Jump to content


Photo

OP 7.2, 2 crashs yesterday


  • Please log in to reply
39 replies to this topic

Re: OP 7.2, 2 crashs yesterday #21 Pr2

  • PLi® Contributor
  • 6,046 posts

+256
Excellent

Posted 21 January 2020 - 13:18

Connect in telnet to your box, issue a: 

init 4

Search for the epg.dat file:   

find -name epg.dat

Remove the local: epg.dat:

rm /path/to/file/epg.dat

If you mount another box HDD on your network don't delete this epg.dat you need to delete the one from your box.

Then type

reboot

NO SUPPORT by PM, it is a forum make your question public so everybody can benefit from the question/answer.
If you think that my answer helps you, you can press the up arrow in bottom right of the answer.

Wanna help with OpenPLi Translation? Please read our Wiki Information for translators

Sat: Hotbird 13.0E, Astra 19.2E, Eutelsat5A 5.0W
VU+ Solo 4K: 2*DVB-S2 + 2*DVB-C/T/T2 (used in DVB-C) & Duo 4K: 2*DVB-S2X + DVB-C (FBC)

AB-Com: PULSe 4K 1*DVB-S2X (+ DVB-C/T/T2)
Edision OS Mio 4K: 1*DVB-S2X + 1*DVB-C/T/T2
 


Re: OP 7.2, 2 crashs yesterday #22 Raybobo

  • Senior Member
  • 190 posts

+1
Neutral

Posted 24 January 2020 - 05:32

epg.dat deleted, but still sometimes a crash, 1 or 2 times a day...


LG Oled C9

Vu+Duo4KSe

Yamaha 5.1 system

Qnap NAS 2Tb

1 wife

2 childs

2 dogs

4 cats :D


Re: OP 7.2, 2 crashs yesterday #23 WanWizard

  • PLi® Core member
  • 68,312 posts

+1,719
Excellent

Posted 24 January 2020 - 14:42

I think you should stop focussing on the EPG, Littlesat has experienced an issue there once, and now everything is caused by that issue... ;)

 

You said you had it when moving a channel in the list, and when looking at the movielist. Can I assume the crashlog enclosed is of the second crash, not the first?

 

The logs shows corruption of a recordings cue sheet, which might be the cause of that crash. Unfortunately, because it happened in compiled code, there is no stack trace in the logs to pinpoint the location of the problem.

 

edit: Ah, probably not, "something is wrong with this cueSheetEvent calculation" is generated when that issue is detected when you skip through a recording, not when displaying the movielist.


Edited by WanWizard, 24 January 2020 - 14:44.

Currently in use: VU+ Duo 4K (2xFBC S2), VU+ Solo 4K (1xFBC S2), uClan Usytm 4K Pro (S2+T2), Octagon SF8008 (S2+T2), Zgemma H9.2H (S2+T2)

Due to my bad health, I will not be very active at times and may be slow to respond. I will not read the forum or PM on a regular basis.

Many answers to your question can be found in our new and improved wiki.


Re: OP 7.2, 2 crashs yesterday #24 WanWizard

  • PLi® Core member
  • 68,312 posts

+1,719
Excellent

Posted 24 January 2020 - 14:51

@littlesat,

 

It crashes in: _ZN8eBouquet12flushChangesEv, which is https://github.com/O...dvb/db.cpp#L126

 

which is when writing the changes made to the bouquet to the bouquet file. What interfaces with Python there, eServiceReference?


Currently in use: VU+ Duo 4K (2xFBC S2), VU+ Solo 4K (1xFBC S2), uClan Usytm 4K Pro (S2+T2), Octagon SF8008 (S2+T2), Zgemma H9.2H (S2+T2)

Due to my bad health, I will not be very active at times and may be slow to respond. I will not read the forum or PM on a regular basis.

Many answers to your question can be found in our new and improved wiki.


