Hi community,
for a couple of days I have a weird problem and don't seem do be able to find the reason.
I'm running LMS 7.7.2 -r33893 on Win 7 for a while (2-3years, same machine) now, and things went smooth until recently. The machine is not a dedicated server - it is used for various things and every once in a while I install or remove software. I can't tell if the mess began with a change of the system.
Here is what happens:
Whenever I start LMS everything is nice at first - but after a couple of hours (2 - 24h) the same scenario starts. It begins with a line in the sever log saying that LMS is "Unable to get players":
(here a fraction of the log is shown: Server start yesterday evening 21:15h, the problem began today at 16:19h)
You can see, that after an initial phase of longer intervals between logs, it quickly converges to a state of one log statement every minute.
As soon as these messages start, LMS is not accessible anymore through players or WebUI.
Now things get increasingly worse and approx. two hours later, most messages end with "Too many open files"
If this state is reached, the control panel is also kind of dead.
I always run LMS as Administrator (run as) - and also installed it as Administrator (I reinstalled yesterday).
I made the observation, that this behaviour is caused by (or it causes) a wildly increasing number of TCP connections innitiated by LMS.
During the "Alive"-period there are something like 10 TCP (~ 2 per SB) connections from LMS shown in the Windows Ressource Monitor.
In parallel to the suspicious logging this number increases quickly and reaches amounts in the thousands.
Following the first bunch of connections as shown in the ressource monitor (TCP connections status page)
( 192.168.1.3: LMS - 192.168.1.1 : my Router)
Abbild PID Lokale Adresse Lokaler Port Remoteadresse Remoteport Paketverlust (%) Latenz (ms)
SQUEEZ~3.EXE 6328 192.168.1.3 9000 192.168.1.1 59277 - -
SQUEEZ~3.EXE 6328 192.168.1.3 9000 192.168.1.1 59278 - -
SQUEEZ~3.EXE 6328 192.168.1.3 9000 192.168.1.1 59260 - -
SQUEEZ~3.EXE 6328 192.168.1.3 9000 192.168.1.1 59261 - -
SQUEEZ~3.EXE 6328 192.168.1.3 9000 192.168.1.1 59235 - -
SQUEEZ~3.EXE 6328 192.168.1.3 9000 192.168.1.1 59234 - -
SQUEEZ~3.EXE 6328 192.168.1.3 9000 192.168.1.1 59226 - -
SQUEEZ~3.EXE 6328 192.168.1.3 9000 192.168.1.1 59227 - -
SQUEEZ~3.EXE 6328 192.168.1.3 9000 192.168.1.1 59221 - -
In total there are 17xx lines like this - all but the Remote port are identical. It seams like LMS is probing the full range of possible ports ...
If I stop LMS, the number of connections decreases to the expected number dependent on the other running processes.
If I restart LMS, LOG is silent for 5min, but the number of LMS TCP connections _immediately_ starts to go up
As LMS restart did not really help, I rebooted the router. And things look slightly better:
After subsequent LMS launch, the TCP connections do _not_ go up - at first ... But after ~20h the whole thing starts from the beginning: Log and TCP connections start to go wild.
Is the router broken? Does LMS kill the router? I can not tell. The scenario is independent from the firewall being on or off.
Any suggestions from a network guy?
I googled for the error messages again and again, because I thought i can't be the first who encounters that. And indeed I found a few posts with similar messages - but unfortunately, nowhere a solution was found.
Any suggestions? Help is urgently needed as LMS is not usable at all ...
Kind regards
Guenther
for a couple of days I have a weird problem and don't seem do be able to find the reason.
I'm running LMS 7.7.2 -r33893 on Win 7 for a while (2-3years, same machine) now, and things went smooth until recently. The machine is not a dedicated server - it is used for various things and every once in a while I install or remove software. I can't tell if the mess began with a change of the system.
Here is what happens:
Whenever I start LMS everything is nice at first - but after a couple of hours (2 - 24h) the same scenario starts. It begins with a line in the sever log saying that LMS is "Unable to get players":
(here a fraction of the log is shown: Server start yesterday evening 21:15h, the problem began today at 16:19h)
Code:
13-05-23 21:15:10.9871 main::init (354) Starting Logitech Media Server (v7.7.2, r33893, Wed Mar 14 05:47:42 PDT 2012) perl 5.014001
13-05-24 16:19:33.0045 Slim::Networking::Discovery::Players::_players_error (146) Unable to get players: Timed out waiting for data
13-05-24 16:39:33.0044 Slim::Networking::Discovery::Players::_players_error (146) Unable to get players: Timed out waiting for data
13-05-24 16:50:33.0039 Slim::Networking::Discovery::Players::_players_error (146) Unable to get players: Timed out waiting for data
13-05-24 16:51:33.0039 Slim::Networking::Discovery::Players::_players_error (146) Unable to get players: Timed out waiting for data
13-05-24 16:52:33.0043 Slim::Networking::Discovery::Players::_players_error (146) Unable to get players: Timed out waiting for data
13-05-24 16:53:33.0042 Slim::Networking::Discovery::Players::_players_error (146) Unable to get players: Timed out waiting for dataAs soon as these messages start, LMS is not accessible anymore through players or WebUI.
