Unfortunately no one can be told what fun_plug is - you have to see it for yourself.
You are not logged in.
Pages: 1 2
Hi all,
just found my way here, checked the list of known issues and will disable the daylight savings thing, if that still is an issue...
I'm having problems in that the unit seems to be active a lot, while I'm not using it at all. I have a 750Gb and a 1Tb disk installed, not running RAID or anything. It's on the home network but nobody is accessing it other than me. As far as I know. Checked the router log/status and the dns-323 status, nothing odd there.
The activity in question is harddisk activity, i.e. I hear the disk and also the light indicator is blinking all the time.
Any idea what might be causing this? It almost sounds like when you're running a virus scan on your harddisks.
Best regards,
Joris
Sweden
Offline
Well, you can check if the problem exists even if the network is unplugged.
You can also check the list of running processes using "ps -ea"
Finally check which processes are accessing the disks using "lsof | grep HD"
Maybe this will shed some light in your issue.
Offline
ERmmm... can't say I'm very good at this stuff... are those commands I can type at the command prompt (anywhere), or how do I do that?
And yes I did check with network cable unplugged, which seemed to not make a difference :S
Thanks for the reply, hope you (or somebody else) can guide me somewhat more specific as to how to perform those checks.
Right now the unit is actually sitting quiet though
Offline
You can also try disabling all servers (iTunes, uPnP, FTP, etc).
Offline
It could be the UPnP Server indexing files, if you don't use it try to deactivate it.
The commands posted by silversurfer are meant for running them through telnet/ssh on the device... the wiki is your friend
Offline
Oh, I'm sorry. I thought that you have some sort of access to the device via telnet/ssh like most people here do. With that you can log into the unit and issue commands like on any other linux machine.
We are using funplug (thank you fonz :-) ) to establish this access and install other services on our systems. I will not encourage you to use it too. You already need some knowledge about linux to use funplug well. Also you will need a reason to use it. The unit runs fine without funplug. We just have certain demands that the "out of the box" system could not fulfill.
If you are interested in funplug nevertheless then please look here -> http://dns323.kood.org/howto:ffp
Don't blame me if you break your system. ;-)
Last edited by silversurfer (2008-07-24 02:00:56)
Offline
Hehehe thanks for the advice. I'll check those uPnP and other things. Currently the unit serves one single purpose only, that of being a NAS; storage for movies and music so that we can put it there from our own pc's and access it from the HTPC and all without having to have all computers powered on etc etc. So nothing fancy right now, as I simply don't have the time to dig into it. Nor do we pay our own powerbill at the moment (thinking of advantages such as using inbuilt torrent client instead of having pc on all night ).
I'm also not quite sure what uPnP is, but will have a read. The activity does indeed sound like some kind of indexing is going on, like after a fresh install and starting up dc++ and it has to hash all files. I did recently swap harddisks and copy over everything to the newer (bigger) ones. So wouldn't be completely unlogical either, I guess.
Thanks!
Offline
My DNS-323 does some strange things at times as well. From time to time the disks spin up and there appears to be some activity. I have funplug installed and the fan controller script but thats it. Everything else is turned off, even tried unplugging the cables, still happens though...... tis a strange thing indeed..
On another note, if I was to activate the 'built-in' FTP server would this cause the disks to spin up even if no-one was accessing the machine?
Tony
Offline
I don't think that any of the standard funplug processes will eventually start up the disks. There are many fan control scripts so I can't tell if this one is responsible for the spin ups. The standard firmware services can be the source of this issue as well, the upnp server for instance. Try turning off services that you don't need.
On my system I'm only using ffp, sshd, samba and a modified fan script and all of this is started from the USB drive. The disks fall asleep when there is nothing to do and wake up only when I access files that reside on them. Samba is the only process which accesses the hard disks but only when I access a share. I can keep the shares connected, the disks will still fall asleep when I don't use a share. That's how it should be.
The same should be true about FTP access. It should only spin up the disks when there is access to one of the files located there.
