--
You received this message because you are subscribed to the Google Groups "Alt-F" group.
To unsubscribe from this group and stop receiving emails from it, send an email to alt-f+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/alt-f/6a06f625-43f1-4fbf-b164-7b63b5ea3362n%40googlegroups.com.
According to <https://sourceforge.net/p/alt-f/featurerequests/32/#a96d> there were some updates to the dns320l-daemon to add ReadAlarm, WriteAlarm, EnableAlarm and DisableAlarm commands. With these, I believe you can schedule the box to power itself up at a particular time.My dns320l-daemon doesn't have these commands.
You might try using the the attached one.
Couldn't get WOL to work, but alarm wakeup is. It has to be rearmed after elapsed.I think to remember that "automatic power up after power fail" means that it will return to the previous state after power fails -- if it was off it will remains off, though not sure.
I have webui and scripts to handle all that, but there are multiple dependencies between them, I might miss some, so not posting them.
I've managed to hack dns320_utils to work without segfaults - at least it now accepts commands to show the status of the alarm flag and to get/set the WOL status, but I haven't tested it for using the alarm. This might be an alternative to those not game enough to replace their dns320l-daemon binary.On Wed, 19 May 2021 at 03:04, João Cardoso <whoami...@gmail.com> wrote:You might try using the the attached one.Thanks João.Couldn't get WOL to work, but alarm wakeup is. It has to be rearmed after elapsed.I think to remember that "automatic power up after power fail" means that it will return to the previous state after power fails -- if it was off it will remains off, though not sure.I had the "auto powerup after power fail" set (GetPowerRecoveryState = 128), but it doesn't power on after I power down using the GUI. So I think what you're saying is that, if I pull the plug instead of powering down, then it will power up again when power is restored?
Would it be a viable option (for XT)
to terminate services and unmount filesystems, then cut power using a smart plug? That way, the power is cut rather than powering itself down, yet no risk of filesystem corruption. I think this can be achieved by running /etc/init.d/rcE after ensuring /boot/.poweroff is removed?
I have webui and scripts to handle all that, but there are multiple dependencies between them, I might miss some, so not posting them.That's fair. Perhaps in some future release of Alt-F?
You said: "According to <https://github.com/martin-vi/dns320_alarm>, the DNS320-revA is just the "L" version but with less RAM."That is incorrect, according to the info I collected at the time (I have neither a DNS-320 rev-A nor rev-B box). It is correct for the DNS-320-rev-B.In Alt-F, the rev-A does not use the dns320l-daemon, instead it uses a shell script, /usr/sbin/dns320-temp.sh, only for reading the temperature. That script is an adaptation of Lentin's findings, see https://jamie.lentin.co.uk/devices/dlink-dns325to terminate services and unmount filesystems, then cut power using a smart plug? That way, the power is cut rather than powering itself down, yet no risk of filesystem corruption. I think this can be achieved by running /etc/init.d/rcE after ensuring /boot/.poweroff is removed?Roughly, but to do a clean shutdown without actually powering itself down it is enough to call /usr/sbin/halt in a cronjob. That calls rcE, which orderly shuts down the system and then just halts the system (theoretically stopping the processor).Only a DNS-320-rev-A owner can confirm that if at that point power is cut and restored the box will power up.
Best regards.
Hi:
I am using DNS-321-Ax.Firmware says "Alt-F 1.0 with kernel 4.4.86, and is flashed with "Alt-F-1.0, initrd" and kernel "Alt-F-1.0, kernel 4.4.86"
I have been trying to follow how to get the DNS-321 to power on after power failure. All I need is for the box to turn on after power is restored. (It is in a difficult location to reach, so when power is lost, it is really a hassle to get to it to push a button to power on.)
I looked at https://jamie.lentin.co.uk/devices/dlink-dns325. However, there is no rc.local file at /etc in my box. I checked via SSH in root and in the Alt-F folder.
Is there something I need to install first? Is there a fix I can download?
Any help is very much appreciated.
Thanks.
On Saturday, May 29, 2021 at 3:22:52 PM UTC-5 ext...@gmail.com wrote:
Thanks all , especially Jezzaaaa for his first answer.
Given the fact I have other issues with my other NAS (Dell Power Edge T20 , the disks are too hot there) , I have rethinking my setup and plan to have my old good DNS-320 turn on all the time , especially as the disks can be spindown , with drives sleeping , I think the whole power consumption should not very huge , actually I don't know how the Dlink consume (I can know it as the smart plug I planned to use to turn on the dlink can give power consumption).
That being said , to make this post even better , I promise to do the test , verify the smart plug can wake up the Dlink shutdown by the schedule task as well as indicate power consumption of everything.
Thanks again for this thread.
See ya
XT
Le jeudi 20 mai 2021 à 18:22:51 UTC+2, João Cardoso a écrit :
On Wednesday, May 19, 2021 at 8:53:48 PM UTC+1 gilles.m...@gmail.com wrote:
Le mer. 19 mai 2021 à 19:50, João Cardoso <whoami...@gmail.com> a écrit :
You said: "According to <https://github.com/martin-vi/dns320_alarm>, the DNS320-revA is just the "L" version but with less RAM."That is incorrect, according to the info I collected at the time (I have neither a DNS-320 rev-A nor rev-B box). It is correct for the DNS-320-rev-B.In Alt-F, the rev-A does not use the dns320l-daemon, instead it uses a shell script, /usr/sbin/dns320-temp.sh, only for reading the temperature. That script is an adaptation of Lentin's findings, see https://jamie.lentin.co.uk/devices/dlink-dns325
to terminate services and unmount filesystems, then cut power using a smart plug? That way, the power is cut rather than powering itself down, yet no risk of filesystem corruption. I think this can be achieved by running /etc/init.d/rcE after ensuring /boot/.poweroff is removed?
Roughly, but to do a clean shutdown without actually powering itself down it is enough to call /usr/sbin/halt in a cronjob. That calls rcE, which orderly shuts down the system and then just halts the system (theoretically stopping the processor).
Only a DNS-320-rev-A owner can confirm that if at that point power is cut and restored the box will power up.
I confirm it works on my DNS-320-rev-A : after calling halt, waiting, and unplugging the cord, the system powered up again when the cord was plugged back in.
Thanks for confirming that.
So, XT can setup a cron job (Services->System, cron, Configure) to halt the box (using the /usr/sbin/halt command) a few minutes before the smart plug cuts power.
Best regards.
--
You received this message because you are subscribed to the Google Groups "Alt-F" group.
To unsubscribe from this group and stop receiving emails from it, send an email to alt-f+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/alt-f/2e45956f-9267-49ac-b918-efe65712e980n%40googlegroups.com.