Cisco advisory: Ethernet cable may press factory reset switch | Hacker News Reader

Cisco advisory: Ethernet cable may press factory reset switch

thenextweb.com

41 pointsJoshTriplett12 comments

I'm lost on why any enterprise kit at all has a reset button. This isn't a Netgear router where the ISP regularly asks people to "press reset" when they forget their password.

I was happy to use a console cable and send a break when I rarely needed to reset Cisco kit. Why these sorts of buttons ever existed I don't understand.

General idea here is that when you have a problematic piece of gear in somebody else's DC across the country, it is usually easier and faster to ask "remote hands" to press and hold the button, compared to asking to get a laptop, serial cable, fire up a terminal, join webex with his laptop and only then try to send break or whatever.

Paper clip reset buttons are so 1990?

> The cables, which are sometimes accidentally used in datacenters, feature a protective boot that sticks out over the top to ensure the release tab isn’t accidentally pressed or broken off

Is this just a typo (misplaced "accidentally"?), or is there really a reason such cables shouldn't be used in datacenters?

It just seems like a poorly written article. Later on it says

> Such a situation could cause a problem in any size datacenter, where these switches and cables are commonly used

which seems to contradict the "accidentally". Similarly, this bit is ungrammatical:

> It’s amazing that Cisco didn’t catch this before the device was released, let alone that the ‘fix’ for the problem which suggests using a different cable or cutting off the boot.

This sort of stuff happens all the time. Usually people just facepalm and get on with it rather than making a headline...

I'm not sure. There might be two design issues here:

1) placing the reset button directly above a port, making it easier to be hit by anything that happens to poke up above the port. Moving the button to one side would be logical.

2) making the reset button stick out. On consumer routers, the reset switch is usually recessed in a hole to make it harder to accidentally hit. Of course, the requirement for a reset switch for a device in a datacentre might be different.

Doing 1) or 2) would prevent the problem.

Of course. However DC kit has a long history of fail. Even old sun kit would drop to the monitor if you yanked the console cable. This on one occasion killed an entire e4500 cluster we had running oracle which failed over and then resumed crashing both instances. This happened due to a port glitch on the console server.

Perhaps I'm just desensitised to the issue.

The Cisco Field Notice implies that it is a problem with excessively large protectors: "Numerous types of snagless cables are available. However, most are designed in such a way that the protective device is less prominent and/or held more tightly to the tab, which prevents the aforementioned drawbacks."

It's an advisory from 2013. I wonder why this is surfacing nearly two years later.

Yeah, I am also curious about this.

http://www.cisco.com/c/en/us/support/docs/field-notices/636/...

It resurfaced on Twitter over the weekend, went viral, and was submitted to HN.

Reply on news.ycombinator.com