Operating System - Tru64 Unix
1827892 Members
1800 Online
109969 Solutions
New Discussion

Link down problem when we restart the system !!

 
Fenny_1
Super Advisor

Link down problem when we restart the system !!

Hi All,
I am facing the problem in DS25 alpha server. The two NIC cards are shown at the boot prompt one is working fine while the others link is down. When we have restated the machine then the one NIC card's link goes down while the other is up. When i decongigure and configure it again then it works ok. So how can i resolve this issue. Waiting for ur reply ...
8 REPLIES 8
Vladimir Fabecic
Honored Contributor

Re: Link down problem when we restart the system !!

Disable autonegotiation on switch port and set fixed speed in console for that interface.
In vino veritas, in VMS cluster
Al Licause
Trusted Contributor

Re: Link down problem when we restart the system !!

Why is it that everyone always seems to blame autonegotiation without ever asking more detailed questions ? Particularly in cases like this where we have so little nformation.

Let's start with the version of Tru64unix along with the patch kit level and the type of network interface you are using...i.e. tu, ee, alt, bcm ?

How is that particular interface configured if other than with default settings ?

When you say, the link is down...how do you konw this ?

Does it have an ip address ? What is the state of the interface ?

Does it pass any traffic at all ?

What does hwmgr think it's doing ?

Lot's of information you need to supply before people start guessing at what the solution should be.

Ivan Ferreira
Honored Contributor

Re: Link down problem when we restart the system !!

Some devices must be configured to disable the autonegotiation at driver level also. Use the lan_config command and add it to the /etc/inet.local file, for example:

/usr/sbin/lan_config -i ee0 -a 0 -s 100 -x 1
Por que hacerlo dificil si es posible hacerlo facil? - Why do it the hard way, when you can do it the easy way?
Vladimir Fabecic
Honored Contributor

Re: Link down problem when we restart the system !!

Why is it that everyone always seems to blame autonegotiation without ever asking more detailed questions ?
Because this is the first step to be done. If it does not help, then I would ask for details.
By the way, which interface is causing the problem, fast ethernet or gigabit?
In vino veritas, in VMS cluster
Ahmed Saleem
Advisor

Re: Link down problem when we restart the system !!

The output taken from the server are as follows
# hwmgr -get att -cat network
33:
name = ee0
category = network
sub_category = Ethernet
model = Intel 8255x
hardware_rev = 9
firmware_rev =
MAC_address = 00-02-A5-20-1D-95
MTU_size = 1500
media_speed = 100
media_selection = Automatic
media_type = Unshielded Twisted Pair (UTP)
loopback_mode = 0
promiscuous_mode = 0
full_duplex = 1
multicast_address_list = CF-00-00-00-00-00 01-00-5E-00-00-01
interface_number = 1
link = Up
autoneg_enable = 1
registration_time = Tue Jun 6 15:52:25 2006
user_name = (null) (settable)
location = (null) (settable)
software_module = (null)
state = available
state_previous = unavailable
state_change_time = Tue Jun 6 16:52:35 2006
event_count = 4
last_event_time = Tue Jun 6 16:52:35 2006
access_state = online
access_state_change_time = none
capabilities = 0
indicted = 0
indicted_probability = (null)
indicted_urgency = (null)
disabled = 0
cabinet_type = (null)
cabinet_id = (null)
position = (null)
chassis_type = (null)
chassis_id = (null)
assembly = (null)
subassembly = (null)
est_seconds = 0
est_bytesent = 88336393
est_bloksent = 1518120
est_mbytesent = 8358
est_mbloksent = 63
est_deferred = 0
est_single = 0
est_multiple = 0
est_collis = 0
est_unrecog = 0
est_userbuf = 0
est_latecoll = 0
est_excesscoll = 0
est_carrierfail = 0
est_shortcirc = 0
est_opencirc = 0
est_sndlong = 0
est_sendfail = 0
est_bytercvd = 10803353424
est_blokrcvd = 7148661
est_mbytercvd = 60
est_mblokrcvd = 1
est_overrun = 0
est_sysbuf = 0
est_unaligned = 0
est_longframe = 0
est_shortframe = 0
est_fcsfail = 0
est_badframe = 0
est_symbolerror = 0
est_recvfail = 0
51:
name = bcm0
category = network
sub_category = Ethernet
model = DEGXA
hardware_rev = BCM5703_A2
firmware_rev =
MAC_address = 00-02-A5-C6-1A-28
MTU_size = 1500
media_speed = 100
media_selection = Automatic
media_type = Unshielded Twisted Pair (UTP)
loopback_mode = 0
promiscuous_mode = 0
full_duplex = 1
multicast_address_list = CF-00-00-00-00-00 01-00-5E-00-00-01
interface_number = 2
link = Up
autoneg_enable = 1
registration_time = Tue Jun 6 15:52:43 2006
user_name = (null) (settable)
location = (null) (settable)
software_module = (null)
state = available
state_previous = unavailable
state_change_time = Tue Jun 6 17:38:12 2006
event_count = 11
last_event_time = Tue Jun 6 17:38:12 2006
access_state = online
access_state_change_time = none
capabilities = 0
indicted = 0
indicted_probability = (null)
indicted_urgency = (null)
disabled = 0
cabinet_type = (null)
cabinet_id = (null)
position = (null)
chassis_type = (null)
chassis_id = (null)
assembly = (null)
subassembly = (null)
est_seconds = 0
est_bytesent = 146129509
est_bloksent = 192680
est_mbytesent = 0
est_mbloksent = 29
est_deferred = 0
est_single = 0
est_multiple = 0
est_collis = 0
est_unrecog = 0
est_userbuf = 0
est_latecoll = 0
est_excesscoll = 0
est_carrierfail = 0
est_shortcirc = 0
est_opencirc = 0
est_sndlong = 0
est_sendfail = 0
est_bytercvd = 275121083
est_blokrcvd = 612157
est_mbytercvd = 0
est_mblokrcvd = 478227
est_overrun = 0
est_sysbuf = 0
est_unaligned = 0
est_longframe = 0
est_shortframe = 0
est_fcsfail = 0
est_badframe = 0
est_symbolerror = 0
est_recvfail = 0
the bcm0 is creating the problem .... as we have manually configured the NIC but when we reboot the system it goes down ...
Al Licause
Trusted Contributor

