Empty Webif EPG

S

SunnysideUK

Member
This morning my Humax4 Webif EPG is empty of data. The EPG from the TV remote is fine,
I changed my favourites to none, same
I tried fix-flash packages, same.
Naturally rebooted.

I can search within the EPG OK.
My Schedules look OK.

Any ideas on a fix?
 
prpr said:
ls -lR /mnt/hd1
Click to expand...
On last look, parts of the EPG has returned after many minutes but parts are empty and some channels missing.
On this boot now it looks ok
Code: 
humax4# ls -lR /mnt/hd1
/mnt/hd1:
drwxr-xr-x    2 root     root          4096 Feb 11 10:40 dvbepg
-rw-r--r--    1 root     root       4751360 Feb 11 11:03 epg.db
drwx------    2 root     root          4096 Nov 12  2021 lost+found

/mnt/hd1/dvbepg:
-rw-r--r--    1 root     root       3898044 Feb 11 10:40 epg.dat

/mnt/hd1/lost+found:
humax4#
 
SunnysideUK said:
On last look, parts of the EPG has returned after many minutes but parts are empty and some channels missing.
Click to expand...
Database rebuilds only happen every 15 minutes. If you want to force a rebuild, you have to stop and start the epg service and wait a minute for it to complete.
I've got something better in the works that will update whenever a change is detected in the data the Humax software caches.
Just out of interest, what does diag epgrange report?
 
prpr said:
I've got something better in the works that will update whenever a change is detected in the data the Humax software caches.
Click to expand...
That will help schedchk since it can only detect changes that actually appear in the database
 
prpr said:
Is the epg service running?
Post the output of ls -lR /mnt/hd1
Click to expand...
Yes both times.
This boot this pm and the EPG is empty again.
{edit @18:35: some 2 hours later still empty. Stopped EPG service, started EPG service {note: no notice pops-up on start)}.
Code: 
humax4# ls -lR /mnt/hd1
/mnt/hd1:
drwxr-xr-x    2 root     root          4096 Feb 11 14:35 dvbepg
-rw-r--r--    1 root     root       4829184 Feb 11 15:45 epg.db
drwx------    2 root     root          4096 Nov 12  2021 lost+found

/mnt/hd1/dvbepg:
-rw-r--r--    1 root     root       3988552 Feb 11 14:35 epg.dat

/mnt/hd1/lost+found:
humax4#
 
Last edited:
prpr said:
idiag epgrange
Click to expand...
Code: 
humax4# diag epgrange
Running: epgrange
Tue Sep 17 00:35:00 BST 2024
Tue Feb 18 22:00:00 GMT 2025
humax4#
Although I didn't mention it, I had previously stopped and restarted the EPG service via the GUI Service Management. I'm sure I waited over 10mins, refreshing the EPG page.
 
Does the epg.db file exist right after boot? Is something deleting it?
Obviously both the files exist with content and contemporary timestamps, so there is some data in them. Do the timestamps continue to update (they don't if you're playing a recording)?
What's changed since it worked? I can't help feeling there is something you aren't telling us, but I don't know what.
Anyway, the epg.dat file needs a periodic cleanout to get rid of old data, as the Humax software is hopeless at it. You can do it 'cleanly' from the maintenance mode menu or uncleanly by just deleting the file and immediately rebooting and hoping.
 
Like the OP, my webif EPG is blank and I've been unable to determine the cause.
Any suggestions would be welcome.

I first noticed the issue a couple of days ago. I powered on the box and was greeted by the retune /setup wizard (happens occationally).
I completed the usual retune and rebooted a couple of times. I browsed to webif's epg to add some scheduled recordings and found the epg empty.
The EPG via the TV's UI was fully populated.

Initially I thought the issue was due to the BST clock change so waited until yesterday before checking again, and found the issue still present.

I've tried the following trying to diagnose.
  1. Fully retuned 2 more times.
  2. Cleared the EPG from maintenance mode a couple of times.
  3. Completed a long fix-disk overnight. No faults reported.
  4. Run fixweb from the terminal.
  5. Checked for any package updates (was already fully updated).
  6. Stopped and started the EPG from webif.
  7. epg -p dump outputs a full epg.
  8. I do see some entries in webif-errors.log (attached)
The epgrange diagnostic reports.

humax# diag epgrange
Running: epgrange
Wed Apr 1 01:00:00 BST 2026
Thu Apr 9 00:56:00 BST 2026

humax# ls -lR /mnt/hd1
/mnt/hd1:
drwxr-xr-x 2 root root 4096 Apr 1 11:06 dvbepg
-rw-r--r-- 1 root root 5435392 Apr 1 11:04 epg.db
drwx------ 2 root root 4096 Nov 10 2016 lost+found

/mnt/hd1/dvbepg:
-rw-r--r-- 1 root root 4546584 Apr 1 11:06 epg.dat

Web Interface Version: 1.5.4-4
Custom Firmware Version: 3.13 (build 4028)
Humax Version: 1.02.32 (kernel HDR_CFW_3.13)
Loader Version: a7.30
 

Attachments

Excellent bug report, thanks!
It looks like you have corruption in the /var/lib/humaxtv/rsvp.db database.
It probably won't tell us anything, but if you can preserve/zip it and DM it to me I will take a look.
The solution is then: rm /var/lib/humaxtv/rsvp.db
Terrahawk said:
I powered on the box and was greeted by the retune /setup wizard (happens occationally).
Click to expand...
The two events may not be entirely unconnected, as this usually happens when there is a corrupt database in the flash.
rsvp.db is not one that the native firmware creates though, so it won't delete/re-create it when the wizard runs.
Have you ever run the System Flush on this box? It might be worth doing so, having made note of any settings in the flash, and remember to run fix-flash-packages afterwards and restore your settings. It might stop the occasional wizard annoyance.

I also wondered why you weren't running CF 3.14 but then noticed you were on 1.02.32 base firmware which I never updated... nobody asked for it so I assumed nobody was using it.
 
Thanks.:doublethumbsup: Removing the rsvp.db resolved the issue. I've sent a DM with the database.
I've never ran a system flush. I'll do that next. Do you need to RMA via webif first?

I've not upgraded the base firmware to avoid the slower onbox epg navigation that comes with later versions.
If you have free time at some point, a CF 3.14 on 1.02.32 base would be magic ☺️
 
Terrahawk said:
I've sent a DM with the database.
Click to expand...
Thanks. It's rather a weird fault though. The database is the expected size and is not corrupted as such, but there is absolutely no sign of the "pending" table which was causing the errors, even looking with a hex dump. It's like it was never there. I can't account for it.
I've never ran a system flush. I'll do that next. Do you need to RMA via webif first?
Click to expand...
You don't need to - it makes more work putting all your settings back, rather than those just stored in the flash, but you may prefer a clean slate and start again.
 
Thanks @prpr :thumbsup_:

The upgrade went without issue.

Web Interface Version: 1.5.4-4
Custom Firmware Version: 3.14 (build 1757805000)
Humax Version: 1.02.32 (kernel HDR_CFW_3.14)
Loader Version: a7.30
 
You must log in or register to reply here.
Back
Top