Mailing List Archive: FCX stack problem and maybe move to 8?

FCX stack problem and maybe move to 8?

Dec 9, 2015, 2:56 AM

Post #1 of 6 (1504 views)

I have an F-CX stack of two which exhibits a problem when running 7.4 code
(even latest 7400m). The problem doesn't exhibit with 7300e.

(Summary of problem is that one of the members will fall out of the stack,
after a bit more than a day (although it managed to do it in 10 mins
earlier). Need to reboot both members to get it all working again.)

However, the reason I'm here really is to ask about the 8 code train,
which I have jumped to on F-CX. I have an opportunity in the next couple
of weeks. But there are so many to choose from! 8001, 8010, 8020, 8030.
How is one to decide?

So I'm looking for any useful hints on what has proven stable and a good
bet (assuming it makes my stacking problem go away at all) - or what is
best avoided. Then I can home in on some release notes. I've a note
saved from a few months ago on this list: "8.0.30b is out (aug 2015), and
a better choice if you are looking to upgrade to an 8.0.30 train. There
were a large number of defects closed with that release. 8.0.20d is due
out very soon as well for those that want to venture past the 8.0.10
code.".

Scenario is fairly simple, some L2 VLANs, L3 ves, some OSPF, quite a lot
of multicast traffic and a few acls.

Jethro.

. . . . . . . . . . . . . . . . . . . . . . . . .
Jethro R Binks, Network Manager,
Information Services Directorate, University Of Strathclyde, Glasgow, UK

The University of Strathclyde is a charitable body, registered in
Scotland, number SC015263.
_______________________________________________
foundry-nsp mailing list
foundry-nsp@puck.nether.net
http://puck.nether.net/mailman/listinfo/foundry-nsp

Re: FCX stack problem and maybe move to 8? [ In reply to ]

jethro.binks at strath

Dec 9, 2015, 2:59 AM

Post #2 of 6 (1496 views)

Permalink

On Wed, 9 Dec 2015, Jethro R Binks wrote:

> However, the reason I'm here really is to ask about the 8 code train,
> which I have jumped to on F-CX.

Apologies, this should read "which I have not jumped to yet on F-CX".

Jethro.

. . . . . . . . . . . . . . . . . . . . . . . . .
Jethro R Binks, Network Manager,
Information Services Directorate, University Of Strathclyde, Glasgow, UK

The University of Strathclyde is a charitable body, registered in
Scotland, number SC015263.
_______________________________________________
foundry-nsp mailing list
foundry-nsp@puck.nether.net
http://puck.nether.net/mailman/listinfo/foundry-nsp

Re: FCX stack problem and maybe move to 8? [ In reply to ]

mkallen at gmail

Dec 9, 2015, 10:25 AM

Post #3 of 6 (1493 views)

Permalink

Hi Jethro, current target patch for the FCX is 8.0.10h, though the latest
patch is 'm' (target path guidance usually lags the release schedule a
bit). That said, all of the development efforts are going into 8.0.30 and
above, the latest patch there is 8.0.30d. I have seen pretty good results
on this code for ICX, but not a lot of visibility for FCX running it.

Mike

On Wed, Dec 9, 2015 at 2:59 AM, Jethro R Binks <jethro.binks@strath.ac.uk>
wrote:

> On Wed, 9 Dec 2015, Jethro R Binks wrote:
>
> > However, the reason I'm here really is to ask about the 8 code train,
> > which I have jumped to on F-CX.
>
> Apologies, this should read "which I have not jumped to yet on F-CX".
>
> Jethro.
>
> . . . . . . . . . . . . . . . . . . . . . . . . .
> Jethro R Binks, Network Manager,
> Information Services Directorate, University Of Strathclyde, Glasgow, UK
>
> The University of Strathclyde is a charitable body, registered in
> Scotland, number SC015263.
> _______________________________________________
> foundry-nsp mailing list
> foundry-nsp@puck.nether.net
> http://puck.nether.net/mailman/listinfo/foundry-nsp
>

Re: FCX stack problem and maybe move to 8? [ In reply to ]

Christopher-Howard at utc

Dec 9, 2015, 10:30 AM

Post #4 of 6 (1496 views)

Permalink

We have more ICX than FCX, but every single one of both models are currently on 8.0.30a. I've not had any trouble out of this code for either line of switch. Been on that revision since the day of release (really needed MAC address notification traps).

