#254110 - 13/04/2005 17:05
Note to self: hijack/player prob uploading lots with jempeg 69.0...01
|
enthusiast
Registered: 11/06/2003
Posts: 384
|
Uploading about 8.5GB (1800+ ogg vorbis songs) with jemplode 69.0...01.
v3alpha8, hijack 422, freshly reformatted drive.
Code:
kftpd: ksock_rw(response) failed, rc=0
neighbour table overflow
server_protocol_handler.cpp:1390:Timeout during synchronise, restarting player
Restored terminal settings
(Edit: fixed markup -wfaulk)
Don't have time to search the board about this now, but wanted to preserve the info someplace where I could find it again.
--Nathan
Edited by wfaulk (13/04/2005 19:15)
|
|
Top
|
|
|
|
|
#254111 - 13/04/2005 19:10
Re: Note to self: hijack/player prob uploading lots with jempeg 69.0..
[Re: Mataglap]
|
carpal tunnel
Registered: 29/08/2000
Posts: 14512
Loc: Canada
|
Mmm.. that's very odd. Hijack kftpd even already has a retry loop to handle that situation, and *that* is failing here.
[EDIT]: Oh, wait. I see a flaw in the error checking in kftpd.
Fixed. v423 shortly. It may not fix the problem, but if it happens again then at least we'll find out why.
Cheers
Edited by mlord (13/04/2005 19:31)
|
|
Top
|
|
|
|
|
#254112 - 13/04/2005 23:12
Re: Note to self: hijack/player prob uploading lots with jempeg 69.0..
[Re: mlord]
|
enthusiast
Registered: 11/06/2003
Posts: 384
|
Well, damn. I guess I won't be digging into the board's search function tonight. Thanks Mark!
If I remember right, the neighbor table is the arp cache, right? Why would that overflow?
And thants wfaulk for the UBB code fix!
And where's self? I hoped to flush him out of hiding. 
--Nathan
|
|
Top
|
|
|
|
|
#254113 - 13/04/2005 23:23
Re: Note to self: hijack/player prob uploading lots with jempeg 69.0..
[Re: Mataglap]
|
carpal tunnel
Registered: 29/08/2000
Posts: 14512
Loc: Canada
|
If I remember right, the neighbor table is the arp cache, right? Why would that overflow?
Dunno. But perhaps the real error is something different, and the neighbor table thing just happens as a result of kftpd not correctly handling the real error.
Cheers
|
|
Top
|
|
|
|
|
#254114 - 13/04/2005 23:41
Re: Note to self: hijack/player prob uploading lots with jempeg 69.0..
[Re: mlord]
|
enthusiast
Registered: 11/06/2003
Posts: 384
|
True, true. v423 installed, and I've got 40GB+ to reload. So there's plenty of time to see if something goes wrong again.
(BTW, I'm just gonna say it's amazing what the sparse superblock does to speed up an fsck of an 80GB drive.)
|
|
Top
|
|
|
|
|
#254115 - 14/04/2005 07:14
Re: Note to self: hijack/player prob uploading lots with jempeg 69.0..
[Re: Mataglap]
|
carpal tunnel
Registered: 18/01/2000
Posts: 5688
Loc: London, UK
|
Quote:
If I remember right, the neighbor table is the arp cache, right? Why would that overflow?
At the bottom of this page, it says to issue "ifconfig lo 127.0.0.1" to fix some instances of the "neighbour table overflow" error.
Odd. I don't know why.
_________________________
-- roger
|
|
Top
|
|
|
|
|
#254116 - 14/04/2005 15:15
Re: Note to self: hijack/player prob uploading lots with jempeg 69.0..
