Switches, Hubs, and Modems
cancel
Showing results for 
Search instead for 
Did you mean: 

Blocking "non-genuine" SFPs in newer firmwares

Nico_24
Occasional Advisor

Blocking "non-genuine" SFPs in newer firmwares

Hello,

I quote from the release notes for ProCurve 26xx switches:

"Log (PR_1000323790) . The switch detects a non-genuine ProCurve mini-GBIC as a port
self test failure and subsequently disables the link."

I guess you all know what the means: you will have to buy "genuine" HP SFPs(which do not differ from Finisar SFPs at all!).
I won't do that. I suppose we will simply stop buying anymore HP equipment and choose a different vendor. After all I'm talking about several hundrets of 26xx and 28xx switches.

There is nothing more to say but: Bye, bye HP, it was a nice time.

Regards,
Nico
7 REPLIES
PDnet
Advisor

Re: Blocking "non-genuine" SFPs in newer firmwares

Nico,
What is the reasoning behind using non-HP mini-GBICs? I have found them to be attractively priced and they certainly do work fine. Do you have another reason? Or is it just that you don't like being tied to a particular vendor? Have you ever tried calling Cisco TAC with a non-Cisco piece of hardware installed? Maintaining the integrity of the platform is the only way the vendors can provide the support we demand.

Curious about your thoughts.


stu...
Nico_24
Occasional Advisor

Re: Blocking "non-genuine" SFPs in newer firmwares

Right, they just work as fine as our almost thousand Finisar SFPs do. The HP SFPs do not differ in any way from the Finisar parts, it's just the code that makes the switch recognize it as a "genuine" part.
I don't know where you buy but we saved a lot of money by buying Finisar SFPs instead of "relabeled" Finisar SFPs from HP. If you ever need LX or LH, you will see the difference in price.

[quote]Maintaining the integrity of the platform is the only way the vendors can provide the support we demand.[/quote]

Oh please.. again: HP sells Finisar hardware with a different firmware but it is always the same product. So that has nothing to do with the integrity of the platform, it is just about money, as usual.

Please don't compare HP to Cisco, I know exactly why I am maintaining an almost 100% Cisco-free network(Juniper, Foundry & HP).

Nico
Matt Hobbs
Honored Contributor

Re: Blocking "non-genuine" SFPs in newer firmwares

I have given my opinion on this before and I believe a large part of this decision came from the time wasted simply dealing with problems that lower quality 3rd party components can cause in network.

I'm not saying that Finisar SFP's are in anyway worse, however there are also counterfeit Finisar SFP's out there that are of definite poor quality.

Unfortunately, because of the counterfeit products out there, everyone has been tarred by the same brush and a decision was made to reduce the ongoing costs associated with such activity.

One of the things that people enjoy about HP is the advanced exchange warranty process. If you have a faulty SFP, HP will send you a replacement next day with only a few basic questions.

If HP were to support all 3rd-party SFP's, then the only way I can see of dealing with that would be for people to first send back their faulty SFP's, then after it has been verified as an HP SFP, send the replacement. This is obviously time consuming and not acceptable when you have a network down.

In the past where 3rd party SFP's did work, I'm willing to bet a large percentage of faulty SFP's that came back to HP were non-genuine, either from people not being honest or simply being unaware that they did not have a geniune HP product. Either way this drives up support costs.
Matt Hobbs
Honored Contributor

Re: Blocking "non-genuine" SFPs in newer firmwares

By the way, with the absolute latest 2800 firmware 3rd party SFP's should not be disabled, only a warning in the event log:

ftp://ftp.hp.com/pub/networking/software/Mini-GBIC-Support-Oct2006.pdf

Similarly the 2600 depending on if it's a 2600A of B version, you may have better luck with the latest code.
Sergej Gurenko
Trusted Contributor

Re: Blocking "non-genuine" SFPs in newer firmwares

Hello Matt,

Can you please more details about changes in non-genuine Gbic policy? Do HP allows some Gbic flexibilities in a new firmwares on a newest models of switches?

Please share you findings. Thanks in advance.
Matt Hobbs
Honored Contributor

Re: Blocking "non-genuine" SFPs in newer firmwares

With the 2600 'A' versions, 2800's and 5300's, it looks as though a decision was made with the latest firmware to bring those switches back into line with how they originally would have operated with the first ever firmware releases for those products. I believe you will only receive a warning in the event-log now, but it will not disable them.

This looks to be a fairer decision as people would have purchased switches in the past when support for 3rd party gbics was more common. And then when updating to some of those in between versions of firmware, you would have been in for a surprise to find out that your gbics no longer worked!

With the current products though it has been well advertised now that only geniune HP mini-gbics will work and have been that way since day one. Therefore I cannot imagine any changes to those products in relation to 3rd party gbics.
sthen
Occasional Visitor

Re: Blocking "non-genuine" SFPs in newer firmwares

> By the way, with the absolute latest 2800 firmware 3rd party SFP's should not be disabled, only a warning in the event log:

That's great news.

> [...] Therefore I cannot imagine any changes to those products in relation to 3rd party gbics.

Unless HP will be offering CWDM gbics, that's quite a disadvantage for 3400cl compared to Cisco 3560/3750 for some users.