Webif browse files missing bottom section

Thaddeus

Member
Hello,

I changed the hard disc in my HDR-Fox-T2 two days ago and although everything seemed fine, the schedule retained all my programmes etc. and has recorded okay since, I'm now missing the bottom section of the Browse files in the Webif, i.e. this part (the photo is from my other HDR-Fox-T2):

1608888368608.png

All I have when I go to Browse Media files is a list of the recorded programmes in alphabetical order.

I thought it might be something in the Webif package settings but I've put the relevant screens from my two boxes side by side and I can't see any difference.

I don't know if this is a clue but the other anomaly is that although I've got the undelete package installed, there's no [Deleted items] folder showing in the Browse Media files list either. It is showing when I go to the media section on my TV via the remote.

Any ideas about how I could fix it?

Thanks.
 

Attachments

  • Humax Browse Files snip 251220.PNG
    Humax Browse Files snip 251220.PNG
    12.8 KB · Views: 7
Sometimes simply clearing the browser cache fixes WebIF anomalies, and it's always worth trying a reboot.

I changed the hard disc in my HDR-Fox-T2 two days ago and although everything seemed fine
Please describe exactly what you did to "change the hard disc" - in what way did you reinstate the CF packages?

When you initially install CF, it replaces the standard firmware in Flash with standard firmware plus just the facilities necessary to execute user code (and a few other things not important to this discussion). The second stage of installation pulls down a minimal set of "packages" to the HDD (including what's needed to run the WebIF).

Fitting a new HDD results in the packages being missing. Accessing the WebIF network address should prompt you to download the packages again (which will then need configuring to your requirements).

Assuming you went through those stages, clearly something has gone wrong. If you didn't go through those stages, I presume you must have tried to copy the old installation across from the old HDD (because you wouldn't even get the results you did if it was a "fresh" HDD). The first thing to try is "fixweb" on the Telnet menu.

https://wiki.hummy.tv/wiki/Telnet
 
Last edited:
Please describe exactly what you did to "change the hard disc" - in what way did you reinstate the CF packages?

When you initially install CF, it replaces the standard firmware in Flash with standard firmware plus just the facilities necessary to execute user code (and a few other things not important to this discussion). The second stage of installation pulls down a minimal set of "packages" to the HDD (including what's needed to run the WebIF).

Fitting a new HDD results in the packages being missing. Accessing the WebIF network address should prompt you to download the packages again (which will then need configuring to your requirements).

Presuming you went through those stages, clearly something has gone wrong. The first thing to try is "fix webif" on the Telnet menu. If you didn't go through those stages, I presume you must have tried to copy the old installation across from the old HDD (because you wouldn't even get the results you did if it was a "fresh" HDD).

https://wiki.hummy.tv/wiki/Telnet
Thanks.

When I switched the Humax back on after I'd changed the hard disc, I formatted it via the TV and then when I put my IP address in the browser on my laptop, it had a screen which invited me to install the webif.

Once the webif was installed, I installed the packages (comparing the settings on my other box) and thought everything was working fine. It was only last night that I noticed the Browse screen issue.

I've just tried "fixweb" via Telnet but it hasn't fixed the issue. It did come up with this error though:

Code:
Please select option: fixweb
Removing package webif from root...
Not deleting modified conffile /mod/webif/html/favicon.ico.
Installing webif (1.4.8-8) to root...
Downloading http://hpkg.tv/hdrfoxt2/base/webif_1.4.8-8_mipsel.opk.
Configuring webif.
/mod/bin/jimsh: can't load library 'libcrypto.so.1.1'
/mod/bin/jimsh: can't load library 'libcrypto.so.1.1'
Collected errors:
 * file_md5sum_alloc: Failed to open file /mod/webif/html/favicon.ico: No such file or directory.
 
It did come up with this error though
That's a red herring; it always happens and we wish it didn't (because it frequently raises eyebrows).

Once the webif was installed, I installed the packages (comparing the settings on my other box) and thought everything was working fine. It was only last night that I noticed the Browse screen issue.

