webif and rs feature request

On the Web-If you can click on the pie-chart to show disk usage, this screen has a sort by size option :-
View attachment 1486
what webif version does that? mine the latest does not do anything when clicking on pie chart.

If I do diagnostics /disk utilisation then I get the same, how the hell I have not noticed until now sort by size is beyond me. Could it have been the browser? Thanks.
 
Not sure what version of Web-If introduced this but it has been there some time, I am presently on Ver. 1.0.18, I would guess it is a problem with your browser, you could try to clear your browser cache. It is the same screen as the Diagnostics >> Disk Utilisation you mention above
 
The screen shot I linked to claims to be 1.0.18.

Update: belay that, the WebIF version is in the signature panel and therefore may not be what was present at the time of the post.
 
Mihaid : what webif version does that? mine the latest does not do anything when clicking on pie chart.
Clicking on the pie-chart to display the Disk Utility option was introduced around July 2013 (or before), although the pie-chart function has has changed more recently
 
Last edited:
And the click was broken for some of the 1.0.17 train unless you clicked at a very specific place. That should be better with 1.0.18.
 
And the click was broken for some of the 1.0.17 train unless you clicked at a very specific place. That should be better with 1.0.18.
yes that's what the problem was, I knew clicking on the pie chart but it just was not working. Thanks for the update, it's all fine now.
 
Since most of the functions I use it for are automatic or rarely used (undelete, flatten, redring, fan, multimode) it can be days or even weeks between my accessing the WebIF GUI. So it's a bit alarming when I do so and find an "It crashed" message, because it could have happened some time ago and I'll find that something has been disabled for several days or more.
So my request is for another 'alarm' indication that doesn't require action by the user beforehand. An on-screen warning would be nice, but I know that won't happen, but two other possibilities come to my mind. One is the rs e-mail facility, which has the advantage that you'll get a warning even if you aren't home (assuming you have the technology) and the other is via the redring package - perhaps making the ring blink, or (not sure if possible) displaying a message on the panel.
Ideally all of the above of course :)
 
So it's a bit alarming when I do so and find an "It crashed" message, because it could have happened some time ago and I'll find that something has been disabled for several days or more.
So turn off the disable with the plugin_autodisable/off diagnostic.
I thought the disable had been changed to a counter, rather than a single shot, so you shouldn't get things being disabled for a single random crash anyway.
 
I thought RS provides an email alert if the linked HDR-FOX hasn't been seen for a period?
On the occasions it's happened I haven't had any e-mail. It's possible this has only happened on the non-RS box (only one of the two has RS enabled), but I'm fairly sure both have had 'unseen' crashes in the last few months.

So turn off the disable with the plugin_autodisable/off diagnostic.
I thought the disable was to prevent the box getting stuck in a crash/reboot loop should it have been tripped up by a CFW feature. It is paramount that the box does its 'normal' functions of recording and playback as I'm not always around to sort problems out (and if there is going to be a problem 9/10 times it's when SWMBO is home alone - she has this effect on technology).
So unless the disable function is considered of no value now (in which case let's lose it) I'd rather not disable it.

I thought the disable had been changed to a counter, rather than a single shot, so you shouldn't get things being disabled for a single random crash anyway.
I must have missed that one coming in. It would still be nice to be alerted when a disable has been effected even if a single crash is not alarmed.
 
I thought the disable was to prevent the box getting stuck in a crash/reboot loop
It is, but I've never suffered from it.
So unless the disable function is considered of no value now (in which case let's lose it) I'd rather not disable it.
You'll have to accept that your plugins may get disabled if there is a crash shortly after boot then. The presence of a crash message doesn't mean they have been though.
I must have missed that one coming in.
It was my suggestion of a counter, which is why I remembered it.
It got implemented as a two stage thing, rather than a counter, so you need two crashes after less than a minute of uptime to trigger the plugin disable now.
Except, there's a bug in it... it creates a flag file after the first crash and after the second crash, it disables plugins but doesn't remove the flag file.
So, once you have re-enabled all the plugins, the first time it crashes after that, it will disable them all again, instead of the second time.
It's easy enough to fix though:
Code:
humax ~ # diff /mod/boot/posttvcrash~ /mod/boot/posttvcrash
--- /mod/boot/posttvcrash~
+++ /mod/boot/posttvcrash
@@ -73,6 +73,7 @@
elif [ -z "$uptime" ] || [ "$uptime" -lt 60 ]; then
  if [ -f $ffile ]; then
  echo "Second fast crash - disabling plugins"
+  rm -f $ffile
  disable_plugins
  else
  echo "First fast crash"
 
Last edited:
What about the block above that?

Code:
if [ -f $ffile ]; then
  lct="`/bin/date -r $ffile '+%s'`"
  now="`/bin/date '+%s'`"
  [ $((now - lct)) -gt 300 ] && rm -f $ffile
fi

Works as intended as far as I can see..
 
Last edited:
Apologies. I completely missed that block.
(I'll go and tend my sore head somewhere else...)
 
Ooh. Spooky. I just WebIFed into a box to set something regarding another thread - and it had the crash message :eek:
What's a bit weird is that it says it was Uptime 29s (which they mostly have been) but from the time it must have occurred at the END of a recording. ?!!?

Re the few posts above, is the consensus that disable now only happens if there are two early uptime crashes in quick succession? (so eg, one a day for a week wouldn't trigger disable.)
If that is the case it's not as big a concern and I'll leave it activated as my crash log only shows discrete crashes (so far).
But an 'on the box' alert would still be nice to have :)
 
...is the consensus that disable now only happens if there are two early uptime crashes in quick succession? (so eg, one a day for a week wouldn't trigger disable.)
That's right. You need two quick crashes (within the first minute following startup) to occur within 5 minutes of each other.
 
Back
Top