Java errors keep causing Sage to lose connection

[dvd_maniac]

I have a few files in my SageTV folder evertime Sage loses connection with the SageService. An example of the filenames (hs_err_pid2320.log).
The files are created at the very time it happens and when I open them I get the following:

#
# An unexpected error has been detected by Java Runtime Environment:
#
# Internal Error (nmethod.cpp:1720), pid=2320, tid=5704
# Error: guarantee(cont_offset != 0,"unhandled implicit exception in compiled code")
#
# Java VM: Java HotSpot(TM) Client VM (11.0-b16 mixed mode, sharing windows-x86)
# If you would like to submit a bug report, please visit:
# http://java.sun.com/webapps/bugreport/crash.jsp

I have the latest Java (1.6.0_11 along with Sage 6.5.4
Are others having the same issue?

[ntisdale]

I too am having the same problem. The files are generated by the java run-time when a java program crashes. If you read through the file it gives information about which thread of the java program caused the crash. My problem is that everytime sage crashes a different thread seems to be the culprit which is a nightmare in terms of tracking down the cause of the crashes. Anyhow, I would recommend you file a bug report if you haven't already done so and pass the logs onto the sage team in order to help them track down the source of the errors.

Neil

[jasmithvr6]

Originally Posted by dvd_maniac

I have a few files in my SageTV folder evertime Sage loses connection with the SageService. An example of the filenames (hs_err_pid2320.log).
The files are created at the very time it happens and when I open them I get the following:

#
# An unexpected error has been detected by Java Runtime Environment:
#
# Internal Error (nmethod.cpp:1720), pid=2320, tid=5704
# Error: guarantee(cont_offset != 0,"unhandled implicit exception in compiled code")
#
# Java VM: Java HotSpot(TM) Client VM (11.0-b16 mixed mode, sharing windows-x86)
# If you would like to submit a bug report, please visit:
# http://java.sun.com/webapps/bugreport/crash.jsp

I have the latest Java (1.6.0_11 along with Sage 6.5.4
Are others having the same issue?

I'm having the same issue running an HD-200 via Sage Server. When I lose connection at the extender level, it's because Sage is no longer running on the Server. I need to restart Sage on the Server and then it works fine. I'm running Sage 6.5.4 and Java 1.6_11. Any insight?

[jasmithvr6]

Any resolution to this? System continues to crash every 3 or 4 days.

[stuckless]

These types of errors are usually bugs in the JVM (or native libraries that it loads). I'd try to use a different version of the JVM and see if the issues persist. The log files that are generated, can actually tell you quite a bit of information, if you know how to read them (i don't).

[msm]

If you look further down in the error log my guess is you will some reference to the actual Java code that causes this. I had one when resuming from sleep to around the setting up of tuning and capture devices. I raised a bug and haven't tried 6.5.8 yet but I am guessing it it not yet resolved as I can see nothing in the change log that would be realted.

You can mitigate the issue slightly by setting the sageTV service to restart after a crash. This will wait a min and then restart. If a minute is too long you can set up a short bat script and do the same (at least in vista). Mine waits about 8 secs and then restarts and since mine only happens after resuming from sleep this works for me.

[Graygeek]

msm,

What you describe fits a crash I see to a "T". Was this JAVA error related to "seeker"? If so I'll submit a bug report also to maybe help push the point.

[bmac.to]

What version of Java does Sage recommend for running SageTV? To put it another way, what version of Java is tested when they release a new version of SageTV? If it is version 11, is it version 11 on all platforms (ie. XP, Vista, Vista x64, etc)?

Perhaps that information could help fix these issues.

Thanks.
________
Ford mainline history

[msm]

I cant quite remember. I delte all my error files prior to a new version os I can easily monitor when it has happened and it hasn't happened in 6.5.8 yet. If you post your error log I will take and look and tell you if it is the same.

Matt

[Graygeek]

msm,

Here is my error log info, the trigger for this is identical to your description.

http://forums.sagetv.com/forums/showthread.php?t=38949

Thanks, Tom

[msm]

Yep thats the one. What cards do you have in your system?

EIDT: I see from your sig that we do not have any cards in common. I have 2 DVB-T and a DVB-S2 card (all hauppauge). I am actually trying out a pinnacle 7010i at the moment but also not letting my system sleep so it can serve as a slimserver setup.
Matt

[srothwell]

I don't know if this is related but about 4 days ago, my wife reported that all the digital channels said NO SIGNAL.

I have 4 HDHRs.

I had to stop the Sage Service and restart and it was fine again. Honestly I didn't know what to look for in the log. Are the examples you guys gave earlier in the log or an error on the screen?

