Bug #2638
Inverto IP-LNB hangs again
100%
Description
After 13 days of smooth usage ... there we go again ...
During a recording the iplnb showed the known symptom of hanging and no web access to the iplnb possible, only power shutdown helps. Failure occured between 14:37:01 and 14:38:04.
25.01.2015 14:37:01 ping_iplnb.py 6592 (tvheadend is running) returncode=0 wget 25.01.2015 14:38:04 ping_iplnb.py 6733 (tvheadend is running) returncode=4 wget
Different from before where the tvh status would show one or more "stalled entries" under stream and nothing at all under subscriptions, this time there is an entry under subscription (see screenshot) and nothing under stream.
And strange also that after stopping tvh by "sudo stop tvheadend" I don't see the "[ NOTICE]:STOP: Exiting HTS Tvheadend" entry in the log?
Also myterious entries in sub-millisec intervall about tune?
Full log attached
Files
History
Updated by Jaroslav Kysela almost 10 years ago
- Status changed from New to Fixed
- % Done changed from 0 to 100
Applied in changeset commit:tvheadend|cde875ecb2296a42adc909b5336d4027734bc637.
Updated by Jaroslav Kysela almost 10 years ago
The last "last tune diff" problem and the dead-lock and the tvh shutdown problem are fixed in v3.9-2427-gcde875e .
The IPLNB deadlock is similar like before:
2015-01-25 14:38:00.817 [ TRACE]:satip: Status string: 'ver=1.0;src=1;tuner=1,171,1,13,12421,h,dvbs,qpsk,,,27500,34;pids=0,1,16,17,18,5501,5502,5503,5504,5505,5506,5530' 2015-01-25 14:38:01.018 [ TRACE]:satip: Status string: 'ver=1.0;src=1;tuner=1,171,1,13,12421,h,dvbs,qpsk,,,27500,34;pids=0,1,16,17,18,5501,5502,5503,5504,5505,5506,5530' 2015-01-25 14:38:01.218 [ TRACE]:satip: Status string: 'ver=1.0;src=1;tuner=1,171,1,13,12421,h,dvbs,qpsk,,,27500,34;pids=0,1,16,17,18,5501,5502,5503,5504,5505,5506,5530'
... more status strings should be here .. between 14:38:01.218 and 2015-01-25 14:38:11.000 tvheadend didn't send any new command in this time
2015-01-25 14:38:11.000 [ TRACE]:subscription: 00BF: unlinking sub 0x2700a50 from svc 0x27683d0 2015-01-25 14:38:11.000 [ TRACE]:mpegts: table: mux 0x275e5a0 destroy pmt 02/FF (2) pid 159A (5530) 2015-01-25 14:38:11.000 [ DEBUG]:mpegts: 12421.5H in SATIP - close PID 159A (5530) [2/0x37bba30] 2015-01-25 14:38:11.000 [ TRACE]:mpegts: table: mux 0x275e5a0 free pmt 02/FF (2) pid 159A (5530) 2015-01-25 14:38:11.000 [ DEBUG]:mpegts: 12421.5H in SATIP - close PID 159A (5530) [1/0x27683d0] 2015-01-25 14:38:11.000 [ DEBUG]:mpegts: 12421.5H in SATIP - close PID 157D (5501) [1/0x27683d0] 2015-01-25 14:38:11.000 [ DEBUG]:mpegts: 12421.5H in SATIP - close PID 157E (5502) [1/0x27683d0] 2015-01-25 14:38:11.000 [ DEBUG]:mpegts: 12421.5H in SATIP - close PID 157F (5503) [1/0x27683d0] 2015-01-25 14:38:11.000 [ DEBUG]:mpegts: 12421.5H in SATIP - close PID 1580 (5504) [1/0x27683d0] 2015-01-25 14:38:11.000 [ DEBUG]:mpegts: 12421.5H in SATIP - close PID 1582 (5506) [1/0x27683d0] 2015-01-25 14:38:11.000 [ DEBUG]:mpegts: 12421.5H in SATIP - close PID 1581 (5505) [1/0x27683d0] 2015-01-25 14:38:11.000 [ DEBUG]:mpegts: 12421.5H in SATIP - stopping mux 2015-01-25 14:38:11.000 [ DEBUG]:satip: SAT>IP DVB-S Tuner #1 (10.0.0.71) - stopping 12421.5H in SATIP
Updated by Jaroslav Kysela almost 10 years ago
About the subscription entry in the status tab - DVR subsystem tries to subscribe service again and again in the DVR time window.. It seems like this reason, becase IPLNB was dead since 15:38 and the tvheadend was restarted at 14:45 - so the recording was probably active.
