Thread: [Ample-devel] Ample 0.5.5 on RH7.3...
Status: Alpha
Brought to you by:
alphix
Menu
▾
▴
ample-devel
|
From: Andy P. <and...@fr...> - 2002-10-27 18:30:31
|
Hi David, Great to see Ample is under development again, after your various woes! I just downloaded and built 0.5.5 on my RH7.3 box (my client is now running 8.0 but I don't want to upgrade my server just yet for a few reasons, so 7.3 it is). It compiles fine. I configured it for tcp wrappers and syslogging. Dropped it in place of the old version (I was on 0.5.2 until now, it worked well for me). I didn't change the config file as I saw no reason to add filters or authentication to my setup, I'm firewalled and tcp-wrappered. I run ample out of xinetd and have that working just fine. Trying to connect back to the box with Galeon and XMMS - no response. In the syslog I saw this: Oct 27 17:44:02 phoenix ample[6759]: Ample/0.5.5 started Oct 27 17:44:02 phoenix ample[6753]: XXXXXXXXXXXXXXXXXXXXXXXX Oct 27 17:44:02 phoenix ample[6753]: BitRate is 128, FileLength is 3987584 Oct 27 17:44:02 phoenix ample[6753]: TPF is 0.026122 and BPF is 417.959184, we have 38.281250 frames in one second Oct 27 17:44:02 phoenix ample[6753]: XXXXXXXXXXXXXXXXXXXXXXXX Oct 27 17:44:02 phoenix ample[6753]: BitRate is 128, FileLength is 5081675 Oct 27 17:44:02 phoenix ample[6753]: TPF is 0.026122 and BPF is 417.959184, we have 38.281250 frames in one second Oct 27 17:44:02 phoenix ample[6753]: XXXXXXXXXXXXXXXXXXXXXXXX Oct 27 17:44:02 phoenix ample[6756]: XXXXXXXXXXXXXXXXXXXXXXXX Oct 27 17:44:02 phoenix ample[6759]: XXXXXXXXXXXXXXXXXXXXXXXX Oct 27 17:44:02 phoenix ample[6762]: Ample/0.5.5 started Oct 27 17:44:02 phoenix ample[6765]: Ample/0.5.5 started Oct 27 17:44:02 phoenix ample[6753]: BitRate is 112, FileLength is 3842322 Oct 27 17:44:02 phoenix ample[6756]: BitRate is 128, FileLength is 3987584 Oct 27 17:44:02 phoenix ample[6759]: BitRate is 128, FileLength is 3987584 Oct 27 17:44:02 phoenix ample[6762]: XXXXXXXXXXXXXXXXXXXXXXXX Oct 27 17:44:02 phoenix ample[6765]: XXXXXXXXXXXXXXXXXXXXXXXX Oct 27 17:44:02 phoenix ample[6768]: Ample/0.5.5 started Oct 27 17:44:02 phoenix ample[6753]: TPF is 0.026122 and BPF is 365.714286, we have 38.281250 frames in one second Oct 27 17:44:02 phoenix ample[6756]: TPF is 0.026122 and BPF is 417.959184, we have 38.281250 frames in one second Oct 27 17:44:02 phoenix ample[6759]: TPF is 0.026122 and BPF is 417.959184, we have 38.281250 frames in one second Oct 27 17:44:02 phoenix ample[6762]: BitRate is 128, FileLength is 3987584 Oct 27 17:44:02 phoenix ample[6765]: BitRate is 128, FileLength is 3987584 Oct 27 17:44:02 phoenix ample[6768]: XXXXXXXXXXXXXXXXXXXXXXXX Oct 27 17:44:02 phoenix ample[6753]: XXXXXXXXXXXXXXXXXXXXXXXX Oct 27 17:44:02 phoenix ample[6771]: Ample/0.5.5 started Oct 27 17:44:02 phoenix ample[6756]: XXXXXXXXXXXXXXXXXXXXXXXX [...] Oct 27 17:45:52 phoenix ample[6856]: TPF is 0.026122 and BPF is 626.938776, we have 38.281250 frames in one second Oct 27 17:45:52 phoenix ample[6856]: XXXXXXXXXXXXXXXXXXXXXXXX Oct 27 17:45:52 phoenix ample[6856]: BitRate is 128, FileLength is 4626099 Oct 27 17:45:52 phoenix ample[6856]: TPF is 0.026122 and BPF is 417.959184, we have 38.281250 frames in one second Oct 27 17:45:52 phoenix ample[6856]: died - failed to bind tcp socket , errno: Address already in use I was only trying to connect once with XMMS when this happened. It looks as if multiple copies get spawned and then race for the socket. I'm not overly keen on this stuff getting spewed into syslog to be honest - with 0.5.2 the logging was far less verbose. Either way, this doesn't look good for me at the moment, so I've reverted to 0.5.2 for now. Any ideas? (I've just joined the ample-devel mailing list, but it looks dormant since January... copied there anyway). Best wishes Andy -- Andy Piper - Farnborough, Hampshire (UK) and...@fr... http://www.andypiper.co.uk/ |
|
From: David <da...@2g...> - 2002-10-28 13:26:20
|
On Sun, Oct 27, 2002 at 06:30:26PM +0000, Andy Piper wrote: > Hi David, > > Great to see Ample is under development again, after your various woes! > <snip> Thanks =) Whops, the very verbose syslog stuff is really a bug, I didn't realize I'd left that there (since I'm always running in debug mode with a lot of extra info being "spewed" anyway). I'll look at your problem and the syslogging stuff and hopefully put out a new version in a few days. Regards, David |
|
From: David <da...@2g...> - 2002-10-28 13:37:26
|
