Opened 16 years ago
Closed 15 years ago
#5210 closed patch (fixed)
Automatic shutdown for SBE
Reported by: | Owned by: | cpinkham | |
---|---|---|---|
Priority: | minor | Milestone: | 0.22 |
Component: | scheduler | Version: | head |
Severity: | low | Keywords: | auto shutdown for a SBE |
Cc: | Ticket locked: | no |
Description
This patch allows a SBE to shutdown it self independently of the MBE. It works like a standalone master backend with the idleTimeoutSecs as if no MBE was present. A new setting enables this feature : idleTimeoutSlaveSecs. It can be configured through mythtv-setup in the screen "Shutdown/Wakeup? Options".
Attachments (2)
Change History (10)
Changed 16 years ago by
Attachment: | ticket.diff added |
---|
comment:1 Changed 16 years ago by
comment:2 Changed 16 years ago by
Milestone: | 0.21.1 → 0.22 |
---|---|
Severity: | medium → low |
Status: | new → infoneeded_new |
Version: | 0.21-fixes → head |
This is a new feature. New features usually do not go into a -fixes branch. Can you update your patch to trunk?
Goob job, btw.
comment:3 Changed 16 years ago by
Yes this is a new feature. Unfortunately, I don't have time to setup mythtv with version 0.22 without breaking my installation running the version 0.21. The time I wrote it, 0.21 was just released ant I thought it was the best way to submit it.
However, I just found that live tv is broken whith this feature : I think I found what happended : when calling FillRecordListFromMaster?() a new socket is open and disrupt the live tv.
Here is a new patch againt 0.21 that handle this correctly but it doesn't call FillRecordListFromMaster?() anymore, it relies FillRecordList?().
Changed 16 years ago by
Attachment: | ticket.2.diff added |
---|
comment:4 Changed 15 years ago by
ticket.2.diff appears to be corrupted. Is there any way you could re-upload?
comment:5 Changed 15 years ago by
Status: | infoneeded_new → new |
---|
I have a mostly-complete patch to allow the master backend to bring slaves up and down as needed. This may obviate the need for the patch in this ticket. The patch I've written allows the master to consider the slave's tuners for recordings and wake a slave up when necessary and tell a slave to go to sleep (shutdown/whatever) when the slave won't be needed for a while (and isn't in use).
comment:6 Changed 15 years ago by
Great news.
In my solution, the slave backend was awaken by the master backend with a magic wol packet sended by a "at" job... this mecanism was done outside mythtv but your work seems promising!
Let me know if you need more informations. Thanks :) Guillaume
comment:7 Changed 15 years ago by
Owner: | changed from bjm to cpinkham |
---|---|
Status: | new → assigned |
comment:8 Changed 15 years ago by
Resolution: | → fixed |
---|---|
Status: | assigned → closed |
I don't think this patch is necessary anymore since the master backend can now shutdown and wakeup slaves on-demand when they are required for recording. This functionality went into trunk in [20084]. Closing as 'fixed, but in a slightly different manner'
Here is some informations I forgot to tell : it asks the MBE about new pending recordings with QUERY_GETALLPENDING at startup, at the end of a recording and when a SCHEDULE_CHANGE is received. It also avoid a MBE to send again a BACKEND_MESSAGE it just send a second before and thus avoid flooding a SBE.