SGE2000 (horrible) problems.

[satanovskyl]

SGE2000 problems:

1) Layer 3 stack routing (?) problem after slave unit restarts.

Greetings!

Firmware: 3.0.0.18

Boot code: 20003

The scheme:

----- LAN1         SGE2000 Level 3 stack:

               `-----------Unit 2 (slave)

                           Unit 1 (master)

                           /

                           \

                            `----------(ws1)

The problem scenario:

            NB: It is only applicable to the situation when we use different subnets, in other words,

            the Layer 3 functionality of SGE2000 is involved. No such issues appear when all the hosts

            are in one VLAN, whether they are connected to different units in a stack or not.

            If the connection between Unit 1 and 2 is lost (I unplug the cable), slave unit ( Unit 2 ) resets,

            and comes up back to operation as a slave. It's OK.

            But the host ws1 (from LAN2) is not able to communicate to cmputers in LAN1 after the event.

            If I go to the web interface of the stack and do the ARP-table flush, some pings may pass

            but the next second the connectivity is lost again.

            All the ports involved in communication are in "forwarding" state.

            Sometimes, if I do a manual flush on Dynamic table in Bridging, communication

            becomes OK. Sometimes it doesn't help. Even if flushing arp cache too.

            If I unplug ws1 from the switch and plug it back, the connectivity becomes OK.

                 Flushing arp cache on ws1 doesn't help.

            If I restart the whole stack, the connectivity is OK.

            What's the problem with restarting a slave unit? Why the routing connectivity is lost?

--------------------------------------

2) Another problem is when trying to bind an ACL to the interface

     it simply reports:

Line No.	Error Type	Value	Diagnostic
1		Unknown value	Cannot apply because lack of HW resources..

     =) Great!

[alissitz]

Hello,

This sounds pretty odd, and thanks for explaining everything as you do.

Have you considered calling our support?

https://www.myciscocommunity.com/community/smallbizsupport

Look to the right hand side and you can find the link for the support numbers.  I do not see anything in the release notes as a known limitation or problem ... unless someone on this community has more to add, I would suggest calling our support.

Our support will likely request the configs and system logs (if you have any)

HTH,

Andrew Lissitz

[KelvinKN3]

I am also suffering from 'Problem 2' applying the ACL to a port.

It was OK - then suddenly has stopped working.  Have saved the config and bounced the device with no joy.

Any thoughts?

Regards,

[satanovskyl]

UNFORTUNATELY,

regarding to the official Cisco tech support (what comes out of my discussions with them),

all the discussed bugs (see also bugs regarding SRW* ) are

not considered critical, and there are no plans to fix anything in the nearest time.

...

On ACL problem: they say, take the textual config of the switch (exprot it to a txt file), edit it and upload it back, that's the way to edit ACLs... =)

Sounds promising )

[ripcisco]

Hi Leonid,

I have the same error re applying the ACLs at a customer site (Cannot apply because lack of HW resources) which is odd as I removed the existing ACL, editted it then tried to re-apply it.

What was the outcome of your dealings with Cisco support and did you manage to resolve this by using the 'textual config' method ?

If so do you have documentation of the 'textual config' method ?

Many thanks

Paul.

[w.surma]

Cannot apply because lack of HW resources

1. Check if every numer of your Rule Priority are unique.

2. I think that main rule (original "ACL_IP_Traffic") must be unchanged ( ANY, ANY, ANY, ANY permit) and must have rule priority number = 1.