Restore/Backup Schedule and Favourites no longer works

[Cabrider]

The backup function appears to work but the restore will no longer work. Today I had to completely reselect all my favourites and restore my scheduled recordings by hand. Viewing the backup shows everything is there but it just won't restore. It loads the disable OTA and that's it! Any causes and solutions? I am running the latest firmware update and all packages have been updated.

 

[MymsMan]

The backup function appears to work but the restore will no longer work. Today I had to completely reselect all my favourites and restore my scheduled recordings by hand. Viewing the backup shows everything is there but it just won't restore. It loads the disable OTA and that's it! Any causes and solutions? I am running the latest firmware update and all packages have been updated.

What version of the webif do you have installed?
Are there any messages in webif-error.log?

I think the latest update to the webif ( 1.4.8-4 ) was a fix to a problem in restore

 

[Cabrider]

Webif 1.4.8-4. Nothing in webif error log.

 

[/df]

What about rsvsync.log? Also, this is a CF issue...

 

[Cabrider]

Says there is no favourites table to restore. There was because I could view it in the backup.

4998	Wed Jun 10 11:13:00 2020: Ignoring: no such table: fav (no such table: fav)
4997	Wed Jun 10 11:13:00 2020: Restoring any favourites.

 

[Cabrider]

Should have copied the full restore session.

5000	Wed Jun 10 11:13:00 2020: Final schedule entries: 12
4999	Wed Jun 10 11:13:00 2020: Slots:
4998	Wed Jun 10 11:13:00 2020: Ignoring: no such table: fav (no such table: fav)
4997	Wed Jun 10 11:13:00 2020: Restoring any favourites.
4996	Wed Jun 10 11:13:00 2020: Opening /var/lib/humaxtv/rsvp.db
4995	Wed Jun 10 11:13:00 2020: rsvsync starting, version 1.1.13.

 

[Cabrider]

I have now also used the Reset Custom Firmware which has reinstalled the custom firmware and have reinstalled only those packages deemed useful to me, such as auto schedule restore; auto update; disable ota; red ring; undelete and of course Webif. I don't know if this will cure the problem and I'm not about to find out right now as I don't want to have to restore everything myself TWICE in one day!

 

[bottletop]

Web interface version: 1.4.8-4
Custom firmware version: 3.13 (build 4028)
Humax Version: 1.03.12 (kernel HDR_CFW_3.13)
Loader Version: a7.30
System ID: 80bc.7e00
I think I also have this issue, as I tried a restore and the recording schedule file is there and correct, but the restore only restores the first item on the backup file.
It gets worse, as after the reboot, the schedule events is empty excpt for the disable OTA reminder.
And if that wasn't bad enough - it seems to have lost 'nugget' functionality even though 'nugget' is installed in the correct location etc.
I couldn't get nugget to work even after checking out the previous solutions after a forum search.

humax# nugget status
Can't connect to nugget, No such file or directory.
humax# nugget ping
Can't connect to nugget, No such file or directory.
humax# sqlite3 /mod/etc/webif.db "select nval from settings where name='rtschedule';"
1
humax# nugget
nugget <command>
  ping             - Test nugget connectivity.
  status           - Show nugget status and version.
  schedule.load    - Load recording schedule from disk.
  schedule.save    - Save recording schedule to disk.
  schedule.slot    - Update timers for schedule slot.
  schedule.db      - sqlite3 database handle info.
  schedule.epg     - Display EPG information for slot.
  schedule.timers  - Show internal timers.
  cryptokey        - Set or display encryption key.
  recordings       - Show in-progress recordings.
humax# which nugget
/mod/bin/nugget
humax# ls -la /mod/bin/nu*
-rwx------ 1 root root 11936 Apr 24  2019 /mod/bin/nugget
humax#

Also I have large number (4000) of /mod/sbin/recmon: can't load library 'libinotifytools.so.0' in the recmon log

 

[Black Hole]

Maybe something is broken.

@af123

 

[bottletop]

The webif-error log is empty. Some of the other logs attached. I couldn't resolve the nugget issue so I did an diagnostics RMA, then installed the webif again.
After the schedule restore I lost nugget again so have stopped playing with this backup HDR.

 

[/df]

I restored the latest automatic backup and got this:

Wed Jun 10 18:23:15 2020: rsvsync starting, version 1.1.13.
Wed Jun 10 18:23:15 2020: Opening /var/lib/humaxtv/rsvp.db
Wed Jun 10 18:23:15 2020: Moving pending entry 0 to spare slot 0
Wed Jun 10 18:23:15 2020: Restoring any favourites.
Wed Jun 10 18:23:16 2020: Slots: +0
Wed Jun 10 18:23:16 2020: Final schedule entries: 1

This appears to have worked as expected: after restore but before restart, channel.db/TBL_FAV was empty while rsvp.db/fav was full of favourites data; after restart favourites were as before and channel.db/TBL_FAV was full. However there's only one event scheduled on this system.

The log entry "Ignoring: no such table: fav (no such table: fav)" presumably occurs because rsvsync runs on each startup, even if there's nothing to be restored, and the favourite-related tables only exist in the pending restore database /var/lib/humaxtv/rsvp.db if a restore is in progress and are dropped afterwards.

 

[xyz321]

After the schedule restore I lost nugget again so have stopped playing with this backup HDR.

Try reinstalling the inotify-tools and openssl packages to fix the missing library issues in the logs.

What is in boot.log?

 

[MymsMan]

After the schedule restore I lost nugget again

I believe the restore process deliberately disables nugget until the next boot (which you would normally need following a successful restore)

Hopefully @af123 will have some time look into the problem's with restore but he has been working away from home recently.

 

[bottletop]

Try reinstalling the inotify-tools and openssl packages to fix the missing library issues in the logs.

