SekaPalle

 

...

Looking at Os*am log instead I've noticed a strange error - I think correlated to the channel change.

 

The error is the following:   "(dvbapi) Error: error on poll of 6 fd's (errno=2 No such  file or directory)"

 

Tomorrow I'll try to search about this... do you think this could be the problem at the root of the loose of control of the decoder?!?

...

 

Hi friends,

better reflecting on what I've wrote above and about my doubt regarding if the dvbapi could be the cause of the partial crash of the decoder, i'm quite sure that this error and others similars I've observed yesterday and today, always regarding dvbapi on channels my cards normally decripts, are a consequence of the problem rather than the cause.

 

Pr2 is supposing it's a consequence of poor memory of my decoder, but dolps above said he has an ET10000 (if my memory don't fail... comparable to my ET8500) functioning without any issues, even if he doesn't specify clearly if using openvpn and what else.

WanWizard thinks probably is due to a bad config of my VPN tunnel, but I don't know what to look at and where.

 

So... I don't know what to test or try.

 

Thanks for the help

 

Hi friends,
I come back to this my old post to inform that my problem was solved.

Suspecting that Pr2 had correctly found the root of my problem, locating it in a lack of memory of my old Xtrend ET8500 related to greater memory needs of OpenPLI 8.1, I've tried to use another my decoder with a larger memory as server.

I've installed OpenPLI on a Mutant HD51 and afterwards i configured on it, as server, OpenVPN and Oscam, mantaining more or less the same configs I used on Xtrend. I had to resume an old smargo too, because the Mutant has only 1 card reader.

The result was positive and everithing worked like a charm ... effectively It was a poor memory problem of my old decoder

So I've decided to by a new decoder selecting a VU+ Duo 4K SE.

I've got some troubles discovering the new stb had an FBC DVB-S2 tuner defective, but when I've received the substitutive stb, after a certain workaround regarding tuners configuration, I'got it working perfectly and now I feel happy... no moore sudden crash of enigma2 services .

So thank a lot to Pr2, but also to WanWizard and every friend who try helping the others.

Best regards

Hi friends, I'm still looking to focus the problem on my decoder, but with poor progress.

I would like to try transferring the role of server (and various configs) from my old Xtrend ET-8500 to a new and more performant STB so to exclude the problem is in a my misconfiguration of the old decoder.

Anyway I need another decoder so this could be the manner, I hope, to obtain a double result.

I will ask you a hint about a new decoder to buy. I suppose to look for:

• a quite recent model, still produced, mantained and findable on sale;
• a product with 2 card-readers (I've got 2 cards) and the capability to have installed an HD inside;
• a model with a certain amount of memory, a modern processor and working well with OpenPLI.

I've recently bought a GigablueTrio 4K but I evaluate bad its RCU, I don't like the necessity to use an external HD with USB interface in a stb with a lot of unused space inside, just one card reader, a poor led display etc.

I've seen some GigaBlue UHD Quad 4K on sale but I think this is not a very recent model  (on market since 2018?). If haven't misunderstood there is also the problem Gigablue doesn't colaborate with OpenPLI development so this isn't a good thing for the decoder's future support.

Something newer and possibly better?!?

Thanks a lot for yours tips.

...

Looking at Os*am log instead I've noticed a strange error - I think correlated to the channel change.

 

The error is the following:   "(dvbapi) Error: error on poll of 6 fd's (errno=2 No such  file or directory)"

 

Tomorrow I'll try to search about this... do you think this could be the problem at the root of the loose of control of the decoder?!?

...

Hi friends,

better reflecting on what I've wrote above and about my doubt regarding if the dvbapi could be the cause of the partial crash of the decoder, i'm quite sure that this error and others similars I've observed yesterday and today, always regarding dvbapi on channels my cards normally decripts, are a consequence of the problem rather than the cause.

Pr2 is supposing it's a consequence of poor memory of my decoder, but dolps above said he has an ET10000 (if my memory don't fail... comparable to my ET8500) functioning without any issues, even if he doesn't specify clearly if using openvpn and what else.

WanWizard thinks probably is due to a bad config of my VPN tunnel, but I don't know what to look at and where.

So... I don't know what to test or try.

Thanks for the help

These evening the problem has come back, after almost two days of correct functioning.

This happened although the creation of 1 Gb swap file on the attached pendrive.

The executed operation that formally gave origin to E2 hung-up was a simple change of channel remaining in the same bouquet.

Connecting with telnet to the decoder I've seen it's alive (command ps shows the various processes in execution) and looking at Os*am Webif I've seen it is still functioning and decoding.

Looking at Os*am log instead I've noticed a strange error - I think correlated to the channel change.

The error is the following:   "(dvbapi) Error: error on poll of 6 fd's (errno=2 No such  file or directory)"

Tomorrow I'll try to search about this... do you think this could be the problem at the root of the loose of control of the decoder?!? It becomes irresponsive to the RCU and I've to command a reboot from telnet or to bring away the power through the switch in the rear side of stb.

@WanWizard:

If you don't have split tunneling active, then all local traffic on both locations will work, and traffic between the two boxes will work, but the client box will not have any connection beyond that, as all traffic will be routed into the VPN tunnel instead of to the local router.

I thinked it was the opposite, that is, with split tunnel the two decoders in different LANs uses the tunnel just to speak eachother (Os*am comunication protocols in my case), meanwhile for normal traffic they use the local resources they see (their local default gateway, DNS servers etc.).

I preferred my remote decoder doesn't use the tunnel to navigate forward the WEB, even if it's unusual for me such use of decoder.

My openVPN tunnel works in this way (I think :-) ).

Thankyou for your appreciated help

Hi WanWizard,

thanxs for your contribute too.

I don't know what's nevil's case, so I just speak about my situation. I'm not a networking expert and I've got my troubles in making my self-configured openvpn, although it was the simpliest possible configuration (just a server with one client).

My only requirement was to share TV programs among family members, between two places in which we live, so I haven't used routers to built my VPN but I choose the solution to construct it directly using two my decoders.

In the udp tunnel just passes protocols used by oscam and nothing more. Each of the two decoders uses the resorces of the LAN they belongs (to navigate, to stream to the tablet, ecc.) and they only see each other with no access to other members of respective LANs.

I think I could define it a sort of split tunnel if you (@WanWizard) were intending something similar.

I don't know how and where to look for something wrong in my configs, but it seems strange to me thinking it's due to a network problem since, as I've explained in my first post, both openvpn tunnel and oscam decoding still work correctly, assuring vision to my remote decoder.

It seems to me more reasonable it could be a consequence of insufficient memory, as supposed by Pr2.

After his tip regarding a memory problem, I kept on controlling memory status ot the decoder through OpenPLI's Main menù > Informations > Systems Informations > Memory Informations and I realized that sometimes (I don't know the reason) the information I've reported at post #3 above, regarding the memory state of my decoder, are not always true.

In fact I've seen situations in which MemFree has drastically fallen to 3600-4900 Kb.

I can't undestand the meaning of MemAvailable which still remains at high values 211000 - 215000 Kb.

Anyway yesterday in late evening I made what Pr2 suggested to me: I put a 2 Gb pendrive in the decoder and with the plugin "swap-manager" I've given a present to my Xtrend activating a 1 Gb swapfile... since than the decoder is well functioning... I'll keep monitoring it in next days.

The usual menu I spoke above reports a SwapTotal of 1048572 Kb and an equal SwapFree.

I ask you: before to use the swapfile the decoder ha to reach MemFree = 0 Kb ?!?!?

Thankyou very much for your help guys. Have a good night