Mailing List archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[linux-dvb] Re: possible free disk space bug in vdr ?
Martin Hoffmann wrote:
>
> As you please, master ;-)
>
> root@linux:/video0 > df -m -P '/video0'
> Filesystem 1048576-blocks Used Available Capacity Mounted on
> /dev/hda3 31584 10591 19389 36% /
>
> What are you looking for ?
I guess I was "hoping" to find something wrong with the line that
shows the free disk space for the /video0 partition, but everything
appears to be fine.
> As said before the problem occured this afternoon - and this evening the
> drivers crashed :-(
> So i rebooted vdr and everything just works fine ...
> I tried a manual recording and it works ...
>
> ul 16 20:23:38 linux vdr[657]: switching to channel 7
> Jul 16 20:23:38 linux vdr[657]: record
> /video0/@Pro-7/2001-07-16.20:23.50.50.rec
> Jul 16 20:23:38 linux vdr[657]: creating directory
> /video0/@Pro-7/2001-07-16.20:23
> .50.50.rec
> Jul 16 20:23:38 linux vdr[657]: recording to
> '/video0/@Pro-7/2001-07-16.20:23.50.5
> 0.rec/001.vdr'
> Jul 16 20:23:39 linux vdr[657]: creating directory
> /video1/@Pro-7/2001-07-16.20:23
> .50.50.rec
> Jul 16 20:23:39 linux vdr[747]: output thread started (pid=747)
> Jul 16 20:23:39 linux vdr[748]: input thread started (pid=748)
> Jul 16 20:47:20 linux -- MARK --
> Jul 16 20:55:52 linux vdr[747]: recording to
> '/video0/@Pro-7/2001-07-16.20:23.50.5
> 0.rec/002.vdr'
> Jul 16 20:56:27 linux in.telnetd[848]: connect from 192.168.222.1
> (192.168.222.1)
> Jul 16 20:56:49 linux su: (to root) martin on /dev/pts/0
> Jul 16 20:56:49 linux PAM-unix2[860]: session started for user root, service
> su
> Jul 16 20:59:00 linux /USR/SBIN/CRON[870]: (root) CMD ( rm -f
> /var/spool/cron/last
> run/cron.hourly)
> Jul 16 21:06:37 linux vdr[747]: ERROR: can't record MPEG1!
>
> Except that i don't get what the last line means :-( But the recording still
> went on, so i think this was an error in the stream ? Never had this error
> before...
I have stripped down the code taken from tuxplayer which repacks the TS
into a PES, and thereby removed the parts that handled MPEG1 (since no
tv broadcast is in MPEG1). If there are data corruptions in the stream
they might trigger the MPEG1 detection that is still in there.
> Can't it be that the check which Directory has more space left (video0 or
> video1) goes wrong somehow if both are equal ? Maybe then vdr thinks that
> nothing is left ?
> Just an idea ...
I don't think so, because te initial error message at 15:58:41 ("low disk space...")
comes from cRecordBuffer::RunningLowOnDiskSpace() in dvbapi.c, and that one just
examines the partition on which the current recording is being stored.
Unfortunately the log file lines in your original post were cut off.
Could you please locate the line starting with
Jul 16 15:58:41 linux vdr[3931]: recording to '/video0/Star_Tr
in your /var/log/messages and retry the
df -m -P '/video0/Star_Tr...'
with the full path name from that error message (except for the "/002.vdr" part)?
So if, for instance, the complete line reads
Jul 16 15:58:41 linux vdr[3931]: recording to '/video0/Star_Trek/2001-07-09.21:13.99.99.rec/002.vdr'
then execute
df -m -P '/video0/Star_Trek/2001-07-09.21:13.99.99.rec'
Klaus
>
> Martin.
>
> ----- Original Message -----
> From: "Klaus Schmidinger" <Klaus.Schmidinger@cadsoft.de>
> To: "linux-dvb" <linux-dvb@linuxtv.org>
> Sent: Monday, July 16, 2001 9:02 PM
> Subject: [linux-dvb] Re: possible free disk space bug in vdr ?
>
> Martin Hoffmann wrote:
> >
> > root@linux:/home/martin > df -m -P
> > Filesystem 1048576-blocks Used Available Capacity Mounted on
> > /dev/hda3 31584 9800 20180 33% /
> > /dev/hda1 23 6 17 25% /boot
> > /dev/hdc1 23639 4136 19259 18% /video1
> > shmfs 137 0 137 0% /dev/shm
> >
> > in fact it changed, because i did a new recording at the moment :-)
> >
> > Martin.
>
> Hmm, I can't see anything wrong here.
> Please try this:
>
> df -m -P '/video0'
>
> and send me the result.
>
> Klaus
>
> >
> > ----- Original Message -----
> > From: "Klaus Schmidinger" <Klaus.Schmidinger@cadsoft.de>
> > To: "linux-dvb" <linux-dvb@linuxtv.org>
> > Sent: Monday, July 16, 2001 8:20 PM
> > Subject: [linux-dvb] Re: possible free disk space bug in vdr ?