[Re: Mataglap]
|
enthusiast
Registered: 11/06/2003
Posts: 384
|
Picking up where I left off yesterday... I'm now seeing a constant series of output like this: Code:
ksock_rw(): error: 16
ksock_rw(): error: 12
ksock_rw(): error: 17
ksock_rw(): error: 12
ksock_rw(): error: 17
ksock_rw(): error: 8
ksock_rw(): error: 11
ksock_rw(): error: 24
ksock_rw(): error: 13
ksock_rw(): error: 30
ksock_rw(): error: 21
ksock_rw(): error: 23
ksock_rw(): error: 11
ksock_rw(): error: 30
ksock_rw(): error: 30
ksock_rw(): error: 28
ksock_rw(): error: 11
ksock_rw(): error: 8
ksock_rw(): error: 18
ksock_rw(): error: -104
kftpd: ksock_rw() failed, rc=-104
Which seem to correspond to the end of each transfer activity in jEmplode, i.e., at the end of transferring a song and at the end of transfering a playlist change. The upload seems to keep on trucking though. My scrollback buffer wasn't big enough to capture what happened first, so when I start the next batch, I'll make sure to log to disk. --Nathan
|
|
Top
|
|
|
|
|
#254117 - 14/04/2005 15:16
Re: Note to self: hijack/player prob uploading lots with jempeg 69.0..
[Re: Roger]
|
enthusiast
Registered: 11/06/2003
Posts: 384
|
Roger,
Thanks for that tip. the loopback i/f is up and running, so maybe that forces some kind of crazy arp cache reinitialization?
--Nathan
|
|
Top
|
|
|
|
|
#254118 - 14/04/2005 15:35
Re: Note to self: hijack/player prob uploading lots with jempeg 69.0..
[Re: Mataglap]
|
enthusiast
Registered: 11/06/2003
Posts: 384
|
Also seeing the occasional slightly different one: Code:
ksock_rw(): error: 16
ksock_rw(): error: 12
ksock_rw(): error: 17
ksock_rw(): error: 12
ksock_rw(): error: 17
ksock_rw(): error: 8
ksock_rw(): error: 11
ksock_rw(): error: 24
ksock_rw(): error: 13
ksock_rw(): error: 30
ksock_rw(): error: 21
ksock_rw(): error: 23
ksock_rw(): error: 11
ksock_rw(): error: 30
ksock_rw(): error: 30
ksock_rw(): error: 28
ksock_rw(): error: 11
ksock_rw(): error: 8
ksock_rw(): error: -32
kftpd: ksock_rw(response) failed, rc=-32
ksock_rw(): error: -104
kftpd: ksock_rw() failed, rc=-104
|
|
Top
|
|
|
|
|
#254119 - 14/04/2005 18:37
Re: Note to self: hijack/player prob uploading lots with jempeg 69.0..
[Re: Mataglap]
|
carpal tunnel
Registered: 29/08/2000
Posts: 14512
Loc: Canada
|
Okay, Try adding a ReserveCache=16 line to the players config.ini file on the empeg.
Cheers
|
|
Top
|
|
|
|
|
#254120 - 14/04/2005 18:39
Re: Note to self: hijack/player prob uploading lots with jempeg 69.0..
[Re: mlord]
|
carpal tunnel
Registered: 29/08/2000
Posts: 14512
Loc: Canada
|
Oh, wait. I messed up with v423.
Trying again.. v424 shortly.
-ml
|
|
Top
|
|
|
|
|
#254121 - 14/04/2005 19:05
Re: Note to self: hijack/player prob uploading lots with jempeg 69.0..
[Re: mlord]
|
enthusiast
Registered: 11/06/2003
Posts: 384
|
Quote:
Try adding a ReserveCache=16
Shoot, I've completely forgotten to do that when I reformated the drive earlier this week.
Will do that and v424 after the current batch finishes uploading.
|
|
Top
|
|
|
|
|
#254122 - 14/04/2005 20:35
Re: Note to self: hijack/player prob uploading lots with jempeg 69.0...01
[Re: Mataglap]
|
new poster
Registered: 07/01/2004
Posts: 13
|
|
|
Top
|
|
|
|
|
#254123 - 14/04/2005 20:42
Re: Note to self: hijack/player prob uploading lots with jempeg 69.0...01
[Re: self]
|
carpal tunnel
Registered: 15/08/2000
Posts: 4859
Loc: New Jersey, USA
|
Thank you. I was waiting for that...