Re: OP 7.2, 2 crashs yesterday #25 littlesat

  • PLi® Core member
  • 56,123 posts

+685
Excellent

Posted 24 January 2020 - 15:00

Most likely called via channelselector and servicelist.py... Then something is currupt within a userbouquet list... some settings editors might make trouble here.

Frans also reported such thinks from the latest hansat settings.

 

This functions writes a userbouquet.... (the flush means write a userbouquet).

 

-BUT-

 

I'm also 100% sure I had the same issue here during a flash userbouquet where I confirmed it was triggered due to a corrupted memory due to EPG stuff... But when we can exclude this I better go for this comment in this thread

 

 


Yes, I do, from a script, but I have created my favorites list with E-Channelizer.

 

Maybe the script does create invalid characters (utf-8 stuff?) so this flush bouquets code 'crashes'....


Edited by littlesat, 24 January 2020 - 15:04.

WaveFrontier 28.2E | 23.5E | 19.2E | 16E | 13E | 10/9E | 7E | 5E | 1W | 4/5W | 15W


Re: OP 7.2, 2 crashs yesterday #26 WanWizard

  • PLi® Core member
  • 68,312 posts

+1,719
Excellent

Posted 24 January 2020 - 15:03

eServiceReference uses string.length() to create a buffer, but length() is not multibyte safe?

 

@Raybobo, any non-ASCII characters in the bouquet name of the bouquet you were trying to edit?


Currently in use: VU+ Duo 4K (2xFBC S2), VU+ Solo 4K (1xFBC S2), uClan Usytm 4K Pro (S2+T2), Octagon SF8008 (S2+T2), Zgemma H9.2H (S2+T2)

Due to my bad health, I will not be very active at times and may be slow to respond. I will not read the forum or PM on a regular basis.

Many answers to your question can be found in our new and improved wiki.


Re: OP 7.2, 2 crashs yesterday #27 littlesat

  • PLi® Core member
  • 56,123 posts

+685
Excellent

Posted 24 January 2020 - 15:10

RESULT eBouquet::flushChanges()
{
                std::string filename = eEnv::resolve("${sysconfdir}/enigma2/" + m_filename);
                {
                               CFile f((filename + ".writing").c_str(), "w");
                               if (!f)
                                               goto err;
                               if ( fprintf(f, "#NAME %s\r\n", m_bouquet_name.c_str()) < 0 )
                                               goto err;
                               for (list::iterator i(m_services.begin()); i != m_services.end(); ++i)
                               {
                                               eServiceReference tmp = *i;
                                               std::string str = tmp.path;
                                               if ( fprintf(f, "#SERVICE %s\r\n", tmp.toString().c_str()) < 0 )
                                                               goto err;
                                               if ( i->name.length() )
                                                               if ( fprintf(f, "#DESCRIPTION %s\r\n", i->name.c_str()) < 0 )
                                                                               goto err;
                               }
                               f.sync();
                }
                rename((filename + ".writing").c_str(), filename.c_str());
                return 0;
err:
                eDebug("[eBouquet] couldn't write file %s", m_filename.c_str());
                return -1;
}

My knowledge to cpp is not that good to check if tmp.toString().c_str or i->name.c_str or something else might cause issues here when something is weird in userbouquets?


WaveFrontier 28.2E | 23.5E | 19.2E | 16E | 13E | 10/9E | 7E | 5E | 1W | 4/5W | 15W


Re: OP 7.2, 2 crashs yesterday #28 littlesat

  • PLi® Core member
  • 56,123 posts

+685
Excellent

Posted 24 January 2020 - 15:10

check on i->name.length() OMG can go wrong... better check on !i->name.empty() ???


Edited by littlesat, 24 January 2020 - 15:12.

WaveFrontier 28.2E | 23.5E | 19.2E | 16E | 13E | 10/9E | 7E | 5E | 1W | 4/5W | 15W


Re: OP 7.2, 2 crashs yesterday #29 WanWizard

  • PLi® Core member
  • 68,312 posts

+1,719
Excellent

Posted 24 January 2020 - 15:11

eServiceReference.toString() is called there, and that does:

ret.reserve((6 * sizeof(data)/sizeof(*data)) + 8 + path.length() + name.length()); /* Estimate required space */

what if path or name are UTF8 multibyte?