Stacy

[jphipps]

I been having the same issue for a while... The JVM would run out of memory, and sage would stop responding and I would have to cycle it for it to work again. I have raised the JVM heap size to 768 and put a schedule task in to cycle sage every night at 02:30 and it has been stable for a few weeks. It just locked up today and the heap was at 768. I am going to bump it up to 1024 and keep the daily cycle...

Thanks,
Jeff

[DaveS]

Originally Posted by srothwell

I don't know if this is related but about 4 days ago, my wife reported that all the digital channels said NO SIGNAL.

I have 4 HDHRs.

I had to stop the Sage Service and restart and it was fine again. Honestly I didn't know what to look for in the log. Are the examples you guys gave earlier in the log or an error on the screen?

Stacy

I've been having the same issues since I built a new computer a couple of weeks ago and loaded the latest Java (1.6, I think.) My old computer used the previous version (1.5, I think.) Both computers were running AMD processors and both used the same video card (NVidia 8600GT) and the same tuners (ATI Theaterpro 550 & Plextor TV402). My old computer never crashed. Could it be the latest version of Java is the issue?

[Graygeek]

Originally Posted by msm

Yep thats the one. What cards do you have in your system?

EIDT: I see from your sig that we do not have any cards in common. I have 2 DVB-T and a DVB-S2 card (all hauppauge). I am actually trying out a pinnacle 7010i at the moment but also not letting my system sleep so it can serve as a slimserver setup.
Matt

Did you have any luck with the newer version of SAGE and this error. I reworked my server thinking the issue was related to M780 Tuners I had, I still have the JAVA error related to seeker and I find no relationship to specific tuners. Have kicked two errors in two days when recording starts after wake from sleep. I guess it is finally time to ping Sage support.

[msm]

Nope, it still seems to happen. I have gone back to my hauppauge cards for reliability.

If your service actually fails then it is simple to automatically restart it. I have actually written a script that waits about 8 secs and then restarts so I do not get much inconvienience (the miniumum windows recovery restart time is 1 minute and this felt too long).

It only happens after a deep sleep (as the log calls it) and my guess is that it is trying to use the tuner before it is fully initialised, but that is just a hunch.

Matt

[Graygeek]

Yes, I have Sage Service set to restart, that is a savior. I have had it occur even an hour after resume from sleep, i.e. tuner #1 is recording, then tuner #2 will start and ZAP! JAVA error .... this creates a gap in the first tuners recording. Not good for WAF. I've played with system waking up several minutes prior to tuning and still get the error. It does almost seem to happen more with some tuners than others, but then it may only be a more used tuner. I did sent in a trouble report last night so we will see what is next. I really hate to give up on S3 Sleep an leave system on 24/7 .... getting "Green" in my old age and want to save power, plus it is that much less dust & grap being sucked in by cooling fans. S3 is working great ... except for this on bug.

Tom aka Graygeek

[jasmithvr6]

Originally Posted by Graygeek

Yes, I have Sage Service set to restart, that is a savior. I have had it occur even an hour after resume from sleep, i.e. tuner #1 is recording, then tuner #2 will start and ZAP! JAVA error .... this creates a gap in the first tuners recording. Not good for WAF. I've played with system waking up several minutes prior to tuning and still get the error. It does almost seem to happen more with some tuners than others, but then it may only be a more used tuner. I did sent in a trouble report last night so we will see what is next. I really hate to give up on S3 Sleep an leave system on 24/7 .... getting "Green" in my old age and want to save power, plus it is that much less dust & grap being sucked in by cooling fans. S3 is working great ... except for this on bug.

Tom aka Graygeek

Doesn't appear to be tuner card related. I have two HVR-1600's, but Sage Server will crash both while recording or just powering up from sleep mode with no scheduled recordings. I rolled back from Java 1.6_11 to Java 1.6_10 to see if that helped any, but no such luck. I may try rolling back to 1.5 just for grins. Sage Server crashes at least every other day if not more...which is causing serious WAF.

[Graygeek]

I have been emailing support on this, they asked I update to 6.5.9 and reinstall JAVA, (the link pionted to update 12?)

Anyhow, 6.5.9 broke playback of live & recorded content sourced from my HAVA tuner via my HD100 extenders, so I had to roll back to 6.5.7. Once the dust settles again I'll reload JAVA update 12 and see if that helps.

[msm]

I have now updated to 6.5.9 (the release version) and java u12 and will report if I see this again with this setup. Only had it for a day but no problems yet after a deep sleep.