Hi Andy, I'm having some problems recreating the problems you are seeing. Could you please try 0.5.5 again with debugging turned on and send me the output? If you are worried about contaminating your syslogs I can recommend adding --disable-syslog when you run the configure script before you compile Ample. Then you can add a logfile=/some/where.log directive to ample.conf. If you could do a test run with debugging turned on and send me your ample.conf and the log file (privately to me, no need to send it to the list) I'd appreciate it. Also, to make sure that there are no funky (damaged) MP3 files messing it up, could you try to create a dir with just one or two good MP3's in it and try to run Ample with only that dir. > Oct 27 17:45:52 phoenix ample[6856]: died - failed to bind tcp socket , > errno: Address already in use This is really weird, the TCP socket is only bound once by the parent process at the time Ample is started. Are you sure all instances of Ample were killed before you started it again? Regards, David |
|
From: David <da...@2g...> - 2002-11-01 11:47:35
Attachments:
patch.apiper
|
Andy Piper wrote: > Running ample out of xinetd seems to prevent this from working. > Currently I have no output going anywhere. Hi, I've looked into this and currently the support for (x)inetd is broken :-) I've attached a patch, could you please try applying it in the src dir of ample 0.5.5 and see if it helps? Regards, David |
|
From: Andy P. <and...@fr...> - 2002-11-08 11:05:43
|
On Fri, 2002-11-01 at 00:18, David H=C3=A4rdeman wrote: > I've looked into this and currently the support for (x)inetd > is broken :-) >=20 > I've attached a patch, could you please try applying it in > the src dir of ample 0.5.5 and see if it helps? Hmm, only a week's delay... sorry about that, things have been hectic. Well that's a definite improvement, in that I can now get to my music again... ------------------------------------------------------ The behaviour when running from xinetd is as follows - connect using XMMS: an error pops up with the message "Couldn't connect to host phoenix Server reported: Ample/0.5.5 started" (this is the same as before). Then the music starts playing. - connect using web browser: web page says "LOG[28523]: Ample/0.5.5 started LOG[28523]: Connection from 192.168.1.12:12270 accepted, currently 1/2 clients HTTP/1.1 200 OK Server: Ample/0.5.5 Accept-Range: none Content-Type: text/html Connection: close" Nothing appears in syslog or any other log.=20 ------------------------------------------------------ Running standalone, options -d 9 -t - connect using XMMS: DBG[28590]: Opened UDP socket, port 33190 DBG[28590]: Opened TCP socket, port 1234 LOG[28590]: Connection from 192.168.1.12:46720 accepted, currently 1/2 clients DBG[28590]: No-fork/inetd mode: connection from 192.168.1.12:46720 DBG[28590]: Requested path / DBG[28590]: Entering MP3-Metadata mode (range 1 - 465) DBG[28590]: In playrange with range 1 - 465 and order 0 DBG[28590]: Going to play Ben Folds Five - Crosstown Traffic (live).mp3 - index 67 - order 0 DBG[28590]: Playing file /home/andyp/music/Ben Folds Five - Crosstown Traffic (live).mp3 DIE[28590]: (errno: Connection reset by peer) Error writing to client - connect using web browser: nothing on client. Server reports DBG[28548]: Opened UDP socket, port 33190 DBG[28548]: Opened TCP socket, port 1234 LOG[28548]: Connection from 192.168.1.12:45696 accepted, currently 1/2 clients DBG[28548]: No-fork/inetd mode: connection from 192.168.1.12:45696 DBG[28548]: Requested path / DBG[28548]: Entering HTML mode Segmentation fault Hope that helps. --=20 Andy Piper - Farnborough, Hampshire (UK) and...@fr... http://www.andypiper.co.uk/ |
|
From: David <da...@2g...> - 2002-11-15 06:45:51
|
On Fri, Nov 08, 2002 at 11:07:48AM +0000, Andy Piper wrote: > > Hmm, only a week's delay... sorry about that, things have been hectic. > > Well that's a definite improvement, in that I can now get to my music > again... > Ok, good, thanks for the testing... I have a good idea what the cause of the rest of the problems are. I have exams coming up real soon though so I'll have to get back to you sometime next week :-) Regards, David |
|
From: David <da...@2g...> - 2002-11-21 14:54:31
|
On Fri, Nov 08, 2002 at 11:07:48AM +0000, Andy Piper wrote: > - connect using web browser: nothing on client. Server reports > DBG[28548]: Opened UDP socket, port 33190 > DBG[28548]: Opened TCP socket, port 1234 > LOG[28548]: Connection from 192.168.1.12:45696 accepted, currently 1/2 > clients > DBG[28548]: No-fork/inetd mode: connection from 192.168.1.12:45696 > DBG[28548]: Requested path / > DBG[28548]: Entering HTML mode > Segmentation fault Hi, Exams are over and I've spent some time sorting out which messages goes where dpending on how the program is started... Could you please try to grab the latest sources from the cvs server, compile that and see if it works better for you. And if you still get the segfault when connecting to the html pages (I can't reproduce it with CVS sources), then please run ample under gdb and mail me the output from a backtrace after it segfaults. Thanks, David |
Thanks for helping keep SourceForge clean.
X