Re: PC 'Stuttering' after UMS running for a while
Posted: Tue Apr 28, 2020 2:33 am
As you said, most things looks good in the log file. I do however see something I find somewhat unusual:
There are quite a lot of these, and I have read many such logs in my time. I can't remember having seen this. It also looks to me like it receives an unusual amount of UPnP "NOTIFY" messages. These are part of the UPnP discovery protocol (SSDP), and one theory could be that UPnP/SSDP interface is constantly being reset, resulting in UMS broadcasting its presence and receiving replies much too often.
That said, I don't have any good explanation of why this would happen. It must be said that I haven't run UMS since 6.x, so there might have been some new bugs introduced since then. I am also doubtful to whether such a bug and the resulting traffic would be enough to cause the behavior you describe. I would think that it would take much more than that, but it's hard to know exactly what such a network interface initialization potentially might include on the OS level. If a socket is reserved each time, it's possible that significant OS resources would be used for this, but I still find it somewhat unlikely.
Except for this, I really don't have any good idea of what might be causing this. I would try a much older version and see if the problem is the same, preferably 6.x or 7.x. With these versions you also need to have Java installed, which could solve some potential issues between Windows Defender (or other such "scanners") and Java.
Code: Select all
TRACE 2020-04-27 02:29:04.202 [UPNP-AliveMessageSender] net.pms.network.UPNPHelper Setting SSDP network interface: name:eth2 (Broadcom NetLink (TM) Gigabit Ethernet)
That said, I don't have any good explanation of why this would happen. It must be said that I haven't run UMS since 6.x, so there might have been some new bugs introduced since then. I am also doubtful to whether such a bug and the resulting traffic would be enough to cause the behavior you describe. I would think that it would take much more than that, but it's hard to know exactly what such a network interface initialization potentially might include on the OS level. If a socket is reserved each time, it's possible that significant OS resources would be used for this, but I still find it somewhat unlikely.
Except for this, I really don't have any good idea of what might be causing this. I would try a much older version and see if the problem is the same, preferably 6.x or 7.x. With these versions you also need to have Java installed, which could solve some potential issues between Windows Defender (or other such "scanners") and Java.