Custom 1.0.7.8 firmware is ready to download

[fireflash]

time to nit-pick. Something keeps accessing the hard drive, so the drive won't power down. MLDonkey is disabled, and so is the media server.. Not much is running on it except what it needs to do it's job. I could try and kill watchdog just for giggles and see what it does.. Any suggestions?

[exbliss]

thanks everyone and most specially to kfurge for all your effort...

anyway, i am getting this warning.. any thoughts?

i just installed this custom firmware version 1.0.7.8 from version 2.0.0.7 via web upload.. at first nothing happened because my box doesnt reboot by itself nor even by the power button in front.. yeah, i know thats really weird.. so i then reset the box totally... and after the reboot, it reverted back to custom firmware version 1.0.4.6... i didnt know where it came from.. anyway, i tried installing the custom firmware version 1.0.7.8 via web upload and everything looks ok... except that i could not log in to the telnet using admin account so i login as root...

i read somewhere you mentioned that i kindoff store the previous firmware as sort of backup or something maybe thats why it reverted back to 1.0.4.6 from 2.0.0.7...

then i noticed that mldonkey is not starting up on its own via the rc.local script.. so i just start it myself after i login via telnet and check the logs and get this messege...

WARNING: Directory /shares/MYVOLUME1/MYSHARE1/.mldonkey is full, MLDonkey shuts down

does this means i dont have enough free space?

[root@(none) MYSHARE1]$ df -h
Filesystem                Size      Used Available Use% Mounted on
rootfs                   67.1M     67.1M         0 100% /
/dev/root                 1.3M      1.3M         0 100% /mnt
/dev/ide/host2/bus0/target0/lun0/part2
                         67.1M     67.1M         0 100% /

if its is and so... how do i clean up and free up some space? HELP! ='(

[kfurge]

WARNING: Directory /shares/MYVOLUME1/MYSHARE1/.mldonkey is full, MLDonkey shuts down

You have some strange stuff going on. Your primary data partition on the HDD is not there! You should have a /dev/se/2 mounted on /shares/MYVOLUME1.

My output looks like:

[root@WL700gE cups]$ df -h
Filesystem Size Used Available Use% Mounted on
rootfs 67.1M 67.1M 0 100% /
/dev/root 1.3M 1.3M 0 100% /mnt
/dev/ide/host2/bus0/target0/lun0/part2 67.1M 67.1M 0 100% /
/dev/se/2 232.3G 170.6M 232.1G 0% /shares/MYVOLUME1
/dev/se/2 232.3G 170.6M 232.1G 0% /opt


I'd recommend completely purging the router of the 2.0.xx settings by pressing the recessed reset button while powering up the router. You may have to reflash the 1.0.7.8 image after this.

- K.C.

[exbliss]

You have some strange stuff going on. Your primary data partition on the HDD is not there! You should have a /dev/se/2 mounted on /shares/MYVOLUME1.

I'd recommend completely purging the router of the 2.0.xx settings by pressing the recessed reset button while powering up the router. You may have to reflash the 1.0.7.8 image after this.

- K.C.

Oh my god! now that you mentioned, yeah i realized its missing.. in the web it says unmounted and there isnt any option to mount it... in the explorer, \\192.168.1.1 doesnt have the share forlder...

what have i done?!?!

i have already reset it a couple of times already and reflash it again and its still not mounting... is there anything i can mount this manually?

[exbliss]

Oh my god! now that you mentioned, yeah i realized its missing.. in the web it says unmounted and there isnt any option to mount it... in the explorer, \\192.168.1.1 doesnt have the share forlder...

what have i done?!?!

i have already reset it a couple of times already and reflash it again and its still not mounting... is there anything i can mount this manually?

i did my homework and took crash course on unix 101 and... eureka! it turns outs i just had corrupted partition table.. so i just fixed with fdisk and rebooted a couple of times after reconfiguring stuff and al... and walla! everything is perfect! Thanks kfurge! your the best!

[kfurge]

time to nit-pick. Something keeps accessing the hard drive, so the drive won't power down. MLDonkey is disabled, and so is the media server.. Not much is running on it except what it needs to do it's job. I could try and kill watchdog just for giggles and see what it does.. Any suggestions?

It's not watchdog. Try disabling more applications. You can speed up the trial and error process by issuing the following command from a terminal window:

hdparm -y /dev/ide/host2/bus0/target0/lun0/disc

Keep disabling applications and doing this until the drive stays spun down for at least 60 seconds.

FWIW, the drive does not completely spin down with this firmware. It enters a low RPM mode which reduces drive wear and tear at spin up.

- K.C.

[fireflash]

It's not watchdog. Try disabling more applications. You can speed up the trial and error process by issuing the following command from a terminal window:

hdparm -y /dev/ide/host2/bus0/target0/lun0/disc

Keep disabling applications and doing this until the drive stays spun down for at least 60 seconds.

FWIW, the drive does not completely spin down with this firmware. It enters a low RPM mode which reduces drive wear and tear at spin up.

- K.C.

if I disable anything else the poor thing will stop working! I'll see what else I can kill, but it's running just enough to do what it needs to do, nothing special.

