Nate Iverson wrote:
I did a couple things. I pulled the PVR-150 so it is not sharing an
interrupt with the graphics card, that seemed to make it stable for 2
days. The audio card is now sharing the interrupt with the graphics
card.
[EMAIL PROTECTED] ~]$ cat /proc/interrupts
CPU0 CPU1
0: 37935159 54664878 IO-APIC-edge timer
1: 10259 1656 IO-APIC-edge i8042
8: 2 0 IO-APIC-edge rtc
9: 0 1 IO-APIC-fasteoi acpi
12: 68081 6197 IO-APIC-edge i8042
14: 2869562 32876 IO-APIC-edge ide0
15: 141208 4326 IO-APIC-edge ide1
16: 0 0 IO-APIC-fasteoi ohci_hcd:usb1
17: 870225 140957 IO-APIC-fasteoi megaraid
18: 10473848 65 IO-APIC-fasteoi eth1
19: 731601 1640 IO-APIC-fasteoi ivtv0
20: 6080727 82891 IO-APIC-fasteoi CS46XX,
[EMAIL PROTECTED]:0000:01:05.0
21: 335821 1302 IO-APIC-fasteoi ivtv1
NMI: 0 0
LOC: 92608893 92608893
ERR: 0
MIS: 0
I then updated the kernel, drivers, and mythtv suite. That was a huge
mistake. I started getting DMA timeout errors with 5 minutes of starting
to record. I have a spare 1Ghz Celeron, so I decide to put 2 pvr-150 in
that system and build a slave backend. To make a long story short I have
a complete frontend/backend system and a slave backend running the
default mythdora 4 distro kernel(2.6.20-1.2944.fc6) with the newest
myth-suite. The master appears to be pretty stable after reverting to
the older kernel. The slave can access the masters database, but the
slave doesn't record. I haven't been able to find any hint on why this
is. I have the the NFS share setup correctly and can manually create a
file on the mounted share from the slave system. I used the backend only
option in Mythdora setup for the slave. The mythtvsetup/capture card
screen shows only the local system capture cards. However when I exited
mythtvsetup on either system before setting up the Schedule Direct query
I got a warning message saying that all 4 capture didn't have a default
channel set. So myth know of all 4 capture cards. Any Ideas? Or maybe a
link to a slave backend howto? Google didn't pop up much besides the
mysql commands that allowed the slave to see the masters database.
Thanks for the help,
Nate
Boot up order is VERY important when using slaves. I have found that
after I make capture card changes on the slave, I have to restart the
master backend processes to get them to see the slave card changes.
I have also found that any errors on the slave machine's capture
card parameters will cause the slave mythtv-backend process to immediately
stop, and usually with no indicating error messages.
My suggestion is to shutdown both mythtv-backend processes (master/slave),
then go into mythtv-setup on the master, and delete ALL capture cards.
Then add the cards back in, in the order that you want them to be used
for capturing. So, if you want the capture cards on the slave to be used
first, add them into the slave first.
Then bring up the master backend, then boot up the slave backend.
Hopefully your cards will all be present and system will be useable.
begin:vcard
fn:Mark Paulus
n:Paulus;Mark
org:MCI;Lec Interfaces / 40419
adr;dom:;;2424 Garden of the Gods Rd;Colorado Springs;CO;80919
email;internet:[EMAIL PROTECTED]
title:Mark Paulus
tel;work:719-535-5578
tel;pager:800-pagemci / 1406052
tel;home:v622-5578
version:2.1
end:vcard
_______________________________________________
ivtv-users mailing list
[email protected]
http://ivtvdriver.org/mailman/listinfo/ivtv-users
