What is the best code to use on older 3750v2 and 3750e switches? I noticed Cisco seems to recommend 12.2.55(SE11). Is there any benefit to running 15.0.2(SE11)? I don't want to hit memory or CPU issues, which I did read about on early version of 15 a few years ago.
I recommend the following link, actually the version 15 was the newest for these devices, some vulnerabilities has been mitigated, also I suggest use the Cisco Navigation Tool in order to see the features of each other. Currently the new devices are using IOS XE. Now before to migrate from 12 to 15 should should see the hardware requeriments on the Cisco download website. Link
I would opt for v15 on all my 3750 platforms. For the 3750v2 go for the latest 15.0(2)SE11 and on the 3750e 15.2.2 or 15.2.4 depending on features or hardware support that you may be after.
12.2(55)SE11 is over a year old and has several vulnerabilities that are mitigated by upgrading to 15.x
Checking uptime on my core 3750 switch stack, its 3 years, 45 weeks, 3 days, 20 hours, 38 minutes (at the time of this writing, 11:53 AM EST, Saturday 9/23/2017).
Some 3 years, 45 weeks, 3 days ago we tried going from 12.2(55)SE8 to one of the 15.x train's that was out at the time, and it created an extremely unstable network. I was called in to assist our other network admin to break up the stack and downgrade each switch one at a time.
Since then we've been on 12.2(55)SE8 and the stack consists of the following members:
Switch Ports Model SW Version SW Image
------ ----- ----- ---------- ----------
* 1 52 WS-C3750G-48TS 12.2(55)SE8 C3750-IPSERVICESK9-M
2 52 WS-C3750V2-48TS 12.2(55)SE8 C3750-IPSERVICESK9-M
3 52 WS-C3750V2-48TS 12.2(55)SE8 C3750-IPSERVICESK9-M
5 52 WS-C3750V2-48TS 12.2(55)SE8 C3750-IPSERVICESK9-M
6 52 WS-C3750V2-48TS 12.2(55)SE8 C3750-IPSERVICESK9-M
7 54 WS-C3750X-48P 12.2(55)SE8 C3750E-UNIVERSALK9-M
8 54 WS-C3750X-48P 12.2(55)SE8 C3750E-UNIVERSALK9-M
9 30 WS-C3750X-24P 12.2(55)SE8 C3750E-UNIVERSALK9-M
That day we also removed switch 4 out of the stack since we had enough free ports on other switches to move devices over to. We thought maybe a stack of 9 switches just couldn't handle the 15.x train.
Our switch stack does ip routing, vtp server, multiple vlans, mac sticky security on most ports, has multiple trunk ports, has some SFP's installed, and is acting as a "collapesd core". There is a mix of 3750 switches as you can see above, one 3750G, 4 3750G2's and 3 3750X.
It was a real nightmarish event, hence once we got it back on 12.2(55)SE8, we never rebooted it. It's always been a thorn in our past so we purchased gear to create two core switch stacks from another vendor and will do vrrp between them. We will break up this big mixed 3750 stack and it will become an "access stack" where I will feel more comfortable upgrading to 15.0.2SE11 since it will likely only consist of the 3750x's.
I have another location with 3 3750x's and its currently running 15.0.2SE10a without issues. I have a mixed stack of 2 3750x's and 1 3750v2 running 15.0.2SE4 without issues (I probobly should upgrade).
I have another stack at another building of 2 3750x's running 12.2(55)SE8 but I'm considering upgrading that to 15.0.2SE11. I have it downloaded, just have to change the boot entry, write mem and reload it.
Actually my 3 3750x stack needs to stay on 12.2. I rely on ip sla on that stack and it is ipbase. It looks like in feature navigator you can only use ip sla responder in ipbase in the 15.x train. That's sad.
Anyway I'll upgrade 12.2.55 SE8 to SE11.
I came across a bug recently where the v12 erroneously reports discards. This behaviour is present on the most recent v12 build, in spite of the fact it has been cured on recent v15 builds.