_________________________
Paul Grzelak
200GB with 48MB RAM, Illuminated Buttons and Digital Outputs
|
|
Top
|
|
|
|
|
#254124 - 14/04/2005 20:47
Re: Note to self: hijack/player prob uploading lots with jempeg 69.0..
[Re: Mataglap]
|
enthusiast
Registered: 11/06/2003
Posts: 384
|
Alright, with v424 & the reservecache=16, using jEmplode to make both changes. Rebooted the empeg and then connected with jEmplode over the network. Immediately get this before any attemps to load music. The last batch of song loading didn't crash out at any point, but still getting lots of the ksock error messages., Code:
hijack: removed menu entry: "Serial Port Assignment"
khttpd: listening on port 80
kftpd: listening on port 21
Using non-standard cache size 84 (bonus 0Mb, adjustment 16)
player.cpp : 579:empeg-car 3.00-alpha8 2004/07/07.
Prolux 4 empeg car - 2.1490 Jul 7 2004
Vcb: 0x4063a000
ksock_rw(recv): error: -104
kftpd: ksock_rw() failed, rc=-104
ksock_rw(recv): error: -104
kftpd: ksock_rw() failed, rc=-104
ksock_rw(recv): error: -104
kftpd: ksock_rw() failed, rc=-104
ksock_rw(recv): error: -104
kftpd: ksock_rw() failed, rc=-104
Thanks again, Mark, for continuing to work with me on this! best, --Nathan
|
|
Top
|
|
|
|
|
#254125 - 14/04/2005 20:48
Re: Note to self: hijack/player prob uploading lots with jempeg 69.0..
[Re: pgrzelak]
|
enthusiast
Registered: 11/06/2003
Posts: 384
|
Quote:
Thank you. I was waiting for that...
Me too!
|
|
Top
|
|
|
|
|
#254126 - 14/04/2005 21:05
Re: Note to self: hijack/player prob uploading lots with jempeg 69.0..
[Re: Mataglap]
|
enthusiast
Registered: 11/06/2003
Posts: 384
|
And as soon as I try to load the next batch of songs, something hangs.
jEmplode is showing 20% complete on the very first item in the transfer activity list, "add g-h", which is the parent playlist I imported, and eventually jEmplode decided that it wasn't connected to anything. See attachment. (Before the popup window, the status of add g-h was 20%, and the completion bars reflected that.)
EDIT: Oh, and the VFD is still showing the moving synchronizing display, so presumably the player process is still active?
The console showed these additional lines right after the startup stuff I quoted a few posts back: Code:
ksock_rw(recv): error: -104
kftpd: ksock_rw() failed, rc=-104
ksock_rw(recv): error: -104
kftpd: ksock_rw() failed, rc=-104
Adding Swap: 16028k swap-space (priority -1)
e2fsck 1.19, 13-Jul-2000 for EXT2 FS 0.5b, 95/08/09
I telnet'ed into the unit and grabed this stuff:
Code:
Configured for mataglap-work.
Configuring loopback interface...done.
Setting hostname to mataglap...done.
Running ntpdate to update clock...