Offline
Same here,
I am running the ftp service, samba, fun_plug0.5, mt-daapd (firefly) and a fan control script mentioned here http://dns323.kood.org/forum/viewtopic. … 85&p=3 by arokor. Ffp is installed to USB Stick and as backup identically on the Harddisks. Regardless if I boot from USB or from the internal disks the disks spin down if there is nothing to do.
Offline
It seems that my little box has calmed down now, at least. Although sometimes the disks do spin up without me doing much.... (nothing?). That doesn't bother me though. Thanks again!
Offline
I use the 'general use ctrl_fanspeed.sh script' at http://dns323.kood.org/hardware:fan so it shouldn't be causing disk spin ups (which seem to be random anyway, not at fixed times/intervals). Upnp is not activated or running. I have activated the ftp server since yesterday and it does not seem to have made it any worse.
I must say this problem is very annoying though as, like most mechanical things, the damage is done during 'start-up' or in this case 'spin-up'. Leaving them spinning all the time might be better than spinning them up and down every couple of hours or so.....
Is there any way i can see how many times a day they are spinning up? I don't know Linux so please explain any commands you may advise ....
PS I do have ffp 0.5 and am running 1.05 firmware (and have a very basic knowledge of telnet)
Thanks
Tony
Offline
There are entries in "dmesg" that show that the disks were spinning up or falling asleep. Unfortunately there are no times attached to the entries.
I didn't dig into the system so far, so I don't know if dmesg can easily be manipulated to show the date and time of the entries too.
Offline
silversurfer wrote:
There are entries in "dmesg" that show that the disks were spinning up or falling asleep. Unfortunately there are no times attached to the entries.
I didn't dig into the system so far, so I don't know if dmesg can easily be manipulated to show the date and time of the entries too.
How do I access/use dmesg? Am a bit worried I might stuff it up !!! (again)
Offline
flashman1207 wrote:
silversurfer wrote:
There are entries in "dmesg" that show that the disks were spinning up or falling asleep. Unfortunately there are no times attached to the entries.
I didn't dig into the system so far, so I don't know if dmesg can easily be manipulated to show the date and time of the entries too.How do I access/use dmesg? Am a bit worried I might stuff it up !!! (again)
With telnet/ssh access. 'cat /var/log/messages' should show these entries with date and time, along some other info.
Offline
hell0 wrote:
With telnet/ssh access. 'cat /var/log/messages' should show these entries with date and time, along some other info.
This command would be useful if there was a syslog server installed. On a standard system this is not the case. So there is no /var/log/messages file, at least not on my CH3SNAS. I think the manufacturer just didn't know where to put the logs. The RAM disk is to limited for this purpose.
@flashman1207
The command "dmesg" shows the kernel ring buffer. It is available once logged in via telnet or ssh. You can run almost every command with the "--help" option to see what its for and what its parameters are. No harm done that way. ;-)
Offline
On the CH3SNAS the syslog deamon is not running by default but you can start the syslog and kernellog deamon manually from telnet/ssh prompt:
/ffp/sbin/syslogd -O /tmp/syslog/syslog.txt -s 300 /ffp/sbin/klogd -c 3
The -s 300 option will limit the size of syslog.txt file so the ram disk would not be filled completely with loggings.
After that you can see all dmesg messages and more with time stamps by typing the following command from the telnet/ssh prompt:
cat /tmp/syslog/syslog.txt | more
Last edited by EnricoM (2008-07-26 08:22:40)
Offline
Gonna have to show my ignorance further. I'm guessing a CH3SNAS is similar to a DNS-323 so can do what EnricoM suggests. I'm pretty sure that there is no syslog server running as I definately do not have a var/log/messages file anywhere. Anyway, have the following questions:
1. After typing in the commands EnricoM suggests, will my disks stay spun down or will the commands cause them to spin up?
2. Should these commands always be in place ie should I put them in the funplug script? If so, where?
3. How do I know how much RAM I am using?
Thanks to all who have helped
Tony
Offline
1. Yes, the disk should stay spun down since the command tells syslog to log to /tmp which is in memory and not on your hard disk.