Now things get increasingly worse and approx. two hours later, most messages end with "Too many open files"
Code:
13-05-24 17:58:47.0050 Slim::Networking::Discovery::Players::_players_error (146) Unable to get players: Timed out waiting for data 13-05-24 17:59:11.5113 Slim::Plugin::UPnP::Events::error (323) Subscribe/unsubscribe error: 412 Precondition Failed
13-05-24 17:59:11.5828 Slim::Plugin::UPnP::Events::error (323) Subscribe/unsubscribe error: 412 Precondition Failed 13-05-24 17:59:23.0062 Slim::Networking::Discovery::Players::_players_error (146) Unable to get players: Timed out waiting for data
13-05-24 17:59:26.3078 Slim::Networking::SqueezeNetwork::Players::_players_error (334) Unable to get players from SN: Timed out waiting for data, retrying in 3600 seconds[13-05-24 18:00:22.0159] Slim::Networking::Discovery::Players::_players_error (146) Unable to get players: Timed out waiting for data
13-05-24 18:00:40.1627 Slim::Networking::SqueezeNetwork::_error (596) Unable to login to SN: Timed out waiting for data
13-05-24 18:00:40.1816 Slim::Networking::SqueezeNetwork::_error (596) Unable to login to SN: Timed out waiting for data
13-05-24 18:00:40.1819 Slim::Networking::SqueezeNetwork::Players::_players_error (334) Unable to get players from SN: Timed out waiting for data, retrying in 7200 seconds
13-05-24 18:01:22.0049 Slim::Networking::Discovery::Players::_players_error (146) Unable to get players: Timed out waiting for data
13-05-24 18:02:22.0053 Slim::Networking::Discovery::Players::_players_error (146) Unable to get players: Timed out waiting for data
13-05-24 18:03:22.0058 Slim::Networking::Discovery::Players::_players_error (146) Unable to get players: Timed out waiting for data
13-05-24 18:04:22.0062 Slim::Networking::Discovery::Players::_players_error (146) Unable to get players: Timed out waiting for data
13-05-24 18:05:22.0126 Slim::Networking::Discovery::Players::_players_error (146) Unable to get players: Timed out waiting for data
13-05-24 18:05:36.0254 Slim::Networking::SqueezeNetwork::_error (596) Unable to login to SN: Connect timed out: Too many open files
13-05-24 18:05:46.0004 Slim::Utils::Prefs::Namespace::savenow (339) Error: can't save C:\ProgramData\Squeezebox\prefs\server.prefs: Too many open files at /<C:\PROGRA~1\SQUEEZ~1\server\SQUEEZ~3.EXE>Slim/Utils/Prefs/Namespace.pm line 328.
13-05-24 18:06:12.0025 Slim::Networking::Discovery::Players::_players_error (146) Unable to get players: Connect timed out: Too many open files
13-05-24 18:07:12.0032 Slim::Networking::Discovery::Players::_players_error (146) Unable to get players: Connect timed out: Too many open files
13-05-24 18:08:12.0024 Slim::Networking::Discovery::Players::_players_error (146) Unable to get players: Connect timed out: Too many open files
13-05-24 18:09:12.0018 Slim::Networking::Discovery::Players::_players_error (146) Unable to get players: Connect timed out: Too many open files
13-05-24 18:10:12.0052 Slim::Networking::Discovery::Players::_players_error (146) Unable to get players: Connect timed out: Too many open filesI always run LMS as Administrator (run as) - and also installed it as Administrator (I reinstalled yesterday).
I made the observation, that this behaviour is caused by (or it causes) a wildly increasing number of TCP connections innitiated by LMS.
During the "Alive"-period there are something like 10 TCP (~ 2 per SB) connections from LMS shown in the Windows Ressource Monitor.
In parallel to the suspicious logging this number increases quickly and reaches amounts in the thousands.
Following the first bunch of connections as shown in the ressource monitor (TCP connections status page)
( 192.168.1.3: LMS - 192.168.1.1 : my Router)
Abbild PID Lokale Adresse Lokaler Port Remoteadresse Remoteport Paketverlust (%) Latenz (ms)
SQUEEZ~3.EXE 6328 192.168.1.3 9000 192.168.1.1 59277 - -
SQUEEZ~3.EXE 6328 192.168.1.3 9000 192.168.1.1 59278 - -
SQUEEZ~3.EXE 6328 192.168.1.3 9000 192.168.1.1 59260 - -
SQUEEZ~3.EXE 6328 192.168.1.3 9000 192.168.1.1 59261 - -
SQUEEZ~3.EXE 6328 192.168.1.3 9000 192.168.1.1 59235 - -
SQUEEZ~3.EXE 6328 192.168.1.3 9000 192.168.1.1 59234 - -
SQUEEZ~3.EXE 6328 192.168.1.3 9000 192.168.1.1 59226 - -
SQUEEZ~3.EXE 6328 192.168.1.3 9000 192.168.1.1 59227 - -
SQUEEZ~3.EXE 6328 192.168.1.3 9000 192.168.1.1 59221 - -
In total there are 17xx lines like this - all but the Remote port are identical. It seams like LMS is probing the full range of possible ports ...
If I stop LMS, the number of connections decreases to the expected number dependent on the other running processes.
If I restart LMS, LOG is silent for 5min, but the number of LMS TCP connections _immediately_ starts to go up
As LMS restart did not really help, I rebooted the router. And things look slightly better:
After subsequent LMS launch, the TCP connections do _not_ go up - at first ... But after ~20h the whole thing starts from the beginning: Log and TCP connections start to go wild.
Is the router broken? Does LMS kill the router? I can not tell. The scenario is independent from the firewall being on or off.
Any suggestions from a network guy?
I googled for the error messages again and again, because I thought i can't be the first who encounters that. And indeed I found a few posts with similar messages - but unfortunately, nowhere a solution was found.
Any suggestions? Help is urgently needed as LMS is not usable at all ...
Kind regards
Guenther