cancel manual recording?

[pBS]

well, did a cleanup of db, and looks like it's fixed it completely..
i emptied both rec schedule and programme tables and re-imported recordings and epg and now it's zippy and i can quick-record multiple files no problem...

the cancelling of in-progress still takes 2 tries, but it goes away right away on second try and no long pauses.... [and no second file cuz it didn't have time to make one]
i bet it was a bad character or entry in programme...probably because i changed epg sources since install...[plus all the fiddling that goes with doing that]

i just rebooted so will see how it goes after being on a long time and recording a bunch...
but so far so good...back to normal..
i made a batch file to do all the cleanup chores serially so i will share it if all goes well...
kinda like a refresh of db without losing anything...

one note, i noticed if i do a empty epg from config and exit, all my recordings are gone too! i thought it didn't delete programme info for existing recordings, so glad i backed up!
[i guess it just clears the whole table]
it mentions in the warning about losing pending, but not existing...
tho that may have been a glitch on my end, but i tested all on another machine...then first thing i did was empty epg on real pvr, and i saw all my rec info was gone...hadn't even started messing with that db yet...will see if i can confirm that..

[pBS]

i looked thru my old db and found an entry with a blank programme_oid....[a manual recording]
after the cleanup it was gone...so i think that was the problem...not sure how it got in there but it was probably me..
almost finished with refresher bat...just working out how to save/restore/adjust the prog info for reoccurrings..
i'm trying to 'defrag' the existing/reoccurr. programee info so my db is nice and tight and clean.. [no huge numbers or gaps]
working nicely so far...

[pBS]

still having same problem...it's faster but still has problems committing data in a timely fashion...definintely the recording service as when i stop it, things happen quickly again..
[and no errors in logs]
just got my dg45ID motherboard, so will be doing completely fresh rebuild in next week..
[and a 50$ slim case that's smaller then my vcr! ]

wow, this board is very efficient, fan runs at about 50!rpm and still keeps a c2d7200 cool..
and hd playback within 5-10% tops...so it is accellerating more..

also found a Atom based mini barebones from Shuttle [X27] for 189$ total..
[about the size of a mac mini and excellent looking with gigabit]

[fvfv]

Pbs, I am beginning to show the same problems with cancel recordings or just about anything related to db and SQlite. PVRX2 just hangs and needs to be killed from task manager. Same with the config app.

What tool did you use to clean your database ?

We need to have some sort of Vacuum utility that may be can run from a task scheduler.

[pBS]

well, i did mine manually kinda....but the problem still exists, just a clean database doesn't mind as much...it's odd, only second quick record is slow, after that, they seem to go pretty quickly like normal till you get to around 10 or so...then slowdown again..
doesn't seem to be epg data related actually..just looks faster with clean db..

so yea, seems like a commit isn't being done at some point,hanging the db till it times-out..
i'll look at the sql logs again to see if there isn't something missing in the problem areas..

[pBS]

more info, seems to happen *only* when quick scheduling shows adjacent to eachother..
so seems like conflict resolution is slowing things up...[the parts where it adjusts padding,looks for conflicts,picks tuner]

i can go in and schedule shows that don't 'touch' several in a row without slowdowns..
but first one i do that 'touches' another, it slows... [and it doesn't get scheduled]

[BitterBastard]

I was hoping to bump this thread because I have a lot of similar problems. I get database lockups a lot when try to schedule manual recordings. I'm currently working through an issue with GraphRecorder, but in my testing I schedule two manual recordings in close proximity (but not back-to-back). Because of my GR issue, the second recording never proceeds, but I also can't ever delete it from the list of Pending recordings (or I should say, I can't cancel it). When I reboot, the recording is gone, and the logs always show a db lock-up issue.

But basically, I just wanted to know if pBS ever solved his problem.

[dennit]

I've recently injcreased the size of my database by adding hundreds of channels on dvb-s. I'm unable to cancel scheduled recordings. The logs say the db is locked.

Code:

2008-12-02 17:41:52.973    VERBOSE    [5]    Received request to cancel recording: 750
2008-12-02 17:41:52.973    VERBOSE    [5]    Now cancelling recording: 750
2008-12-02 17:41:52.973    VERBOSE    [5]    Stopping recording oid:3
2008-12-02 17:41:53.269    VERBOSE    [3]    Recording cancelled
2008-12-02 17:41:53.379    VERBOSE    [5]    RecordingStatus is now RecordingStatus.RECORDING_COMPETED
2008-12-02 17:41:53.379    VERBOSE    [5]    RecordingFactory.save()
2008-12-02 17:42:23.566    ERROR    [5]    Unabled to save ScheduledRecording: The database file is locked
database is locked
2008-12-02 17:42:23.582    ERROR    [5]       at System.Data.SQLite.SQLite3.Reset(SQLiteStatement stmt)
   at System.Data.SQLite.SQLite3.Step(SQLiteStatement stmt)
   at System.Data.SQLite.SQLiteCommand.ExecuteNonQuery()
   at GBPVR.Backend.Common.x396614b3dae3b1da.xc979432fcc5a510a(ScheduledRecording xe41d2b9d98dc1269)

This thread looks related.

[sub]

Email your database to me and I'll try to reproduce the problem here. I'll PM you my email address.

[dennit]

Email your database to me and I'll try to reproduce the problem here. I'll PM you my email address.

You should have it now -Thanks!