2. The standard funplug has a start script (syslogd.sh) in its /ffp/start folder. Doing a "chmod 755 /ffp/start/syslogd.sh" will enable syslog on every start of your NAS.
This syslog daemon is running with standard options since there is no config file for it. You can set options by changing the syslogd.sh script a bit and adding flags for syslogd.
Mine looks like that at the moment:
... name="syslogd" syslogd_flags="-O /mnt/USB/var/log/messages" klogd_flags="-c 3" ...
It is logging to my USB drive. Just change the path to your liking, /tmp for example.
3. You can see the total RAM usage with the command "free". If you want to know how much space is left on your RAM disk then issue the command "df /".
Offline
silversurfer,
Ok, have
1. updated syslogd.sh with following....
name="syslogd" syslogd_flags="-O /tmp/syslog/syslog.txt -s 300" klogd_flags="-c 3"
2. did the "chmod 755 /ffp/start/syslogd.sh" command
3. Rebooted
4. Issued "ps -ea" and it says:
1815 root /ffp/sbin/syslogd -O /tmp/syslog/syslog.txt -s 300
However, when I issue command "cat /tmp/syslog/syslog.txt | more" I get:
"No such file or directory"
Have I done something wrong?
Thanks
Tony
Offline
I'm not sure but I don't think that syslog will create directories on it own. It will create the file syslog.txt if it is not there already but not the /tmp/syslog directory. Better tell syslog to log to /tmp/syslog.txt instead.
Offline
flashman1207 wrote:
syslogd_flags="-O /tmp/syslog/syslog.txt -s 300"
How about;
syslogd_flags="-s 300"
It'll write /var/log/messages, which is the standard location. -s looks like a good idea.
Offline
Btw, if you don't want to have every info message in your syslog, you can limit syslog to only log "important" messages. The -l parameter is for this purpose. Unfortunately I didn't find documentation about what each level represents, not even on the busybox web site.
I think I figured it out anyway.
Levels seem to be:
1 = LOG_EMERG
2 = LOG_ALERT
3 = LOG_CRIT
4 = LOG_ERR
5 = LOG_WARNING
6 = LOG_NOTICE
7 = LOG_INFO
8 = LOG_DEBUG
What about level 0? I don't have a tool to test syslog to find the correct mapping. Maybe level 0 is LOG_EMERG and level 8 is LOG_INFO and there are gaps between the standard syslog levels? I can at least say that I set my syslog to level 5 and it is logging warnings and messages of more critical level. I don't care about info messages on my NAS.
@fonz
Syslogd already has a default log size of "-s 200". Do you really think it is necessary to increase its size?
Do you have the correct mappings for the log levels?
Offline
silversurfer wrote:
Syslogd already has a default log size of "-s 200". Do you really think it is necessary to increase its size?
Do you have the correct mappings for the log levels?
No, didn't mean to increase. Didn't know there's a default limit.
/ffp/include/sys/syslog.h says:
/* * priorities/facilities are encoded into a single 32-bit quantity, where the * bottom 3 bits are the priority (0-7) and the top 28 bits are the facility * (0-big number). Both the priorities and the facilities map roughly * one-to-one to strings in the syslogd(8) source code. This mapping is * included in this file. * * priorities (these are ordered) */ #define LOG_EMERG 0 /* system is unusable */ #define LOG_ALERT 1 /* action must be taken immediately */ #define LOG_CRIT 2 /* critical conditions */ #define LOG_ERR 3 /* error conditions */ #define LOG_WARNING 4 /* warning conditions */ #define LOG_NOTICE 5 /* normal but significant condition */ #define LOG_INFO 6 /* informational */ #define LOG_DEBUG 7 /* debug-level messages */
Offline
Thanks for all the replies, I feel I am learning a lot.
Fonz...... If I go back to using the default syslogd.sh (thereby logging to /var/log/messages) won't this cause the disks to spin up? Won't it write to disk and not RAM?
silversurfer..... would the format be
syslogd_flags="-O /tmp/syslog.txt -l 5"
Will Log level 5 show the disk spin/down info?
Regards
Tony
Offline
Pages: 1 2