> >
> > Martin Hoffmann wrote:
> > >
> > > Hi,
> > >
> > > today i encountered a strange problem: VDR stopped in the middle of a
> > recording and /var/log/messages says something about disk full (?)
> > > In fact i have 2 Harddisks and both have around 20 GB free (see df
> below,
> > /video0 is located in root)
> > >
> > > So here is the log file, i have marked the strange things with *** in
> > front!
> > >
> > > As you can see the drivers also crashed later on (i had to reboot the
> > machine), but i'm not sure if this was caused by the first problem...
> > >
> > > Jul 16 15:40:34 linux -- MARK --
> > > Jul 16 15:57:00 linux vdr[657]: timer 1 start
> > > Jul 16 15:57:00 linux vdr[657]: switching to channel 6
> > > Jul 16 15:57:00 linux vdr[657]: record
> > /video0/Star_Trek_-_Das_nächste_Jahrhundert
> > > Jul 16 15:57:00 linux vdr[657]: creating directory
> > /video0/Star_Trek_-_Das_nächste
> > > Jul 16 15:57:00 linux vdr[657]: creating directory
> > /video0/Star_Trek_-_Das_nächste
> > > Jul 16 15:57:00 linux vdr[657]: recording to
> > '/video0/Star_Trek_-_Das_nächste_Jahr
> > > Jul 16 15:57:00 linux vdr[3931]: output thread started (pid=3931)
> > > Jul 16 15:57:00 linux vdr[3932]: input thread started (pid=3932)
> > > ***Jul 16 15:58:41 linux vdr[3931]: low disk space (0 MB, limit is 512
> MB)
> > > ***Jul 16 15:58:41 linux vdr[3931]: recording to '/video0/Star_Tr'
> > > Jul 16 15:59:00 linux /USR/SBIN/CRON[3938]: (root) CMD ( rm -f
> > /var/spool/cron/las
> > > Jul 16 16:20:34 linux -- MARK --
> > > ***Jul 16 16:34:53 linux vdr[3931]: ERROR: max number of files (255)
> > exceeded
> > > Jul 16 16:34:53 linux vdr[3931]: output thread ended (pid=3931)
> > > Jul 16 16:34:53 linux vdr[3932]: input thread ended (pid=3932)
> > > Jul 16 16:59:00 linux /USR/SBIN/CRON[4091]: (root) CMD ( rm -f
> > /var/spool/cron/las
> > > Jul 16 17:05:01 linux vdr[657]: buffer stats: 49040 (4%) used
> > > Jul 16 17:05:01 linux vdr[657]: timer 1 stop
> > > Jul 16 17:05:01 linux vdr[657]: deleting timer 1
> > > Jul 16 17:20:34 linux -- MARK --
> > > Jul 16 17:40:34 linux -- MARK --
> > > Jul 16 17:59:00 linux /USR/SBIN/CRON[4215]: (root) CMD ( rm -f
> > /var/spool/cron/las
> > > Jul 16 18:20:34 linux -- MARK --
> > > Jul 16 18:40:34 linux -- MARK --
> > > Jul 16 18:59:00 linux /USR/SBIN/CRON[4338]: (root) CMD ( rm -f
> > /var/spool/cron/las
> > > Jul 16 19:20:34 linux -- MARK --
> > > Jul 16 19:40:34 linux -- MARK --
> > > Jul 16 19:40:43 linux vdr[657]: switching to channel 5
> > > Jul 16 19:40:56 linux vdr[657]: switching to channel 5
> > > ***Jul 16 19:40:56 linux vdr[657]: ERROR: channel 5 not sync'ed!
> > > Jul 16 19:40:56 linux vdr[657]: retrying
> > > Jul 16 19:40:58 linux vdr[657]: ERROR: timeout while tuning
> > > *** Jul 16 19:40:58 linux vdr[657]: ERROR: channel 5 not sync'ed!
> > > Jul 16 19:40:58 linux vdr[657]: retrying
> > >
> > > df
> > > Filesystem 1k-blocks Used Available Use% Mounted on
> > > /dev/hda3 32341888 9998396 20700448 33% /
> > > /dev/hda1 23300 5309 16788 25% /boot
> > > /dev/hdc1 24205416 3185400 20770108 14% /video1
> > > shmfs 227220 0 227220 0% /dev/shm
> > >
> > > Any idea about that ?
> > >
> > > Is it a bug ? Could it be that vdr interprets something wrong if both
> > video0 and video1 have exactly the same amount of free space ?
> > > Maybe you could comment on this, Klaus ?
> >
> > Can you please do a
> >
> > df -m -P
> >
> > oy your machine and post the result?
> >
> > Klaus
--
_______________________________________________________________
Klaus Schmidinger Phone: +49-8635-6989-10
CadSoft Computer GmbH Fax: +49-8635-6989-40
Hofmark 2 Email: kls@cadsoft.de
D-84568 Pleiskirchen, Germany URL: www.cadsoft.de
_______________________________________________________________
--
Info:
To unsubscribe send a mail to listar@linuxtv.org with "unsubscribe linux-dvb" as subject.
Home |
Main Index |
Thread Index