[mythtv-users] mythbackend crashes every 6-7 hours

[Rob Verduijn]

Hi,

I have no intention of sending rapidshare any money as wel, I'm open for
other suggestions or a howto on how to get that 2Mb on pastbin.

no solid state drives in my machine.
Ordinary harddisks in there.
I don't mind it takes it's time to boot cause it's a backend that rarely is
turned off.

OK it's 250Gb and not 256 see the type of the drive at this link
http://www.seagate.com/ww/v/index.jsp?vgnextoid=c2be99f4fa74c010VgnVCM100000dd04090aRCRD&locale=en-US

Furthermore there we're no recordings during the test and I didn't get any
further than booting up the living room frontend.
The frontend was up and running but doing nothing. No tv shows were viewed
during the testing.

fragmentation is really bad
I'm starting defrag in screen session right away, this could take awhile
Never thought fragmentation would bother me again since I left windows
behind me a decade ago.
I'll whip up a nice script that keeps an eye on it from the cron.

xfs_db -c frag -r /dev/system/data2
actual 100214, ideal 1163, fragmentation factor 98.84%

the freespace is not much better
xfs_db -c freesp -r /dev/system/data2
   from      to extents  blocks    pct
      1       1      70      70   0.00
      2       3      22      52   0.00
      4       7      41     240   0.00
      8      15     110    1256   0.01
     16      31    1372   22422   0.22
     32      63     979   36749   0.35
     64     127     769   67759   0.65
    128     255    5468  909456   8.75
    256     511      47   14417   0.14
    512    1023       6    5174   0.05
   1024    2047       1    1171   0.01
   2048    4095       1    3535   0.03
   4096    8191       2   10872   0.10
  16384   32767       4   98051   0.94
  32768   65535       5  226895   2.18
  65536  131071       6  492906   4.74
 131072  262143       8 1593529  15.34
 262144  524287       5 2106212  20.27
 524288 1048575       5 3630188  34.94
1048576 2097151       1 1168264  11.24

2010/11/1 Gavin Hurlbut <gjhurlbu at gmail.com>

> On Mon, Nov 1, 2010 at 12:32 PM, Rob Verduijn <rob.verduijn at gmail.com>
> wrote:
> > system-data2 /data2 443Gb XFS
> > system-data2 is used for the recordings.
> > There is 59Gb free space left on that partition
> > No redundancy has been provided because I  don't care about the local
> data.
>
> OK, by my calculation, this is about 87% full.  XFS is known to have
> issues with filesystems that are approaching full (bad fragmentation,
> IIRC).
>
> xfs_db -c frag -r /dev/whatever/system-data2
> xfs_db -c freesp -r /dev/whatever/system-data2
>
> The first shows fragmentation in the used drive space, and the second
> shows the fragmentation in the free drive space (which is more likely
> to be the issue at hand).
>
> Also, it seems that you may be running on SSD drives (I don't remember
> any 256GB drives that aren't SSD)?  That may somehow be coming into
> play.  Maybe your 1G+ is data trying to get to your drives and making
> it there slower than its coming in?
>
> I can't download your valgrind logs from rapidshare as it says all the
> free downloads are used up, and I'm not gonna send them money :)
>
> From the capture hardware you are using, there should be no reason for
> excessive memory use at all.  I use far more capture devices, and my
> backend sits at < 700M virtual and < 50MB resident.  And for quite
> some time, I used IVTV cards in there, and never saw memory leaks.
> This is why I'm fairly confident that it's something specific to your
> setup.
>
> I'll try the rapidshare link again in a bit to see if it will let me
> download.
> _______________________________________________
> mythtv-users mailing list
> mythtv-users at mythtv.org
> http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mythtv.org/pipermail/mythtv-users/attachments/20101101/22346958/attachment.htm>