14 Apr 15:51:40 ntpdate[55]: adjust time server 192.168.168.90 offset -0.100356 sec
mataglap:/# cat /proc/meminfo
total: used: free: shared: buffers: cached:
Mem: 15241216 14495744 745472 3256320 393216 3964928
Swap: 16412672 0 16412672
MemTotal: 14884 kB
MemFree: 728 kB
MemShared: 3180 kB
Buffers: 384 kB
Cached: 3872 kB
SwapTotal: 16028 kB
SwapFree: 16028 kB
mataglap:/# cat /proc/mounts
/dev/root / ext2 ro 0 0
/dev/hda2 /programs0 ext2 ro 0 0
none /proc proc rw 0 0
/dev/hda4 /drive0 ext2 rw 0 0
mataglap:/# ps -ef
UID PID PPID C STIME TTY TIME CMD
root 1 0 0 15:40 ? 00:00:05 init
root 2 1 0 15:40 ? 00:00:00 [kflushd]
root 3 1 0 15:40 ? 00:00:00 [kupdate]
root 4 1 0 15:40 ? 00:00:00 [kpiod]
root 5 1 0 15:40 ? 00:00:00 [kswapd]
root 6 1 0 15:40 ? 00:00:00 [kftpd]
root 7 1 0 15:40 ? 00:00:00 [khttpd]
root 8 1 0 15:40 ? 00:00:00 [menuexec]
root 11 1 0 15:40 ? 00:00:00 /programs0/bin/telnetd
root 12 1 0 15:40 ttyS1 00:00:02 /empeg/bin/player
root 13 12 0 15:40 ttyS1 00:00:00 /empeg/bin/player
root 14 13 0 15:40 ttyS1 00:00:00 /empeg/bin/player
root 15 13 0 15:40 ttyS1 00:00:00 /empeg/bin/player
root 16 13 0 15:40 ttyS1 00:00:03 /empeg/bin/player
root 17 13 0 15:40 ttyS1 00:00:00 /empeg/bin/player
root 18 13 0 15:40 ttyS1 00:00:00 /empeg/bin/player
root 20 13 0 15:40 ttyS1 00:00:00 /empeg/bin/player
root 21 13 0 15:40 ttyS1 00:00:01 /empeg/bin/player
root 22 13 0 15:40 ttyS1 00:00:00 /empeg/bin/player
root 23 13 0 15:40 ttyS1 00:00:00 /empeg/bin/player
root 24 13 0 15:40 ttyS1 00:00:00 /empeg/bin/player
root 25 13 0 15:40 ttyS1 00:00:00 /empeg/bin/player
root 26 13 0 15:40 ttyS1 00:00:01 /empeg/bin/player
root 27 13 0 15:40 ttyS1 00:00:00 /empeg/bin/player
root 28 13 3 15:40 ttyS1 00:00:27 /empeg/bin/player
root 34 6 0 15:50 ? 00:00:00 [kftpd <defunct>]
root 36 11 0 15:51 ttyp0 00:00:00 /bin/sh --login
root 58 36 9 15:53 ttyp0 00:00:00 ps -ef
mataglap:/#
Off to reboot and fsck.
--Nathan
Edited by Mataglap (14/04/2005 21:09)
|
|
Top
|
|
|
|
|
#254127 - 14/04/2005 22:59
Re: Note to self: hijack/player prob uploading lots with jempeg 69.0..
[Re: Mataglap]
|
enthusiast
Registered: 11/06/2003
Posts: 384
|
The batch that failed earlier (2gb, ~520 songs) finished successfully this time. Still getting the same ksock errors after every song or playlist modification. There was one -32 error in a sea of -104's.
--Nathan
|
|
Top
|
|
|
|
|
#254128 - 15/04/2005 01:07
Re: Note to self: hijack/player prob uploading lots with jempeg 69.0..
[Re: Mataglap]
|
carpal tunnel
Registered: 29/08/2000
Posts: 14512
Loc: Canada
|
The -104 is "normal".. I'll filter that out in v425.
-ml
|
|
Top
|
|
|
|
|
#254129 - 15/04/2005 11:52
Re: Note to self: hijack/player prob uploading lots with jempeg 69.0..
[Re: mlord]
|
enthusiast
Registered: 11/06/2003
Posts: 384
|
Much thanks Mark!
--Nathan
|
|
Top
|
|
|
|
|
|
Previous Topic
Index