I've just tried "fixweb" via Telnet but it hasn't fixed the issue.
I have little doubt there might be a "clever" way to fix this, but that is beyond my pay grade (and I'm not sure many others will be lurking the forum today of all days). If you want to fix it now, I can offer a sure-fire "nuclear option" (which will involve setting the CF up to your liking from scratch, but I don't find that too arduous):

Reinstate Corrupt CF via RMA Mode (click)
 
Last edited:
That's a red herring; it always happens and we wish it didn't (because it frequently raises eyebrows).


I have little doubt there might be a "clever" way to fix this, but that is beyond my pay grade (an I'm not sure many others will be lurking the forum today of all days). If you want to fix it now, I can offer a sure-fire "nuclear option" (which will involve setting the CF up to your liking from scratch, but I don't find that too arduous):

Reinstate Corrupt CF via RMA Mode (click)
Thanks.

As you hinted, although I posted it today, I wasn't particularly expecting a response so thanks for getting back to me (both today and for the other issues I've had in the last couple of years).

It doesn't feel like something that needs to be sorted in a hurry, so I'll pause a few days in case anyone else has any suggestions before I explore the nuclear option.
 
Have you tried a simple power cycle (which I edited into post 2 later)?
 
Last edited:
Code:
Downloading http://hpkg.tv/hdrfoxt2/base/webif_1.4.8-8_mipsel.opk.
Configuring webif.
/mod/bin/jimsh: can't load library 'libcrypto.so.1.1'
/mod/bin/jimsh: can't load library 'libcrypto.so.1.1'
Try the following in a telnet session:
Code:
diag ssl/fix
opkg update
opkg upgrade
 
Hello all, and season's greetings

I had similar experiences over the years of a directory listing not being complete

What has worked for my own installation is take note of the file in the listing which is showing as it should, then using an FTP or telnet session change directory to show the same list of recordings, then move recordings and sidecar file to a new location from the recording after the recording noted above

There seems to be either a corrupt recording or sidecar file

As stated this solution or workaround has sorted the issue on my system, hopefully, it will benefit others

I use Super-PuTTY for telnet access and FileZilla for FTP access and have betaftpd installed on the Humax HDR Fox T2
 
That's a red herring; it always happens and we wish it didn't (because it frequently raises eyebrows).
...
Indeed, as the reason for the unwanted diagnostic is now understood (ie, a file in the Webif package marked as user-modifiable is a symbolic link and not a regular file), it should be abolished real soon now; not in 1.4.8-11 though.
 
Try the following in a telnet session:
Code:
diag ssl/fix
opkg update
opkg upgrade
Thanks. I've tried this but it hasn't fixed the issue.

Check webif-error.log from the diag menu to see if it is throwing up any messages

I can't spot anything obvious myself but this is everything in the error log:

Code:
9    at file "/mod/webif/lib/ts.class", line 809
8    at file "/mod/lib/jim/oo.tcl", line 73
7    in procedure 'ts genrenib' called at file "/mod/lib/jim/oo.tcl", line 52
6    in procedure '<reference.<ts_____>.00000000000000000080>' called at file "/mod/webif/lib/ts.class", line 813
5    at file "/mod/lib/jim/oo.tcl", line 73
4    in procedure 'ts genre_info' called at file "/mod/lib/jim/oo.tcl", line 52
3    in procedure '<reference.<ts_____>.00000000000000000080>' called at file "/mod/webif/html/browse/index.jim", line 194
2    in procedure 'entry' called at file "/mod/webif/html/browse/index.jim", line 330
1    /mod/webif/lib/ts.class:809: Error: expected integer but got ""


I had similar experiences over the years of a directory listing not being complete
Thanks. I've checked and I haven't got any files missing in the directory listing (comparing the TV view with what's showing in the Browse files section in Webif), so I don't think this is my issue.
 
I can't spot anything obvious myself but this is everything in the error log:

Code:
9    at file "/mod/webif/lib/ts.class", line 809
8    at file "/mod/lib/jim/oo.tcl", line 73
7    in procedure 'ts genrenib' called at file "/mod/lib/jim/oo.tcl", line 52
6    in procedure '<reference.<ts_____>.00000000000000000080>' called at file "/mod/webif/lib/ts.class", line 813
5    at file "/mod/lib/jim/oo.tcl", line 73
4    in procedure 'ts genre_info' called at file "/mod/lib/jim/oo.tcl", line 52
3    in procedure '<reference.<ts_____>.00000000000000000080>' called at file "/mod/webif/html/browse/index.jim", line 194
2    in procedure 'entry' called at file "/mod/webif/html/browse/index.jim", line 330
1    /mod/webif/lib/ts.class:809: Error: expected integer but got ""

I can't say I understand it, but that doesn't look good and does appear to relate to the media browser.

