05-07-2009 12:28 AM - edited 03-06-2019 05:35 AM
Hi,
Strange one here, I have 2 x 3750G switches which connect to our Nortel core switches. The trunk is setup as an Etherchannel, I have 2 ports in each etherchannel goign to the Nortel switches.
(I can add a diagram if required)
On these 3750's is our Exchange 2007 email system and I use Solarwinds Orion tools to monitor these Exchaneg servers and email me if anything goings down.
I am using 3 VLAN's on the 3750's and Solarwinds happy ping sthe servers and I get no issues at all. The problem I am getting is pinging the VLAN's on these 3750's (basically their default gateways), throughout the day I will get email alerts saying VLAN 10 is down then up, then later VLAN 20 is down then up, this can take only 30 seconds, but they never go down at the same time and it might only be one 3750 switch for that day, the other will be fine.
I'm just not sure where to start as the Exchaneg system is fine, but it just concerns me.
Is it the Nortel's or the trunk?
Thanks
05-07-2009 12:57 AM
Hi ,
If logging is enabled on your switches you can the logs using show log on 3750 switches.
Are you seeing any interface flops in logs..
Thanks
satish
05-07-2009 01:05 AM
On both all I have is this, do I ned to change the logging level if so how?
Syslog logging: enabled (0 messages dropped, 1 messages rate-limited, 0 flushes, 0 overruns, xml disabled, filtering disabled)
Console logging: level debugging, 2704 messages logged, xml disabled,
filtering disabled
Monitor logging: level debugging, 1 messages logged, xml disabled,
filtering disabled
Buffer logging: level notifications, 47 messages logged, xml disabled,
filtering disabled
Exception Logging: size (4096 bytes)
Count and timestamp logging messages: disabled
File logging: disabled
Trap logging: level notifications, 2707 message lines logged
Logging to 192.168.21.19, 1400 message lines logged, xml disabled,
filtering disabled
Log Buffer (8192 bytes):
14w6d: %SYS-5-CONFIG_I: Configured from console by vty0 (192.168.7.12)
15w0d: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/10, changed state to down
15w0d: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/19, changed state to down
15w0d: %LINK-3-UPDOWN: Interface GigabitEthernet1/0/10, changed state to down
15w0d: %LINK-3-UPDOWN: Interface GigabitEthernet1/0/19, changed state to down
15w0d: %LINK-3-UPDOWN: Interface GigabitEthernet1/0/10, changed state to up
15w0d: %LINK-3-UPDOWN: Interface GigabitEthernet1/0/19, changed state to up
15w0d: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/10, changed state to up
15w0d: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/19, changed state to up
15w0d: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/10, changed state to down
15w0d: %LINK-3-UPDOWN: Interface GigabitEthernet1/0/10, changed state to down
15w0d: %LINK-3-UPDOWN: Interface GigabitEthernet1/0/10, changed state to up
15w0d: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/10, changed state to up
15w0d: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/19, changed state to down
15w0d: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/10, changed state to down
3750-B#
05-07-2009 01:10 AM
We can see the interface flap is happening...
Can paste the config...
Thanks,
Satish
05-07-2009 01:12 AM
These are 15w0d old though, not recent and are just cables put into port 10 and 19.
05-07-2009 01:15 AM
Hi ,
Clear the logs and check the logs at the end of the day..after getting mail alerts...
Thanks,
Satish
05-07-2009 01:31 AM
Is my logging levels ok to capture this?
sh run | include logging
logging buffered 8192 notifications
logging enable
logging size 200
logging trap notifications
logging facility local4
logging source-interface Vlan1028
logging 192.168.2.9
05-07-2009 06:22 AM
I would make your logging buffer size a lot bigger . the default does not give you many messages before it overwrites itself , I would make it like 100K .
05-07-2009 11:35 AM
Thanks, all done.
Should I be seeing anything using "terminal monitor" via telnet if my config is:
logging buffered 100000 notifications
logging trap notifications
or
logging buffered 100000 debug
logging trap debug
I added these lines, but see nothing
05-07-2009 10:13 AM
Hi. No, they happened 15w0d after last reload of the switch...
Oen reason why the Cisco default of "srvice timestamp log uptime" is annoying.
Always as a best practice user "service timestamp debug/log datetime msec" and if possible NTP in order to have useful time stamping of what goes on in your network.
05-07-2009 11:01 AM
Thanks, I have done this, do I need to change these settings to debug?
logging buffered 100000 notifications
logging trap notifications
05-07-2009 06:43 AM
Hi:
Its hard to say what is wrong. I doubt there is any issue with the Nortel switch or the trunk.
For starters, why dont you configure more granular logging.
service timestamps debug datetime localtime
service timestamps log datetime localtime
Clear the log and lets see what kind of error messages pop up.
HTH
Victor
05-07-2009 07:02 AM
Hi,
I have changed a few settings, this what I have, do I need to change my trap and buffer to debug?
sh run | inc logg
logging buffered 100000 notifications
logging enable
logging size 200
logging trap notifications
logging facility local4
logging source-interface Vlan1028
logging 192.168.2.9
05-07-2009 01:53 PM
If you want to see the debug in the logs then yes change it to logging buffered
05-11-2009 12:35 PM
Hi,
Well after a few days the vlans are still going up and down, but the "sh log" only shows my logins, does this mean the switches are fine? Maybe it is the opposite end of the trunk that is the issue, the Cisco's have a trunk to a Nortel core switch.
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide