 root apkg
1993 root crond
2050 root [scsi_eh_4]
2056 root [usb-storage]
2095 root smbd -D
2163 root /ffp/sbin/sshd
2355 root sshd: root@pts/0
2357 root -sh
2505 root ps -ef
Offline
It looks pretty clean. If you are using a USB stick, then I would assume you don't need the print spooler. So you can get rid of the lpr/lpd processes.
What indicator are you using to assess that the HDDs are awake/asleep? Have you been monitoring the system log with 'dmesg'?
Have you checked/set the LLC value on the HDDs?
Last edited by FunFiler (2011-10-09 00:50:27)
Offline
Thanks- kinda of new to linux, how do i permanently kill the lpr/lpd process?
The indicators i was using were strictly audio/visual; i hear the HDD (#1- on the right) powering up and observe the dimmed-pwr light go bright. Thanks for the dmesg command, results are what i suspected- HD0 (i assume is the first HDD with ffp installed) is alternating between stand by and awake.
What is the LLC value exactly?
I started a new thread concerning my problem here:
http://dns323.kood.org/forum/viewtopic.php?id=6815
Tnx.
Offline
Ok, i found the kill command- easy enough...
Offline
LLC = Load_Cycle_Count
This is one of the S.M.A.R.T. values that can be queried and may affect the drive as discussed in another thread.
As part of my fun_plug_local script, along with other things, I kill the lpd process
# Get rid of the lpd process since we don't use a print spooler echo "`date` Killing the Print Spooler" /ffp/bin/killall lpd
Last edited by FunFiler (2011-10-09 06:53:31)
Offline
Can mserver or apkg be causing it? (I don't know - just mentioning possibilities)
One approach to solving this is to check which files on the hard disk have been accessed or modified in the last 10 minutes or so.
find /mnt/HD_a2 -xdev -mmin -10
Will list all the files on your disk that have been modified 10 minutes or less ago.
find /mnt/HD_a2 -xdev -amin -10
Does the same except will list the files that have been *accessed* within the last 10 minutes. This might be more useful as it would include the files that are just getting read.
If you see a file getting changed or accessed when you expect the disk to be asleep, it can give you a clue to the offending program.
Check your crontab to make sure it isn't doing anything you do not expect.
crontab -l
Might try looking at the output of
lsof
Which is a list of the currently open files. There will be a lot of false positives here, since a process may have a file open on your hard drive but may not be actively reading or writing it. If you have ffp on a usb stick, then there will be far fewer false positives, since you can immediately reject anything on the usb stick. You might get lucky and spot a file that is open on the hard disk that you do not expect.
Other than that, I'd kill off processes one by one until the disk starts sleeping as you expect. Then you know what needs to be reconfigured to avoid accessing the disk.
I would start by killing mserver and apkg.
Shut down Samba with smb stop.
I don't know what xmldb does, but try killing that.
Try killing crond and atd and perhaps webs.
There's not much left to kill at this point. Be sure to wait after each kill to see if the disk sleeps. You can always reboot to get back to normal. This process killing just for investigation.
Check your ssh and sshd settings to be sure some sort of logging is not active. Unlikely.
There is an outside chance that your drive's S.M.A.R.T. is configured to run some sort of self-test. It might be waiting for a few minutes of inactivity and then start a surface scan or something. Unlikely, but you might try:
smartctl --device=marvell --all /dev/sda
That is about all I can think of.
Offline
I have mserver and apkg running and have no issues.
/ # ps -ef
PID USER COMMAND
1 root init
2 root [ksoftirqd/0]
3 root [events/0]
4 root [khelper]
5 root [kthread]
11 root [kblockd/0]
14 root [khubd]
49 root [pdflush]
52 root [aio/0]
189 root [scsi_eh_0]
51 root [kswapd0]
190 root [scsi_eh_1]
200 root [mtdblockd]
217 root [kcryptd/0]
218 root [kmirrord/0]
231 root [loop0]
1274 root xmldb -n config
1330 root [kjournald]
1517 root [kjournald]
1592 root chkbutton
1652 root fancontrol 1
1658 root op_server 3 3 3
1673 root -sh
1728 root atd
1779 root mserver
1780 root apkg
1882 root [scsi_eh_2]
1883 root [usb-storage]
2002 root /usr/sbin/samba/smbd -D
2006 root /usr/sbin/samba/smbd -D
2007 root /usr/sbin/samba/nmbd -D
2038 root /ffp/sbin/telnetd -l /ffp/bin/sh
2056 root /ffp/bin/rsync --daemon --config=/ffp/etc/rsyncd.conf
2122 root /web/webs
2728 root crond
14147 root [pdflush]
16651 root /ffp/sbin/lighttpd -f /ffp/etc/lighttpd.conf
16522 root /ffp/bin/sh
16523 root ps -ef
/ #
Offline
I seem to have resolved my problem, i got rid of the lpr/lpd processes as mentioned above and uninstalled/reinstalled Transmission- the unit now sleeps on a long term basis. Tnx for the tips but sorry if i wasnt able to identify or eliminate the specific process that was causing the problem.
Offline
My disks sleep fine with lpr/lpd running. I use the DNS for printing, so that's important to me.
It is great you got it working, but it seems odd that you had to kill off lpr/lpd. Not a problem if you don't run a printer.
It is possible that your old install of transmission was causing it.
Offline