Having thought about what could cause a web page (which is what the WebIF is - a set of web pages) to stop rendering before the end, logically that could be if the source file is incomplete or corrupt (which should have been sorted out by fixweb), or if the server-side scripting crashes. What I mean by the latter is that instead of the web server sending your browser a straightforward HTML file stored ready-prepared (like a document), it is actually creating the HTML on demand by running code (in this case written in Jim, a version of TCL).

The Jim code is written to inspect the current state of the system (in this case the directory listing of recordings stored in a particular folder, and each recording's status) and present it as a web page. This involves some procedure calls to dig into the .hmt file to display each recording's status, and the impression I get from that log file is that the execution is borking, and could be the reason the rendering terminates early.

Assuming I'm on the right track, this is a situation which needs fixing in the WebIF code, revealed by something unusual about your system. You could (possibly) "fix" it by fixing what is unusual about your system, and the information above might be sufficient for @af123 to find the problem in the code and work out how to fix the WebIF so it doesn't happen again, but if you fix it in your system you are destroying the only test case we have. It would be much better if you can hang onto the test case for the time being, and see whether a WebIF update (or tweak) fixes it.

It could be that what's unusual about your system is something delinquent, or it might be that it is normal but never encountered before - that's something else to be investigated.

There are a handful of people on here who might be able to poke into the code and determine whether it really does look like a code problem, and even offer manual tweaks, but I think only af123 could actually provide a formal WebIF update to download by the usual route.

Apologies if all the above is nonsense, I've got my "middle management" hat on rather than a "software engineering" hat (which doesn't exist in my wardrobe - I might be able to find a "software dabbler" hat somewhere). I'm sure the usual suspects will be along to confirm or refute my deductions.
 
It seems to be borking on an invalid 'genre' setting. Perhaps there is a truncated hmt file somewhere.
Try this to locate any small hmt files:
Code:
/mod/bin/busybox/find '/media/My Video/' -name \*.hmt -size -2072c -exec ls -l {} \;
 
I can't spot anything obvious myself but this is everything in the error log:
Yes, looks like a corrupt .hmt file to me too.
You could try running this (from telnet or Webshell command prompt) and seeing if you can spot the broken one:
Code:
humax# cd '/media/My Video'
humax# for f in *.hmt; do echo "$f"; hmt -p "$f"|cut -f13; done
You are looking for a file that doesn't generate a subsequent line containing a number (which should be divisible by 16).

The Webif code is not very robust in some places. This is one of them.
 
Thanks everyone. I think you're all right as I've noticed that the last file in the list "Who Dares Wins" doesn't have the "+otp" at the end of it, doesn't appear on the TV list (I was comparing the TV list with the webif view but not the other way round) and comes up with this when I click on the file name: "Invalid HMT file, /media/My Video/Who Dares Wins_20201223_2200.hmtSynopsisEpisode"

I've also just noticed I can't connect to the drive via FileZilla (it's coming up with a "Connection attempt failed with "ECONNREFUSED - Connection refused by server"" error.

Unless someone has a better idea then I'll wait until it's finished recording, take the hard drive out to connect it to my laptop and then delete it via Linux.
 
I've also just noticed I can't connect to the drive via FileZilla (it's coming up with a "Connection attempt failed with "ECONNREFUSED - Connection refused by server"" error.

Unless someone has a better idea then I'll wait until it's finished recording, take the hard drive out to connect it to my laptop and then delete it via Linux.
A bit extreme, does the services menu show whether betaftpd is running?

Just use Putty, Telnet or the webshell command line (on diagnostics page) to give you a linux command line where you can rename the errant hmt file
You should then be able to use the sidecar tools to rebuild a valid hmt.
 
Back
Top