24th Jun, 2014, 12:56 AM
24th Jun, 2014, 04:30 AM
(23rd Jun, 2014 06:53 PM)mk01 Wrote: [ -> ]@nsviper
while testing new build system I let build Helix and installed the .deb . and realised one thing we didn't consider until now.
when xbmc job is started, there is timeout 100s waiting for XBMC be fully up. normally that is sufficient for XBMC start. but if HUGE db is being converted to new version (what was my case) 100s wasn't enough, xbmc upstart job run out of timeout considered xbmc.bin as not being started correctly (as xbmc was still converting DB long before notifying system that started ok).
so edit /etc/init/xbmc.conf, go into post-start section and change TIMEOUT=100 to TIMEOUT=100000. (you can just delete the new version DB's from sql before, start xbmc again. it will start conversion again. maybe your case is failing for that reason that xbmc.bin is being killed before it fully converts the DB).
Nice 1 mk01 - all fixed now
24th Jun, 2014, 10:24 AM
(23rd Jun, 2014 07:58 PM)Oliver.O Wrote: [ -> ]IR remote via GPIO still not working after boot until XBMC is restarted (1.0 RC 2, all updates installed).
This fix still applies.
@Oliver.O
please follow this issue (my comments). https://github.com/xbianonpi/xbian/issues/549
also putting "started lirc" into xbmc would effectively disable it for users with disabled lirc - as one of implications you was expecting.
now when I think about it a bit, xbmc-preload could be changed to check for lirc autostart enabled/disabled and according to that lircd as process could by added by it into wait_for list automatically.
can you test if adding
Code:
grep -wqi manual /etc/init/lirc.* || WAITFOR="$WAITFOR lircd"added into line 37 (before "waitfor "$WAITFOR"") would solve the problem ?
(24th Jun, 2014 04:30 AM)nsviper Wrote: [ -> ]Nice 1 mk01 - all fixed now
@nsviper
can you please just open this one on git issues to follow up later (just shortly - xbmc killed before it can finish startup tasks - in case they run for longer time like DB convert)?
as it is quite important but I have to think a bit about elegant solution - so we don't forget meanwhile.
(24th Jun, 2014 12:34 AM)Oliver.O Wrote: [ -> ]do you think all the other changes are really necessary?
they are not because :
- adding "started mountall" to lirc is unnecessary because lirc already depends on stopped module-init-tools and module-init-tools won't start before "started mountall"
- adding "started lirc" into xbmc is unnecessary if already added into xbmc-preload because started xbmc-preload is in xbmc dependencies for start.
24th Jun, 2014, 05:06 PM
24th Jun, 2014, 05:50 PM
24th Jun, 2014, 06:17 PM
25th Jun, 2014, 12:52 AM
(24th Jun, 2014 04:30 AM)nsviper Wrote: [ -> ](23rd Jun, 2014 06:53 PM)mk01 Wrote: [ -> ]@nsviper
while testing new build system I let build Helix and installed the .deb . and realised one thing we didn't consider until now.
when xbmc job is started, there is timeout 100s waiting for XBMC be fully up. normally that is sufficient for XBMC start. but if HUGE db is being converted to new version (what was my case) 100s wasn't enough, xbmc upstart job run out of timeout considered xbmc.bin as not being started correctly (as xbmc was still converting DB long before notifying system that started ok).
so edit /etc/init/xbmc.conf, go into post-start section and change TIMEOUT=100 to TIMEOUT=100000. (you can just delete the new version DB's from sql before, start xbmc again. it will start conversion again. maybe your case is failing for that reason that xbmc.bin is being killed before it fully converts the DB).
Nice 1 mk01 - all fixed now
I may have been too hasty, or my databases are too large
Further testing required at my end, but will raise on git once i've confirmed...26th Jun, 2014, 01:03 AM
26th Jun, 2014, 06:20 AM
(25th Jun, 2014 12:52 AM)nsviper Wrote: [ -> ]I may have been too hasty, or my databases are too large
After experiencing the same trouble I chose a different approach: I upgraded a Windows XBMC installation from Frodo to Gotham and the databases were converted correctly.
29th Jun, 2014, 09:56 PM
Hello
i got a problem after update/upgrade, it doesn´t matter if i try it via xbian config or by sudo -i and:
apt-get clean
apt-get update
apt-get upgrade
this is the message i get:
Fetched 71.3 MB in 8min 13s (145 kB/s)
Create a snapshot of '/tmp/btrfs-snap/root/@' in '/tmp/btrfs-snap/root/@btrfs-auto-snap_apt-run-2014-06-29-1143'
Transaction commit: none (default)
Delete subvolume '/tmp/btrfs-snap/root/@last_good_known'
Create a snapshot of '/tmp/btrfs-snap/root/@btrfs-auto-snap_apt-run-2014-06-29-1143' in '/tmp/btrfs-snap/root/@last_good_known'
@btrfs-auto-snap_apt-run-2014-06-29-1143, 1 created snapshots, 0 destroyed snapshots, 0 warnings.
Preconfiguring packages ...
dpkg: error: reading package info file '/var/lib/dpkg/available': Input/output error
E: Sub-process /usr/bin/dpkg returned an error code (2)
What can i do?
I use Xbian RC2 and Gotham 13.1 on a Raspberry.
Thanks for your Help.
Yami
i got a problem after update/upgrade, it doesn´t matter if i try it via xbian config or by sudo -i and:
apt-get clean
apt-get update
apt-get upgrade
this is the message i get:
Fetched 71.3 MB in 8min 13s (145 kB/s)
Create a snapshot of '/tmp/btrfs-snap/root/@' in '/tmp/btrfs-snap/root/@btrfs-auto-snap_apt-run-2014-06-29-1143'
Transaction commit: none (default)
Delete subvolume '/tmp/btrfs-snap/root/@last_good_known'
Create a snapshot of '/tmp/btrfs-snap/root/@btrfs-auto-snap_apt-run-2014-06-29-1143' in '/tmp/btrfs-snap/root/@last_good_known'
@btrfs-auto-snap_apt-run-2014-06-29-1143, 1 created snapshots, 0 destroyed snapshots, 0 warnings.
Preconfiguring packages ...
dpkg: error: reading package info file '/var/lib/dpkg/available': Input/output error
E: Sub-process /usr/bin/dpkg returned an error code (2)
What can i do?
I use Xbian RC2 and Gotham 13.1 on a Raspberry.
Thanks for your Help.
Yami
30th Jun, 2014, 04:28 AM
1st Jul, 2014, 04:25 AM
1st Jul, 2014, 08:09 AM
1st Jul, 2014, 09:53 PM
2nd Jul, 2014, 05:20 AM