-Christopher

On Dec 9, 2015, at 1:25 PM, Mike Allen <mkallen@gmail.com<mailto:mkallen@gmail.com>> wrote:

Hi Jethro, current target patch for the FCX is 8.0.10h, though the latest patch is 'm' (target path guidance usually lags the release schedule a bit). That said, all of the development efforts are going into 8.0.30 and above, the latest patch there is 8.0.30d. I have seen pretty good results on this code for ICX, but not a lot of visibility for FCX running it.

Mike

On Wed, Dec 9, 2015 at 2:59 AM, Jethro R Binks <jethro.binks@strath.ac.uk<mailto:jethro.binks@strath.ac.uk>> wrote:
On Wed, 9 Dec 2015, Jethro R Binks wrote:

> However, the reason I'm here really is to ask about the 8 code train,
> which I have jumped to on F-CX.

Apologies, this should read "which I have not jumped to yet on F-CX".

Jethro.

. . . . . . . . . . . . . . . . . . . . . . . . .
Jethro R Binks, Network Manager,
Information Services Directorate, University Of Strathclyde, Glasgow, UK

The University of Strathclyde is a charitable body, registered in
Scotland, number SC015263.
_______________________________________________
foundry-nsp mailing list
foundry-nsp@puck.nether.net<mailto:foundry-nsp@puck.nether.net>
http://puck.nether.net/mailman/listinfo/foundry-nsp

_______________________________________________
foundry-nsp mailing list
foundry-nsp@puck.nether.net<mailto:foundry-nsp@puck.nether.net>
http://puck.nether.net/mailman/listinfo/foundry-nsp

Re: FCX stack problem and maybe move to 8? [ In reply to ]

salmanravian at gmail

Dec 14, 2015, 10:53 PM

Post #5 of 6 (1470 views)

Permalink

Hi Jethro,
We have a lot of FCX switches. Had a lot of issues with 7.4d (dhcp
snooping bug), 7.4f (stack sync problem; stack CPU-CPU communication
failure), 8.1d (high CPU due to multicast) and 8.1g (telnet stops working).
Finally upgraded to 8.1m with UPnP blocked via an ACL. 8.1m has been stable
for nearly 3 months now.

BR

On Wed, Dec 9, 2015 at 10:30 PM, Howard, Christopher <
Christopher-Howard@utc.edu> wrote:

> We have more ICX than FCX, but every single one of both models are
> currently on 8.0.30a. I've not had any trouble out of this code for either
> line of switch. Been on that revision since the day of release (really
> needed MAC address notification traps).
>
> -Christopher
>
> On Dec 9, 2015, at 1:25 PM, Mike Allen <mkallen@gmail.com> wrote:
>
> Hi Jethro, current target patch for the FCX is 8.0.10h, though the latest
> patch is 'm' (target path guidance usually lags the release schedule a
> bit). That said, all of the development efforts are going into 8.0.30 and
> above, the latest patch there is 8.0.30d. I have seen pretty good results
> on this code for ICX, but not a lot of visibility for FCX running it.
>
> Mike
>
> On Wed, Dec 9, 2015 at 2:59 AM, Jethro R Binks <jethro.binks@strath.ac.uk>
> wrote:
>
>> On Wed, 9 Dec 2015, Jethro R Binks wrote:
>>
>> > However, the reason I'm here really is to ask about the 8 code train,
>> > which I have jumped to on F-CX.
>>
>> Apologies, this should read "which I have not jumped to yet on F-CX".
>>
>> Jethro.
>>
>> . . . . . . . . . . . . . . . . . . . . . . . . .
>> Jethro R Binks, Network Manager,
>> Information Services Directorate, University Of Strathclyde, Glasgow, UK
>>
>> The University of Strathclyde is a charitable body, registered in
>> Scotland, number SC015263.
>> _______________________________________________
>> foundry-nsp mailing list
>> foundry-nsp@puck.nether.net
>> http://puck.nether.net/mailman/listinfo/foundry-nsp
>>
>
> _______________________________________________
> foundry-nsp mailing list
> foundry-nsp@puck.nether.net
> http://puck.nether.net/mailman/listinfo/foundry-nsp
>
>
>
> _______________________________________________
> foundry-nsp mailing list
> foundry-nsp@puck.nether.net
> http://puck.nether.net/mailman/listinfo/foundry-nsp
>

