1309 replies to this topic

[kalkan99]

Hello to everyone,
my satellite device vu + duo 4k
I love openpli software.
I'm trying to get a backup from my device in both 7.0 and 7.1 software.
The options I use are backupsuite, automatic fullbackup plugins.
seems to have taken a backup, but the device does not turn on again when loading the device.
is this the right method?
or is there another method?
or what could be the error?
thanks.

[master64nl]

Here's the mipsel download from a mipsel machine (et8000)

Ik heb deze geïnstalleerd op mijn dm800se maar hij doet niets. Weet iemand hoe dit werkt?

I installed this ipk on my dm800 but it doesn’t work. Anyone knows how this should work?

[master64nl]

I managed to get it to work, https://wiki.openpli...kup/BackupSuite"]https://wiki.openpli.org/Backup/BackupSuite. See the WIKI. As a test I turned the box off with the USB stick connected. The blue led started blinking but kept on blinking ‘until pids fly’. So I restored the backup although this doesn’t work as described. One can only create a backup on your NAS which succeeded although this isn’t described too how restore this. Anyone has a hint?

Edited by Frenske, 20 July 2019 - 10:00.
Please post in the proper language. You posted in Dutch although this is the English section. Graag voortaan in de juiste taal posten. Indien je die taal niet machtig bent dan maak je (eventueel) een nieuw topic aan in de juiste taal.

[Persian Prince]

@PLi

 

I think it's better to set your "reporefs.conf" to "a9ec02a55887b46cdca1c93a921088b50cd7e0a3" (even for develop) because it is possible that my future commits break the support for you as I want to use our "getBoxBrand" and some other imports which you don't have.

 

As we're using the master branch of OE (which is newer than any other) for Open Vision 9.2, I'm going to do some changes in both python and shell scripts.

 

I will try my best to keep this plugin backward compatible and won't show errors with more "try ... except" but for now use that hash then I give you the green light that everything is ok again.

 

Regards,

Persian Prince

Edited by Persian Prince, 10 September 2019 - 20:58.

[WanWizard]

You're not saying that you've invented yet another flavour of "branding"?

[Persian Prince]

You're not saying that you've invented yet another flavour of "branding"?

 

Not like that, for being compatible with OE-A plugins we have our own branding module which is based on their sources but without remote/webif and extra things: https://github.com/O...branding-module

 

Also we have "getBoxBrand" inside enigma2 just like "getBoxType" which helps a lot to reduce the code and check so many things faster.

 

We have our own Open WebIF flavour supporting all brands/models: https://github.com/O...ionE2/OpenWebif

 

Just take a look at https://github.com/O.../data/rc_models and see how many remotes we provide but at last with less image size thanks to https://github.com/O...els/Makefile.am and brand detection.

 

enigma2-plugins: https://github.com/O...enigma2-plugins

 

tuxtxt: https://github.com/OpenVisionE2/tuxtxt

 

gstreamer1.0-plugin-multibox-dvbmediasink: https://github.com/O...ox-dvbmediasink

 

satip-client: https://github.com/O...E2/satip-client

 

servicehisilicon: https://github.com/O...ervicehisilicon

 

minidlna: https://github.com/O...sionE2/minidlna (which you're using it too)

 

libdvbsi: https://github.com/O...sionE2/libdvbsi

 

blindscan-s2: https://github.com/O...E2/blindscan-s2

 

dvbsnoop: https://github.com/O...sionE2/dvbsnoop (which you're using it too)

 

libdvbcsa: https://github.com/O...ionE2/libdvbcsa

 

And more ...

 

The thing is we're supporting more brands/models in our sources and we can't patch everything, instead we maintain our own.

 

There are so many disagreements, for example we don't like having multiple branches instead we use if conditions which you don't like.

[WanWizard]

The more if conditions you have, the more complex it will be to maintain it. You'll figure that out soon enough.

 

For OE-A compatibility, we inject a dummy, like so: https://github.com/O...xbranding.patch

 

I was just wondering if we will have to do this soon for all your stuff too.

 

Too bad the more people want to do their own thing instead of cooperate, the more complex things become, the less time is spend on new development, and the sooner it will all come to an end...

[Persian Prince]

I disagree, we use if conditions + MACHINE_FEATURES and it's easier than having multiple branches.

 

We support old API (dm800 for example), RaspberryPi and new OpenSSL and OE changes in enigma2 all with if conditions.

 

Injecting a dummy is another type of patching things and we're against patching. We do even maintain sh4 kernel and drivers with less patching. The branding module bring some good imports for us like "getDisplayType" which you can't get it with a dummy.