Edit: Here's what I've got running:

PID Uid VSZ Stat Command
1 root 9116 S init
2 root SW< [keventd]
3 root SW [events/0]
4 root SWN [ksoftirqd_CPU0]
5 root SW [kswapd]
6 root SW [bdflush]
7 root SW [kupdated]
8 root SW [mtdblockd]
9 root SW [khubd]
14 root SW< [mdrecoveryd]
74 root SW< [ksed]
75 root SW [ksechanged]
77 root SW [gpio_out_pulse]
78 root SW [button_notify_u]
106 root SW< [raid1d]
111 root SW< [raid1d]
120 root SW [kreiserfsd]
195 root 3600 S syslogd -R 255.255.255.255
197 root 3552 S klogd
199 root 11040 S httpd 80 y
201 root 11020 S httpd 8080 share_browsing
207 root 4908 S upnp -D -L br0 -W eth0
208 root 4708 S nas /tmp/nas.lan.conf /tmp/nas.lan.pid lan
217 root 1880 S udhcpd /tmp/udhcpd_lan.conf
221 root 2188 S dnsmasq -i br0 -r /tmp/resolv.conf
228 root 8520 S /usr/local/samba/sbin/smbd
241 root 1856 S udhcpc -i eth0 -p /var/run/udhcpc.pid -s /tmp/udhcpc -H Routerhosen
242 root 6792 S /usr/local/samba/sbin/nmbd
267 root 8520 S /usr/local/samba/sbin/smbd
275 root 5824 S waveservermain
277 root 5824 S rcamdmain
278 root 4808 S infosvr br0
289 root 5824 S usbcopymain
512 root 2164 S utelnetd -d -i br0
519 root 1364 S /opt/sbin/dropbear
4235 root 1444 R /opt/sbin/dropbear
4236 admin 2108 S -bash
4237 root 2108 S bash
4242 root 1792 R ps

[kfurge]

if I disable anything else the poor thing will stop working! I'll see what else I can kill, but it's running just enough to do what it needs to do, nothing special.

Edit: Here's what I've got running:

The only think you have running that I don't is, nas, PID 208. What's that? One other thought... Are you using the original HDD or did you change it out?

What happens when you use the hdparm command? Does the drive spin back up immediately?

- K.C.

[marcolino]

many tanks for your prompt respons, but i forget to mention that i have already tried to reflash three times without results.

Just after telneting is shown:

(none) login:

and no passwords are good. No way to gain control. I am stuck.

Any suggestion will greatly appreciated.

[kfurge]

and no passwords are good. No way to gain control. I am stuck.

I'm seeing a trend here... The conversion of rc.local from 1.0.4.6 to 1.0.7.8 can be somewhat difficult and many people are getting stuck with no ability to log in. Unfortunately, unless you put the copy button bypass into *your* rc.local, the bypass does not work.

In retrospect, the bypass is in the wrong spot. It should be in /apps/bin/rcex which mounts the /opt directory and executes rc.local. That way pressing the copy button will *force* a boot with the firmware's /opt directory which is known to work. I'll be uploading an updated firmware image shortly which addresses this issue.

- K.C.

[fireflash]

The only think you have running that I don't is, nas, PID 208. What's that? One other thought... Are you using the original HDD or did you change it out?

What happens when you use the hdparm command? Does the drive spin back up immediately?

- K.C.

I thought NAS was another form of network disc access... fat lot of good that does though, I can't get anything to connect to it :P Samba works with everything, so far.

It is the original drive, the Hitachi drive. I tried killing nas, waveservermain, rcamdmain (which seems to have killed my 'Ready' light ?!), and watchdog, and issued the command and it still spins right back up after a few seconds. I don't know what else to kill.

[kfurge]

It is the original drive, the Hitachi drive. I tried killing nas, waveservermain, rcamdmain (which seems to have killed my 'Ready' light ?!), and watchdog, and issued the command and it still spins right back up after a few seconds. I don't know what else to kill.

OK. There's one final place to look. Make sure "mount" shows nodiratime and noatime as below:

/dev/se/2 on /shares/MYVOLUME1 type reiserfs (rw,noatime,nodiratime)
/dev/se/2 on /opt type reiserfs (rw,noatime,nodiratime)

If these options are not set then any time an applications reads from the HDD the access time will be updated causing a write to the HDD.

You set these options with:

mount -o remount,noatime,nodiratime /shares/MYVOLUME1

- K.C.

[fireflash]

OK. There's one final place to look. Make sure "mount" shows nodiratime and noatime as below:

/dev/se/2 on /shares/MYVOLUME1 type reiserfs (rw,noatime,nodiratime)
/dev/se/2 on /opt type reiserfs (rw,noatime,nodiratime)

If these options are not set then any time an applications reads from the HDD the access time will be updated causing a write to the HDD.

You set these options with:

mount -o remount,noatime,nodiratime /shares/MYVOLUME1

- K.C.

Nope, that ain't it either.. both are mounted correctly. I'm gonna try something silly like disabling my wireless.. Just had a silly thought.. And I'm not using it right now anyways.