1748 replies to this topic

[SpaceRat]

I couldn't care less *vbeg*

As I said, it has been fixed in 1.1 and "we" (That actually means Jörg Bleyel) even implemented a feature request that was almost 2 yrs old:
OWIF 1.1 will not only hide those fillers within the station list, you can even create them inside the Bouquet Editor!

Edited by SpaceRat, 30 November 2016 - 20:12.

[Persian Prince]

Still waiting : "Delete png meta data for libpng >=1.6.x iCCP chunk warning"

 

"E2OpenPlugins"

 

https://github.com/E...Password/pull/2

https://github.com/E...poItalia/pull/2

https://github.com/E...eoItalia/pull/2

https://github.com/E...n-XPower/pull/5

https://github.com/E...urScreen/pull/4

https://github.com/E...nAirPlay/pull/7

https://github.com/E...nmpAgent/pull/3

https://github.com/E...eoViewer/pull/4

https://github.com/E...logClock/pull/2

https://github.com/E...SetPicon/pull/3

https://github.com/E...cheFlush/pull/3

https://github.com/E...ShutDown/pull/8

https://github.com/E...HetWeer/pull/17

https://github.com/E...nterface/pull/3

https://github.com/E...rossEPG/pull/41

https://github.com/E...-Foreca/pull/34

https://github.com/E...toBackup/pull/7

https://github.com/E...ienradar/pull/1

 

"littlesat"

 

https://github.com/l...-PLiHD/pull/476

 

"ims21"

 

https://github.com/i...hBouquet/pull/2

[littlesat]

This was a bit too much... please let me know when I forget one...

[SpaceRat]

You might want to have a look at these two commits:
https://github.com/o...29a47939c4bfb7d
https://github.com/o...7524f6b1033c5af

betacentauri took (one of) the rifle(s) out of some users hands

I consider this an important security fix:
Wikipedia: Mirai (Malware)
Live monitoring of Mirai botnet attacks
October 2016: Blame the Internet of Things for Destroying the Internet Today (Made github, Twitter, Netflix unreachable for hours)
November 2016: Mirai botnet targets Deutsche Telekom routers in global cyberattack (This attack took nearly one million of them down!)

Please be aware that E2 boxes are already one of the known targets of Mirai:
Mirai botnet scanner code example code line:

add_auth_entry("\x50\x4D\x4D\x56", "\x46\x50\x47\x43\x4F\x40\x4D\x5A", 1);              // root     dreambox

Although this isn't OpenPLi's or any oe-a's image's default password, it's clear that Mirai is aware of our boxes.

We can't leave the decision about shooting their feet to the user in this case, as such botnets are a treat to the internet as a whole.

Edited by SpaceRat, 8 December 2016 - 04:31.

[peteru]

Improving security is always a good idea, but I wonder if patching BusyBox telnet server is the right angle of attack.

 

I would think that a system wide configuration using ipchains/iptables would be more sensible. It would work with any telnet server and the same set of rules can be applied to multiple services.

[betacentauri]

iptables need afaik adaptions in the kernel config. Kernel config changes are always dangerous because afterwards box driver maybe cannot be loaded anymore.
I personally never heard that somebody managed it to get it running.

[peteru]

My experience is mainly with ini hardware and Beyonwiz distro. With that combination there are no issues changing kernel config, but there is no need, since all the required features are already enabled. OpenPLi may be more limited in terms of kernel features, but it should be possible to use the dvb driver with a kernel that has more modules enabled.

 

By all means, use whatever method is best for OpenPLi. At least we now both know that you are aware of different methods and that you have a reason for choosing the approach you have taken.

 

For maximum flexibility, the patch that restricts the connections to private or local networks would be selectable at runtime with a command line switch.

[SpaceRat]

For maximum flexibility, the patch that restricts the connections to private or local networks would be selectable at runtime with a command line switch.

Then you can as well keep it outside.
Those who desperately want to open Telnet would quickly find "Howto-shoot-your-feet"-Tutorials for putting the command line switch.

Simply no excuse/reason exists for opening telnet to the outside world.
Tests on honeypots have proven that it only takes minute(s) before attacks come in.

The patch is addressed at those users that are reluctant to learn how to create a key pair for ssh, as opening telnet is sooooo much easier.
Tutorials on how to manually downgrade busybox (Without f*cking up the box on the way) would be harder than those for key pair creation for ssh.

I really hoped that OpenPLi could skip the endless discussion in such an obvious case.

[MastaG]

I see similar matters have been taken for the webif.

If I now tunnel directly trough my router to the openwebif I'm not allowed anymore.

I now create two tunnels, the second (to the receiver) pointing to localhost

 

I'm all for security, great job!

[Persian Prince]

"openpli"

 

https://github.com/O...e-core/pull/144

 

"E2OpenPlugins"

 

https://github.com/E...nterface/pull/3

 

"ims21"

 

https://github.com/i...hBouquet/pull/2

[WanWizard]

E2OpenPlugins merged.

[littlesat]

OpenPLi Merged by Littlesat...

[Abu Baniaz]

Can someone cherry pick this in please? Allows composition of recording filenames to be based on the event name.

https://github.com/O...2f7623c1037aa80

Edited by Abu Baniaz, 22 December 2016 - 21:26.

[Erik Slagter]

Can you explain what it does?

[SpaceRat]

By looking at the code for 3 sec:
It re-orders the filename components
from
1. Date
2. Station
3. Event Name
to
1. Event name
2. Date
3. Station

[littlesat]

Then offer us a patch on our github... this one is openvix...

[SpaceRat]

I was only explaining ...
I don't care about this change ... I bet it even breaks diplaying recordings in EMC and other plugins that rely on the way recordings are labelled now.

[littlesat]

I bet it even breaks diplaying recordings in EMC and other plugins that rely on the way recordings are labelled now. 

->

I bet the same...

[WanWizard]

That happens when you hardcode anything. Eventually it will break. Question is, who's problem is that?

[littlesat]

But what real advantage should it give when the names of recordings are changed, while old recordings stay the same??