Currently in use: VU+ Duo 4K (2xFBC S2), VU+ Solo 4K (1xFBC S2), uClan Usytm 4K Pro (S2+T2), Octagon SF8008 (S2+T2), Zgemma H9.2H (S2+T2)

Due to my bad health, I will not be very active at times and may be slow to respond. I will not read the forum or PM on a regular basis.

Many answers to your question can be found in our new and improved wiki.


Re: OP 7.2, 2 crashs yesterday #30 littlesat

  • PLi® Core member
  • 56,123 posts

+685
Excellent

Posted 24 January 2020 - 15:13

I'm afraid we're discovering here a >15 years old bug....

'estimate required space' WTF


Edited by littlesat, 24 January 2020 - 15:13.

WaveFrontier 28.2E | 23.5E | 19.2E | 16E | 13E | 10/9E | 7E | 5E | 1W | 4/5W | 15W


Re: OP 7.2, 2 crashs yesterday #31 WanWizard

  • PLi® Core member
  • 68,312 posts

+1,719
Excellent

Posted 24 January 2020 - 15:13

No, can't be it:
 

Note that string objects handle bytes without knowledge of the encoding that may eventually be used to encode the characters it contains. Therefore, the value returned may not correspond to the actual number of encoded characters in sequences of multi-byte or variable-length characters (such as UTF-8).

 

which is fine for determining buffer size...

 

What else access Python data structures there?


Currently in use: VU+ Duo 4K (2xFBC S2), VU+ Solo 4K (1xFBC S2), uClan Usytm 4K Pro (S2+T2), Octagon SF8008 (S2+T2), Zgemma H9.2H (S2+T2)

Due to my bad health, I will not be very active at times and may be slow to respond. I will not read the forum or PM on a regular basis.

Many answers to your question can be found in our new and improved wiki.


Re: OP 7.2, 2 crashs yesterday #32 littlesat

  • PLi® Core member
  • 56,123 posts

+685
Excellent

Posted 24 January 2020 - 15:14

:(


WaveFrontier 28.2E | 23.5E | 19.2E | 16E | 13E | 10/9E | 7E | 5E | 1W | 4/5W | 15W


Re: OP 7.2, 2 crashs yesterday #33 WanWizard

  • PLi® Core member
  • 68,312 posts

+1,719
Excellent

Posted 24 January 2020 - 15:18

Lets wait and see if the TS can provide more details on what exactly the situation is and what he was doing.


Currently in use: VU+ Duo 4K (2xFBC S2), VU+ Solo 4K (1xFBC S2), uClan Usytm 4K Pro (S2+T2), Octagon SF8008 (S2+T2), Zgemma H9.2H (S2+T2)

Due to my bad health, I will not be very active at times and may be slow to respond. I will not read the forum or PM on a regular basis.

Many answers to your question can be found in our new and improved wiki.


Re: OP 7.2, 2 crashs yesterday #34 littlesat

  • PLi® Core member
  • 56,123 posts

+685
Excellent

Posted 24 January 2020 - 15:31

Here flushchanged is used...

 

https://github.com/O..._q=flushChanges

 

in python only here lib/python/Screens/ChannelSelection.py


WaveFrontier 28.2E | 23.5E | 19.2E | 16E | 13E | 10/9E | 7E | 5E | 1W | 4/5W | 15W


Re: OP 7.2, 2 crashs yesterday #35 WanWizard

  • PLi® Core member
  • 68,312 posts

+1,719
Excellent

Posted 24 January 2020 - 15:34

The crash happens in flushChanges(), when accessing a Python datastructure. Unfortunately there isn't more info in that crash log...


