JackOfAll wrote:
> John,
>
> logitechmediaserver-7.9.0-0.67.20150220git1424339864 in testing repo.
> 0.66 pushed to "stable".
Thank you, testing version installed and playing music. Busy summer
weekend here, will try new features later in the week.
A camel is a racehors
John,
logitechmediaserver-7.9.0-0.67.20150220git1424339864 in testing repo.
0.66 moved to "stable".
Changelog
Code:
* Fri Feb 20 2015 - 7.9.0-0.67.20150220git1424339864
- Git revision public/7.9 c371802dba0841070f987b9be149ea6ece1ef4c6
Make sure the columns a
One for Michael, a search for - A Killer's Dream - brings the server to
it's knees.
Substring search is the killer, in particular with popular terms like
"a*" and "s*". Some statistics show me:
- there are 71543 search terms starting with "a" resulting in >16 Mio.
occurrences across 2.5 Mio.
Michael, don't fix the playing music during a clear & rescan. It's a
I was kidding :-).
The search for A killer's Dream stopped my server, boom rebuffered etc,
but the next version of LMS handled the query without stopping. Too late
then to withdraw the comment. Its still slow, but certainly n
mherger wrote:
> > JackOfAll wrote:
> >> And obviously a benefit, the current playlist not being "zapped", as
> the
> >> result of a "clear and rescan".
> >
> > Yes, quite disturbing to start a clear & rescan and the music keeps
> > playing. So, fix worked, thank you Michael and JackOfAll.
>
> I
JackOfAll wrote:
And obviously a benefit, the current playlist not being "zapped", as the
result of a "clear and rescan".
Yes, quite disturbing to start a clear & rescan and the music keeps
playing. So, fix worked, thank you Michael and JackOfAll.
I'm sorry about that. Can fix this easily :-P
JackOfAll wrote:
> And obviously a benefit, the current playlist not being "zapped", as the
> result of a "clear and rescan".
Yes, quite disturbing to start a clear & rescan and the music keeps
playing. So, fix worked, thank you Michael and JackOfAll.
One for Michael, a search for - A Killer's
And obviously a benefit, the current playlist not being "zapped", as the
result of a "clear and rescan".
JackOfAll's Profile: http://forums.slimdevices.com/member.php?userid=3069
View this thread: http://forums.slimdevices.
Looking good
Code:
[15-02-15 13:10:03.8936] main::main (203) Starting Logitech Media Server
scanner (v7.9.0, 0.66.20150215git1424002734, Sun 15 Feb 13:06:15 GMT 2015) perl
5.018004
[15-02-15 13:10:22.9837] Slim::Music::Import::runImporter (485) Starting
Slim::
John,
"sudo sc-cleanUpdate-testing log\*", full scan, and report back, please.
LOGITECHMEDIASERVER-7.9.0-0.66.20150215GIT1424002734
Changelog
Code:
* Sun Feb 15 2015 - 7.9.0-0.66.20150215git1424002734
- Git revision public/7.9 c01f83d0f099b75fcffa1c70d106e59a754e3
mherger wrote:
>
> You were right. I've committed a change to make sure that module is
> loaded before it's being used. Could you please give it a try and roll
> out a new build if that commit fixes this? Thanks!
Thanks, Michael. I'll "spin" a new build
---
> Verified. I guess it is a result of this commit, 'Don't clear the
> current playlist when a scan is triggered'
> (https://github.com/Logitech/slimserver/commit/a58532201ba877aea0a1d9004a25da3598eb351c).
> I'll look a bit harder tomorrow.
You were right. I've committed a change to make sure that
Most likely this has nothing to do with the packaging. Please post to a new 7.9
thread and include more information if possible. Is there more in the logs than
just that one line?
> Am 15.02.2015 um 00:04 schrieb Wirrunna
> :
>
>
> JackOfAll, not sure if here is the right place, but Clear &
Verified. I guess it is a result of this commit, 'Don't clear the
current playlist when a scan is triggered'
(https://github.com/Logitech/slimserver/commit/a58532201ba877aea0a1d9004a25da3598eb351c).
I'll look a bit harder tomorrow.
Wirrunna wrote:
>
> Log message -
> [15-02-15 08:06:24.9438] Slim::Control::Request::execute (1890) Error:
> While trying to run function coderef
> [Slim::Control::Commands::wipecacheCommand]: [Can't locate object method
> "getMetadataFor" via package "Slim::Player::Protocols::Volatile" at
> /us
JackOfAll, not sure if here is the right place, but Clear & Rescan
doesn't work on VBox 7.9.0 - 0.65.20150213git1423815993, but does on
Win7 7.9.0 - 1423815993 . Tried on two different VBoxes.