 

Anyway I don't want Backup Suite requires things like this I just wanted to inform you not get the updates till the green light.

 

We always contribute and stay friends, even I brought IanSave's time/date changes from ViX/Beyonwiz to Open Vision with his guid and soon I will send the PR for PLi.

 

If you check https://github.com/O...hs/contributors I'm #9 in contributors and even if you consider my commits useless (which could be as I'm still Senior Member here), did I ever give up sending PRs to PLi? No! As I see you all our friends even if I'm the youngest here because still I like learning from you.

 

I don't want to say I'm the good guy here but at least we try sending PRs upstream, here I want to thank ViX team for being friendly and contributing generously.

 

You should see how PurE2 team bring shame to this scene with their closed-source plugins even skins, they're using our sources with 0 contribution to open source.

[WanWizard]

Not sure what you mean by "multiple branches" vs "if's".

 

But if you mean the BSP's, it is a logical outcome from the fact we don't maintain those. Obviously, if you maintain them yourselfs you can put it all in one repo, and have one recipe per package for all hardware, instead of a recipe per vendor.

 

But we can't / won't, we want commitment from the vendor, instead of them throwing drivers over the fence, we do all the work, and they make money of it...

 

For the rest, Ok. clear. I've pinned it in develop.

[Persian Prince]

I mean enigma2 branches and patching other repos when we can do all with if conditions.

 

Also "athoik" and "betacentauri" have write access to Backup Suite repo so they can help with PLi compatibility if it was needed

[WanWizard]

If you have different build and release environments which you have to build and maintain simultaneously, like we do, you can't do without different branches, it's impossible.

 

If you only build "nighly's" from your current code base, obviously that isn't an issue.

 

And we don't "patch" other repo's, we have to patch OE-A plugins that use their box branding, so we can map it to our SystemInfo.

[littlesat]

Please stop using openatv like branding and use the simpel solution wanwizard invent to get around the openatv’s branding ‘crap’

[Persian Prince]

Please stop using openatv like branding and use the simpel solution wanwizard invent to get around the openatv’s branding ‘crap’

 

What do you mean?

 

Our "getBoxBrand" is inside enigma2 just like "getBoxType", even our "getBoxProc" is inside enigma2 but it's different.

[littlesat]

It is just stupid to invent another thing around this... as what we already have is sufficiant... invent other things without making proper agreements breaks compatibility. It should be ‘ours’ not ‘yours’...
I think it is a bit a pitty you forked and go on your own way instead of somehow work together.

Edited by littlesat, 13 September 2019 - 10:35.

[foxbob]

I test this "a9ec02a55887b46cdca1c93a921088b50cd7e0a3" on thud branch-not work.

 

Plugin version     = git467+a9ec02a

Back-up media      = /hdd

Filesystem                Size      Used Available Use% Mounted on

/dev/sda1               915.9G    679.6G    189.7G  78% /media/usb

Back-up = 2019.09.16_07:43

Version = openpli develop 

Flashed = 2019.09.15_18:43

Updated = 2019.09.15_23:33

Drivers = et-dvb-modules-kodi-et1x000 - 4.1.37+20170927-r2

Enigma2 = 2019-09-15-develop

Working directory  = /hdd/bi

Not Open Vision, Open PLi maybe?

Destination        = /hdd/fullbackup_et1x000/20190916_0743/et1x000

RECEIVER = GI ET11000 

ls: invalid option -- 'e'

BusyBox v1.29.3 (2019-09-15 13:11:35 UTC) multi-call binary.

 

I understand that this branch is only developing.

[betacentauri]

Don’t use thud branch! It’s work in progress.
And if you really want to use it, you should know how to fix it (just look into develop branch )

Edited by betacentauri, 16 September 2019 - 06:13.

[foxbob]

Now only this error, backup is created, but there is no final log.

Attached Files

•  1_0_16_70C_12_70_1680000_0_0_0_20190916175251.jpg   77.45KB   2 downloads

[betacentauri]

@foxbob: Sorry, it was too early this morning.

 

Please report problems with thud in other (own) thread. Most likely these problems have nothing really to do with the plugin. I guess something has changed in busybox which now makes the problems.

[betacentauri]

@Persion Prince: In latest busybox "ls -e" is not supported any longer. It was removed by busybox developers: https://github.com/m...85b31f515ca670b

Can you rewrite the plugin so that -e is not used?

[Persian Prince]

See: https://github.com/O...18e878ed7f564cc

 

I'm working on it