Currently in use: VU+ Duo 4K (2xFBC S2), VU+ Solo 4K (1xFBC S2), uClan Usytm 4K Pro (S2+T2), Octagon SF8008 (S2+T2), Zgemma H9.2H (S2+T2)

Due to my bad health, I will not be very active at times and may be slow to respond. I will not read the forum or PM on a regular basis.

Many answers to your question can be found in our new and improved wiki.


Re: OP 7.2, 2 crashs yesterday #36 littlesat

  • PLi® Core member
  • 56,123 posts

+685
Excellent

Posted 24 January 2020 - 15:35

Now wait for more data... first thing is we need to verify the crash...


WaveFrontier 28.2E | 23.5E | 19.2E | 16E | 13E | 10/9E | 7E | 5E | 1W | 4/5W | 15W


Re: OP 7.2, 2 crashs yesterday #37 Raybobo

  • Senior Member
  • 190 posts

+1
Neutral

Posted 25 January 2020 - 05:49

Raybobo, any non-ASCII characters in the bouquet name of the bouquet you were trying to edit?

 

No, bouquet already there, with standard name, Only move channels in it.

 

Now wait for more data... first thing is we need to verify the crash...

 

I will find the last log and send it asap. At the moment, Duo2 is off.


LG Oled C9

Vu+Duo4KSe

Yamaha 5.1 system

Qnap NAS 2Tb

1 wife

2 childs

2 dogs

4 cats :D


Re: OP 7.2, 2 crashs yesterday #38 WanWizard

  • PLi® Core member
  • 68,312 posts

+1,719
Excellent

Posted 25 January 2020 - 13:27

Ok, so that can't be the cause then...


Currently in use: VU+ Duo 4K (2xFBC S2), VU+ Solo 4K (1xFBC S2), uClan Usytm 4K Pro (S2+T2), Octagon SF8008 (S2+T2), Zgemma H9.2H (S2+T2)

Due to my bad health, I will not be very active at times and may be slow to respond. I will not read the forum or PM on a regular basis.

Many answers to your question can be found in our new and improved wiki.


Re: OP 7.2, 2 crashs yesterday #39 40H3X

  • Forum Moderator
    PLi® Contributor
  • 5,784 posts

+186
Excellent

Posted 25 January 2020 - 14:04

 


Search for the epg.dat file:   

find -name epg.dat

 

Btw it should better be..

find / -name epg.dat

...than you are searching from the root and up, so practically everywhere  ;)


Edited by 40H3X, 25 January 2020 - 14:06.

Hardware: Vu+ Uno 4K SE - Vu+ Duo 4K  - Fuba 78 cm - Tripleblock LNB Quad 19.2/23.5/28.2 - DS918+
Software : OpenPLi - OSCam - Settings van Hans - Autotimer - EPGImport

---------------------------------------------------------------------------------------------------------------------------------------

Remember: Upvote with the rep_up.png button for any user/post you find to be helpful, informative, or deserving of recognition!

---------------------------------------------------------------------------------------------------------------------------------------

Many answers to your question can be found in our new and improved wiki

Note: I do not provide support via PM !.


Re: OP 7.2, 2 crashs yesterday #40 Pr2

  • PLi® Contributor
  • 6,046 posts

+256
Excellent

Posted 26 January 2020 - 11:05

Correct I should have specify it. Thanks 40H3X.

NO SUPPORT by PM, it is a forum make your question public so everybody can benefit from the question/answer.
If you think that my answer helps you, you can press the up arrow in bottom right of the answer.

Wanna help with OpenPLi Translation? Please read our Wiki Information for translators

Sat: Hotbird 13.0E, Astra 19.2E, Eutelsat5A 5.0W
VU+ Solo 4K: 2*DVB-S2 + 2*DVB-C/T/T2 (used in DVB-C) & Duo 4K: 2*DVB-S2X + DVB-C (FBC)

AB-Com: PULSe 4K 1*DVB-S2X (+ DVB-C/T/T2)
Edision OS Mio 4K: 1*DVB-S2X + 1*DVB-C/T/T2
 



1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users