Log message -
[15-02-15 08:06:24.9438] Slim::Control::Request::execute (1890) Error:
While trying to run
Wirrunna wrote:
> Thank you for this, the Full Text Search has been given a polish and
> works superbly.
Thank Michael, John! I'm just the fool re-packaging it. ;)
JackOfAll's Profile: http://forums.slimdevices.com/member
JackOfAll wrote:
> Last weeks 0.64 builds (with CUE changes) have been pushed from testing
> to stable. (I've had one positive report from someone who does
> extensively use CUEsheets.)
>
> This weeks build in testing repo, ("sudo sc-cleanUpdate-testing
> logitechmediaserver\*"), is 7.9.0-0.65.
Last weeks 0.64 builds (with CUE changes) have been pushed from testing
to stable. (I've had one positive report from someone who does
extensively use CUEsheets.)
This weeks build in testing repo is 7.9.0-0.65.20150213git1423815993
Code:
* Fri Feb 13 2015 - 7.9.0-
JackOfAll wrote:
> John,
>
> Can you run on the machine you already upgraded to the test version...
>
> >
Code:
> >
> sudo sc-cleanDowngrade logitechmediaserver{,-firmware-ip3k}
>
> >
>
> This gets you back to stable repo build, with that F
John,
Can you run on the machine you already upgraded to the test version...
Code:
sudo sc-cleanDowngrade logitechmediaserver{,-firmware-ip3k}
This gets you back to stable repo build, with that FTS "fix", minus the
CUE sheet changes.
-
Wirrunna wrote:
> Thank you, installed and playing. FTS search working properly again.
Great!. I will make a new stable build tomorrow with just that FTS
patch. After which, I'll post on VB forum asking if VB users will update
to the testing repo build, but offering a clear downgrade path if the
JackOfAll wrote:
> >
Code:
> >
> sudo sc-cleanUpdate-testing logitechmediaserver\*
>
> >
>
> No risk you running the testing build if you don't use CUE sheets.
>
> Also John, off topic, and I don't know whether you already noticed,
> but all
Wirrunna wrote:
> Well, the world has not stopped dead and I don't use que sheets, so,
> whatever is easiest for you, do nothing, build 0.63+ or point me to the
> post that told me how to use the test build.
Code:
sudo sc-cleanUpdate-testing logitechmediaserver\*
JackOfAll wrote:
> Erm, if that fix is
>
> >
Code:
> >
> - Git revision public/7.9 c126e3331866263868b36370ded6e9be31e04f44
> When using full text search we need to maintain the FTS' sort order when
> searching.
>
> >
>
> It's not in
Wirrunna wrote:
> Did you do this ? 'Cause you usually push the old test version down so
> that us adventurous VortexBox users can get the week old version and I'm
> looking forward to the fix for Full Text Search committed on Jan 28
> that works on my Win 7 test box.
Erm, if that fix is
JackOfAll wrote:
> Later this evening, I'll push the latest weekly LMS 7.9 builds to the SC
> testing repo.
Did you do this ? 'Cause you usually push the old test version down so
that us adventurous VortexBox users can get the week old version and I'm
looking forward to the fix for Full Text Sea
I think it was cue sheets when folder browsing that got some biugfixing
and some other inconsequential things cue vs tags got fixed ? A dual
corner case ?
No, the CUE sheet scanner code has been pretty much re-written to cover
more tags, improve consolidation of album and track level informatio
I think it was cue sheets when folder browsing that got some biugfixing
and some other inconsequential things cue vs tags got fixed ? A dual
corner case ?
Classical listeners are 1/2 forgiven for the need to do this :D
Tagging is broken from the core in general IMO nothing to do with our
system
mherger wrote:
>
> 7.9 is a dev branch, not for fainthearted :-).
>
I'm going to ignore that! ;)
mherger wrote:
>
> I don't use CUE sheets myself. I can't test this.
>
And there's the problem, hence my warning. I don't use them either. But
I know a man that does. And I'm pretty sure that
Later this evening, I'll push the latest weekly LMS 7.9 builds to the SC
testing repo. Just a heads-up There are some very scary changes, (or
> they look scary to me), with regard to the handling of CUE files. Not
7.9 is a dev branch, not for fainthearted :-).
True, it's not a small change.
Later this evening, I'll push the latest weekly LMS 7.9 builds to the SC
testing repo. Just a heads-up There are some very scary changes, (or
they look scary to me), with regard to the handling of CUE files. Not
something I'd be testing anyway as I don't use them, but I know the
people who do
32 matches
Mail list logo
