I followed advice on here and bought a 2nd user 2960 which made a significant difference to the sound of my system, but made file copying, backups, etc very slow compared to it's Netgear predecessor. I therefore have obtained a 2960G (Gigabit) which has restored copying speed and with which I'm happy. I also got myself a conole cable (and USB adapter).
I have made a few settings changes, both from the CLI/console and CNA tool. I've got SPANing between two ports enabled so I can do wireshark traces, I've set the timezone, DST, time, enabled the web interface and set the enable password. I even remembered to save the running config to the startup config 
I was even brave enough to update the firmware
I also set NTP server, but it grumbles it can't see that so I need to investigate what's wrong there - probably need to change a firewall rule.
I made the same changes to the Fast Ethernet version.
This set me thinking though - are there any recommended settings for these devices to optimise either their operation in a home environment, or that might alter (improve) the sound coming out of the speakers ? Mine is connected to my router, NAS and Nova so it's purely for the audio parts of my network.
The obvious other question is whether there are any things you definately shouldn't set ?
I've done a forum search and haven't found a simple guide, though there are bits spread across multiple posts, hence the new thread.
(Useful hint - if you're going to use the CNA tool make sure you download at least v5.8 - the default link from the Cisco documentation takes you to v5.6, which won't let you log in via the CNA tool - keeps grumbling about level 15 access and won't accept any username/password (even if they are correct). Or at least both mine do this.)
Posted on: 06 July 2018 by Simon-in-Suffolk
There is very little I do on the 2960 to ‘optimise’ things... but I do do the following
1. Set the port with the Naim streamer on to Portfast Enable... this allows the port to come up straightaway, which helps the slightly less than fully reliable Naim implementation of client DHCP on their streamers... obviously this removes loop protection, but not needed if the Naim streamer remains on that port.
2. I set up an IGMP Querier on the switch... this allows all the hosts on on IP group 235.255.255.250 to be kept fresh and alive when other devices on your network are IGMP snooping, or have a short ARP timeout.. this allows the Naim app to be instaneously quick and responsive. (It improves what some on this forum call ‘network discovery’). I find IGMPv2 seems to work best with Naim.
3. The uplink from the switch I aggregate, this helps minimise changes to interframe timing when other traffic is being sent to your switch... this probably is more specific to my implementation.
4. I leave frame delivery set to store and forward, as opposed to using cut through.
Other than that all fairly standard... and yes NTP works fine on my setup... though I set a single switch as my stratum n device, and the other devices point to that and are thus stratum n+1
I configure the switch via IOS commands in the running config file via the CLI and a telnet connection.
Posted on: 06 July 2018 by TallGuy
My NTP problem was that the server I selected doesn’t appear to exist any more, I’ve swapped it to one which is live and all is now well. So that’s one to user stupidity.
Thanks for the other tips - Im not sure my basic network needs aggregation, but I’ll make a mental note to give it a go should the network become busier. I’ll give 1 and 2 a go and have a look at 4 (and read up on them, I do like to understand what I’m doing and why).
I’m generally a GUI person nowadays, but was driven to CLI when I couldn’t get in via CNA and I haven’t enabled telnet (as far as I know...). Quite exciting using that - learnt far more than I would have via a GUI.
Posted on: 06 July 2018 by Simon-in-Suffolk
Yes the GUI offers in my experience a limited subset of capabilities.. looks pretty but for me if programming the device there is no alternative to IOS, where you have full control. Enabling telnet is pretty straight forward.. here is a CCNA study guide showing how how to setup telnet on the vty lines.
http://www.ciscopress.com/arti...p=27650&seqNum=4
The aaggregation is more about concurrent transmission of frames rather than overall throughput, thereby reducing serialisation timing variation... all very minimal, but worth playing around with... clearly you might need a bit if trial and error to find the optimum aggregation parameters to separate the flows between the streamer and other hosts.