Updated by ullix tv almost 10 years ago
Well, that sure was fast!
Now running on 3.9.2427~gcde875e~precise
Is there any particularly good way to stress-test the iplnb for this problem?
Updated by ullix tv almost 10 years ago
- File tvh-2015-01-26.log.zip tvh-2015-01-26.log.zip added
It hung again!
26.01.2015 12:05:01 ping_iplnb.py 23065 (tvheadend is running) returncode=0 wget 26.01.2015 12:06:04 ping_iplnb.py 23083 (tvheadend is running) returncode=4 wget
Though I did put quite a bit of load on it with idle-scan and a bunch of recordings. Bitrate on the LAN topped at 150MBit/s.
log attached
Updated by Jaroslav Kysela almost 10 years ago
This hang was independent from the tvheadend handshake:
2015-01-26 12:06:01.516 [ TRACE]:satip: Status string: 'ver=1.0;src=1;tuner=1,152,1,12,10891,h,dvbs2,8psk,on,0.35,22000,23;pids=0,1,16,17,18,5300,5320,5330,5331,5332,5333,5334,5335,5336,5340,5350' 2015-01-26 12:06:01.516 [ TRACE]:satip: Status string: 'ver=1.0;src=1;tuner=5,147,1,12,12480,v,dvbs,qpsk,,,27500,34;pids=0,16,17,18,38,97,98,99,100,101,102,103,104,106,107,108,109,110,1535,1536' 2015-01-26 12:06:01.516 [ TRACE]:satip: Status string: 'ver=1.0;src=1;tuner=2,137,1,11,11361,h,dvbs2,8psk,on,0.35,22000,23;pids=0,1,16,17,18,6100,6110,6120,6121,6122,6123,6130,6131,6300,6400,6410,6420,6421,6422,6423,6430' 2015-01-26 12:06:01.516 [ TRACE]:satip: Status string: 'ver=1.0;src=1;tuner=4,171,1,13,12421,h,dvbs,qpsk,,,27500,34;pids=0,1,16,17,18,300,400,500,600,700,800,1200,1500,5500,5510,5520,5530,5540,5550' 2015-01-26 12:06:01.516 [ TRACE]:satip: Status string: 'ver=1.0;src=1;tuner=6,142,1,12,11493,h,dvbs2,8psk,on,0.35,22000,23;pids=0,1,16,17,18,5100,5101,5102,5103,5104,5105,5106' 2015-01-26 12:06:01.516 [ TRACE]:satip: Status string: 'ver=1.0;src=1;tuner=8,142,1,11,11302,h,dvbs2,8psk,on,0.35,22000,23;pids=0,1,16,17,18,107,108,109,110,111,116' 2015-01-26 12:06:01.516 [ TRACE]:satip: Status string: 'ver=1.0;src=1;tuner=7,147,1,12,11052,h,dvbs2,8psk,on,0.35,22000,23;pids=0,1,16,17,18,5400,5401,5402,5404,5406,5410,5420' 2015-01-26 12:06:09.000 [ TRACE]:mpegts: 11582.25H in SATIP - starting for 'scan' (weight 2, flags 0042)
No tvheadend activity between 12:06:01.516 and 12:06:09.000 . There should be more 'Status string:' lines which mean that the server is live and streaming. These UDP packets are sent every 200ms.
Updated by ullix tv almost 10 years ago
- File tvh-2015-01-27.log.zip tvh-2015-01-27.log.zip added
Another hang after only 6h of runtime, log attached
27.01.2015 14:34:01 ping_iplnb.py 11135 (tvheadend is running) returncode=0 wget 27.01.2015 14:35:04 ping_iplnb.py 11170 (tvheadend is running) returncode=4 wget
I see a whole bunch error-imagecache and discontiuity even well before tha hang. Anything relevant?
When I start and exit tvh normally. I see the
[ NOTICE]:STOP: Exiting HTS Tvheadend
Message at the end. But not so, when the IPLNB hangs. Some other issue elsewhere?