Re: FCX stack problem and maybe move to 8? [ In reply to ]

youssef at 720

Dec 14, 2015, 11:40 PM

Post #6 of 6 (1471 views)

Permalink

Dear Salman,

Could you please elaborate more as to your environnent and how you use them
? Pure L2 or L3 routing ?

BR.

2015-12-15 7:53 GMT+01:00 salman sadiq <salmanravian@gmail.com>:

> Hi Jethro,
> We have a lot of FCX switches. Had a lot of issues with 7.4d
> (dhcp snooping bug), 7.4f (stack sync problem; stack CPU-CPU communication
> failure), 8.1d (high CPU due to multicast) and 8.1g (telnet stops working).
> Finally upgraded to 8.1m with UPnP blocked via an ACL. 8.1m has been stable
> for nearly 3 months now.
>
> BR
>
> On Wed, Dec 9, 2015 at 10:30 PM, Howard, Christopher <
> Christopher-Howard@utc.edu> wrote:
>
>> We have more ICX than FCX, but every single one of both models are
>> currently on 8.0.30a. I've not had any trouble out of this code for either
>> line of switch. Been on that revision since the day of release (really
>> needed MAC address notification traps).
>>
>> -Christopher
>>
>> On Dec 9, 2015, at 1:25 PM, Mike Allen <mkallen@gmail.com> wrote:
>>
>> Hi Jethro, current target patch for the FCX is 8.0.10h, though the latest
>> patch is 'm' (target path guidance usually lags the release schedule a
>> bit). That said, all of the development efforts are going into 8.0.30 and
>> above, the latest patch there is 8.0.30d. I have seen pretty good results
>> on this code for ICX, but not a lot of visibility for FCX running it.
>>
>> Mike
>>
>> On Wed, Dec 9, 2015 at 2:59 AM, Jethro R Binks <jethro.binks@strath.ac.uk
>> > wrote:
>>
>>> On Wed, 9 Dec 2015, Jethro R Binks wrote:
>>>
>>> > However, the reason I'm here really is to ask about the 8 code train,
>>> > which I have jumped to on F-CX.
>>>
>>> Apologies, this should read "which I have not jumped to yet on F-CX".
>>>
>>> Jethro.
>>>
>>> . . . . . . . . . . . . . . . . . . . . . . . . .
>>> Jethro R Binks, Network Manager,
>>> Information Services Directorate, University Of Strathclyde, Glasgow, UK
>>>
>>> The University of Strathclyde is a charitable body, registered in
>>> Scotland, number SC015263.
>>> _______________________________________________
>>> foundry-nsp mailing list
>>> foundry-nsp@puck.nether.net
>>> http://puck.nether.net/mailman/listinfo/foundry-nsp
>>>
>>
>> _______________________________________________
>> foundry-nsp mailing list
>> foundry-nsp@puck.nether.net
>> http://puck.nether.net/mailman/listinfo/foundry-nsp
>>
>>
>>
>> _______________________________________________
>> foundry-nsp mailing list
>> foundry-nsp@puck.nether.net
>> http://puck.nether.net/mailman/listinfo/foundry-nsp
>>
>
>
> _______________________________________________
> foundry-nsp mailing list
> foundry-nsp@puck.nether.net
> http://puck.nether.net/mailman/listinfo/foundry-nsp
>

--
Youssef BENGELLOUN-ZAHR