Re: Link down problem when we restart the system !!

According to the hwmgr output you listed for bcm0, the link is up, so again I would ask how is it you have determined that the link is down ?

Are you viewing console messages or information in a log file ?
If so can you post any of those messages ?

When you say you have manually configured the network card can you please post the commands
you have used to do so and in which files you have placed them ?

According to hwmgr the card is running at 100Mbps fdx and has done so through autonegotiation. Which means it either negotiated to that setting or failed to negotiate with the switch port and used a default setting. More than likely the switch port is not set to autonegotiate.

What speed and duplex mode are you setting for this card ?

Keep in mind that the only way you'll achieve 1000Mbps fdx operation if that is what you want to do will be through autonegotiation. Any attempt to manually set this card to 1000Mbps is ignored by the driver.

I would also double check your switch port settings. Keep in mind that the settings for the switch and network card must match...i.e. they must both be set for autonegotiation or to the same speed and duplex mode if set manually.

"Why is it that everyone always seems to blame autonegotiation without ever asking more detailed questions ? Because this is the first step to be done. If it does not help, then I would ask for details. By the way, which interface is causing the problem, fast ethernet or gigabit?"

I would normally leave a stupid reply like this alone, but I simply have to respond in the negative. To assume a solution without first asking questions makes no sense. I don't know how many times customers have contacted the support center after spendings hours and in some cases days of trying things without really asking the correct questions. When they finally do, the first thing we do is to ask questions to fully understand the problem. Only then can we offer any kind of suggestion or solution. That is how you trouble shoot.

To state that some interfaces require that autonegotiation be disabled is again not to be taken as a general statement as any device will work if it is capable of autonegotiating and can correctly negotiate with the paired switch port. The only time autonegotiation should be considered should be after it has tried and been shown to fail consitantly in a particular configuration. The equipment was designed to operate at it's optimal setting through negotiation. Let it do what it was meant to do.


Rob Leadbeater
Honored Contributor

Re: Link down problem when we restart the system !!

Vladimir said: "Disable autonegotiation on switch port and set fixed speed in console for that interface."

For older machines with Tulip network cards I'd agree with this, particularly when you're talking to Cisco switches, however we had exactly the opposite situation when connecting up our DS25s to the network.

If the DS25s and switch were set to NOT auto-negotiate, then the link wouldn't come up properly. Set everything back to auto-negotiate and it works fine.

As Al says: "The equipment was designed to operate at it's optimal setting through negotiation. Let it do what it was meant to do."

Cheers,

Rob
Ahmed Saleem
Advisor

Re: Link down problem when we restart the system !!

Hi All,
The server is not down, as after every restart/reboot only the on-board NIC
status is down. We reconfigure it and it works fine till next reboot/restart
or shutdown.

Currently the NIC is working fine