What is in boot.log?

I have tried

opkg --force-reinstall install openssl
opkg --force-reinstall install inotify-tools

Thanks it clears those errors in those logs

boot.log has

Starting ahw
Sat Jan  1 00:00:04 UTC 2000
Micom Timestamp: 1591879487 - Thu Jun 11 12:44:47 2020
TZ: BST
Boot: 2
Thu Jun 11 12:44:47 UTC 2020
Cust FW 3.13
Starting bootstrapd init.html
Starting webshell server...
Starting humaxtv...
Waiting for humaxtv to start...
Waiting for humaxtv to start...
Waiting for humaxtv to start... (repeated approx 85 lines)

I believe the restore process deliberately disables nugget until the next boot (which you would normally need following a successful restore)

That is fair enough. But I was expecting nugget function to return after a reboot.
As far as I can tell, the restore failed, taking nugget with it. After many reboots, nugget is still not functioning properly. Is it a critical function?

I tried random things just for the fun of it
I found https://hummy.tv/forum/threads/schedule-without-reboot-option-gone.8200/#post-116055
tried

/bin/wget -qO- "http://hpkg.tv/diag/ssl/fix" | /bin/sh    #because I lost wget at one point
diag fix-flash-packages
opkg --force-reinstall install openssl
opkg --force-reinstall install inotify-tools
opkg --force-reinstall install nugget
opkg --force-reinstall install multienv rsvsync tweak

Beta - [Real-time scheduling] schedule without rebooting

'Course not. But I don't need a Slingbox to do the same. With a Slingbox, you don't need RS at all. Then why would you say: hope I did not find myself in a situation where I could not reboot When you knew I could not find myself in that situation? As to not needing rs, I have to say that...

hummy.tv

opkg files nugget|grep /mod|xargs ls -l
-rwx------ 1 root root 11936 Apr 24  2019 /mod/bin/nugget
-rwx------ 1 root root 44888 Apr 24  2019 /mod/boot/2/libnugget.so.install
-rwxr-xr-x 1 root root   786 Apr 24  2019 /mod/boot/bootstrap.d/xrts
-rw------- 1 root root   222 Aug 14  2016 /mod/boot/env.d/nugget
-rwx------ 1 root root   150 May 26  2016 /mod/boot/xinit.d/install_nugget
-rw-r--r-- 1 root root   127 Aug 21  2016 /mod/webif/plugin/nugget/diag.hook
-rw------- 1 root root    54 Sep  7  2016 /mod/webif/plugin/nugget/diagmeta.hook
-rwxr-xr-x 1 root root  1325 Aug 25  2016 /mod/webif/plugin/nugget/timers.jim
-rw-r--r-- 1 root root  9358 Aug 17  2016 /mod/webif/plugin/nugget/timers.png

Please, if someone has a spare HDR, try the restore schedule with more than 1 set timer and see for yourself. I don't have favourites set.
I just tried it again and enclose the screenshots. The last screenshot was after reboot. No useful schedule (and nugget status or ping doesn't work).

Just tried another reboot and nugget is back working. Not sure why previous reboots didn't sort it out.

Figured out how to get nugget working.
WebIF Schedule restore
Reboot
At this point the scheduled timers are more or less blank due to this bug.
WebIF add 1 timer magnum
Reboot
Then nugget is back working.
Check WebIF / settings / advance settings/real time scheduling is now available - set it to yes
Still need to do /bin/wget -qO- "http://hpkg.tv/diag/ssl/fix" | /bin/sh to get wget back working again
From https://hummy.tv/forum/threads/lots-of-errors-updating-all-packages-help.9557/page-2#post-137919

 

[peterworks]

I have tried this by creating a backup of the schedule then restoring it. I found that all favourites were restored correctly but only two of the five scheduled recordings were reinstated. Screenshots attached.





Hope this helps to work out what is happening. Could see nothing untoward in any logs (to my untrained eye). Anything I can do to assist ?

 

[prpr]

I expect somebody might need to see the restore log, which you don't appear to have provided.

 

[peterworks]

I could not find a restore log in the listed logs in Diag. Is it hidden somewhere ?

 

[/df]

Try /mod/tmp/rsvsync.log, or the last block of entries therein.

 

[peterworks]

Thanks @/df - here is the relevant time in the log:

Sat Jun 13 16:22:33 2020: rsvsync starting, version 1.1.13.
Sat Jun 13 16:22:33 2020: Opening /var/lib/humaxtv/rsvp.db
Sat Jun 13 16:22:33 2020: Restoring any favourites.
Sat Jun 13 16:22:33 2020: Ignoring: no such table: fav (no such table: fav)
Sat Jun 13 16:22:33 2020: Slots: 
Sat Jun 13 16:22:33 2020: Final schedule entries: 5
Sat Jun 13 16:28:45 2020: rsvsync starting, version 1.1.13.
Sat Jun 13 16:28:45 2020: Opening /var/lib/humaxtv/rsvp.db
Sat Jun 13 16:28:45 2020: Moving pending entry 1 to spare slot 0
Sat Jun 13 16:28:45 2020: Moving pending entry 3 to spare slot 1
Sat Jun 13 16:28:45 2020: Restoring any favourites.
Sat Jun 13 16:28:45 2020: Ignoring: no such table: fav (no such table: fav)
Sat Jun 13 16:28:45 2020: Slots: +0,+1
Sat Jun 13 16:28:45 2020: Final schedule entries: 2

 

[/df]

The first group of entries come from the autoschedule restore, which does nothing unless the existing schedule is empty.

The second is your manual restore. Perhaps there is some significance in the fact that the reminders were restored in this case but not the recordings? The "no such table: fav" entry doesn't seem to be consistent with the backup details in your posted screenshot.