Customised Firmware 2.10 (for 1.02.27)

Has anyone noticed a difference in power management on the HD model (but not the HDR) under the new custom firmware 2.10? I've noticed that the external HDD LED is lit much of the time while the box is (apparently) in standby and the disk will be spinning. Power Saving In Standby and Automatic Power Down are enabled, but I still have access to WebIF even when the box is in this state, the "standby" status of the box being confirmed by WebIF itself . However, if I power on the box and then power it off again, the box will power off completely so that the HDD LED is no longer on and WebIF no longer responds. Later on, the box will go back into this half-awake mode for no apparent reason.

I haven't run the standard 1.02.27 firmware , except during the firmware upgrade, so I've no idea if this is a customisation issue or something that Humax has introduced with 1.02.27.

Any ideas?
 
No ideas, but I have noticed I do not have boot problems any more, which may be related. I cannot say now whether this occurred with 1.02.27 or 2.10.

I will make observations and report back.
 
Using a USB lamp I am not seeing any unexpected USB power during standby (HD-FOX, 1.02.27, CF2.10).
 
The box does power down completely of its own accord sometimes, it just seems to spend far too much time in this half-awake state when there doesn't seem to be any reason for it to do so.
 
This seems to happen if you have programmed a sleep timer without also programming a wake-up timer. Not sure why but it seems to confuse the box.
 
This seems to happen if you have programmed a sleep timer without also programming a wake-up timer. Not sure why but it seems to confuse the box.

Ahhh, yes, could be. I recently removed an 18:00 wake-up timer using RS. After reading your post I can see that the 18:15 sleep timer is still there, so I've deleted it and will see how goes. If it fixes things then I'll put a note to that effect in this thread. Thanks to BH and AF123 for taking a look.
 
Any idea why top has stopped working recently?

Code:
humax# top
top: can't load library 'libncurses.so.5'
humax#
Web interface version : 0.9.6-4 Custom firmware version : 1.16
 
Any idea why top has stopped working recently?
Probably because you have installed sysmon. Included in its dependency list is 'procps' which contains a new version of 'top'. If you install 'ncurses' it should fix the problem since it is missing from the dependency list for procps.
 
Probably because you have installed sysmon. Included in its dependency list is 'procps' which contains a new version of 'top'. If you install 'ncurses' it should fix the problem since it is missing from the dependency list for procps.
I've just uploaded a new copy of procps with a fixed dependency list.
 
I've just uploaded a new copy of procps with a fixed dependency list.
Thanks.

BTW There is a bug in this version of procps/libc in that it doesn't recognise the two processor cores. From 'proc/sysinfo.c':

Code:
  // ought to count CPUs in /proc/stat instead of relying
  // on glibc, which foolishly tries to parse /proc/cpuinfo
  //
  // SourceForge has an old Alpha running Linux 2.2.20 that
  // appears to have a non-SMP kernel on a 2-way SMP box.
  // _SC_NPROCESSORS_CONF returns 2, resulting in HZ=512
  // _SC_NPROCESSORS_ONLN returns 1, which should work OK
  smp_num_cpus = sysconf(_SC_NPROCESSORS_ONLN);
  if(smp_num_cpus<1) smp_num_cpus=1; /* SPARC glibc is buggy */
 
Probably because you have installed sysmon. Included in its dependency list is 'procps' which contains a new version of 'top'. If you install 'ncurses' it should fix the problem since it is missing from the dependency list for procps.

Thanks for that, procps ver. 3.2.8-1 fixed the top problem
 
I haven't had the 'annoying retune' issue at all. Mind you, Winter Hill had it's DSO YEARS ago!
 
I've had no further problem after the initial shock, but I'm still hoping af123 will fix that for us.
 
Could you not poke one into the schedule? Or do you need to see one to know what it looks like? I've got a manual backup from March 30, which probably contains a retune event - but you'll have to give me a clue what to do with it.
 
It's not only switch-over (switch off?) - check whether you have any shuffling scheduled in the future (Mendip has another next year).
 
Could you not poke one into the schedule? Or do you need to see one to know what it looks like? I've got a manual backup from March 30, which probably contains a retune event - but you'll have to give me a clue what to do with it.
I could do with a copy of the line from the backup file (/mod/var/backup/....) - it's text so could just be pasted into a post of PM.
A DSO event has 11 in the 6th column so the line looks like:

Code:
event ? ? ? ? 11 ...
 
Back
Top