Login | Register For Free | Help
Search for: (Advanced)

Mailing List Archive: nsp: juniper

Juniper log settings

 

 

nsp juniper RSS feed   Index | Next | Previous | View Threaded


kanagaraj at aims

Aug 28, 2006, 2:31 AM

Post #1 of 5 (737 views)
Permalink
Juniper log settings

Hi,
We are using M7i's in our network setup for a iBGP connection and OSPF as the internal routing protocol. Whenever the iBGP link flaps or goes down, the log only records the down status (refer to the log below) and no indication on the up status. These makes it difficult to estimate the period of downtime refering to the log. I've attached the config and the log output for reference. Are there any other changes that I need to make? Thanks.

Config
-------
syslog {
user * {
any emergency;
}
file messages {
any notice;
authorization info;
}
}


Log
-----
SNMP_TRAP_LINK_DOWN: ifIndex 68, ifAdminStatus up(1), ifOperStatus down(2), ifName t3-0/1/1
bgp_ifachange_group: NOTIFICATION sent to xxx.xxx.xxx.xxx (Internal AS xxxxx): code 6 (Cease) subcode 6 (Other Configuration Change), Reason: Interface change for the peer-group
RPD_OSPF_NBRDOWN: OSPF neighbor xxx.xxx.xxx.xxx (t3-0/1/1.0) state changed from Full to Down due to Kill all neighbors


jeekay at gmail

Aug 28, 2006, 3:09 AM

Post #2 of 5 (718 views)
Permalink
Juniper log settings [In reply to]

On 28/08/06, Kanagaraj Krishna <kanagaraj at aims.com.my> wrote:
> file messages {
> any notice;
> authorization info;
> }

If memory serves, JunOS logs BGP neighbour down messages at NOTICE,
but logs BGP neighbour up messages at INFO. For the same reason, our
management software only sees things going down via syslog but has to
SNMP for things going up :)

If you change that to 'any info' you should be good to go.


erdems at gmail

Aug 28, 2006, 4:43 AM

Post #3 of 5 (697 views)
Permalink
Juniper log settings [In reply to]

Hi,

If you don't want 'too much information' within a single logfile, you
also have options :

1) alter your config with different filenames with different types of
information such as:

file messages {
any notice;
archive size 10m files 30;
}
file infos {
any info;
archive size 10m files 100;
}
file major {
any error;
archive size 5m files 100;
}
file firewall {
firewall any;
archive size 5m files 100;
}
file user-log {
interactive-commands any;
archive size 5m files 30;
}
file Kernel {
kernel any;
archive size 10m files 30;

2) you could use 'traceoptions' under [edit protocols bgp] and/or
[edit protocols bgp group XXX] and enable flag 'state', which will log
up/down status in the traceoptions file you'll specify:

group iBGP {
type internal;
traceoptions {
file ibgp.log size 10m files 20;
flag state;
}

HTH


On 8/28/06, Jee Kay <jeekay at gmail.com> wrote:
> On 28/08/06, Kanagaraj Krishna <kanagaraj at aims.com.my> wrote:
> > file messages {
> > any notice;
> > authorization info;
> > }
>
> If memory serves, JunOS logs BGP neighbour down messages at NOTICE,
> but logs BGP neighbour up messages at INFO. For the same reason, our
> management software only sees things going down via syslog but has to
> SNMP for things going up :)
>
> If you change that to 'any info' you should be good to go.
> _______________________________________________
> juniper-nsp mailing list juniper-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/juniper-nsp
>


--
Erdem


gbachler at juniper

Aug 28, 2006, 7:26 AM

Post #4 of 5 (696 views)
Permalink
Juniper log settings [In reply to]

Hi,

If you configure log-updown option under any of the bgp hierarchy levels,
BGP state transitions will be logged at warning level.

Configuring BGP to Log System Log Messages:
http://www.juniper.net/techpubs/software/junos/junos80/swconfig80-routing/html/bgp-config62.html#1016413


georg at ankara-re0> show log messages |last |match bgp
Aug 28 16:43:09 ankara-re0 rpd[3147]: %DAEMON-4: bgp_traffic_timeout: NOTIFICATION sent to 192.168.1.8 (Internal AS 64512): code 4 (Hold Timer Expired Error), Reason: holdtime expired for 192.168.1.8 (Internal AS 64512), socket buffer sndcc: 38 rcvcc: 0 TCP state: 4, snd_una: 2725219125 snd_nxt: 2725219125 snd_wnd: 16384 rcv_nxt: 2243555251 rcv_adv: 2243571649, keepalive timer 0
Aug 28 16:43:09 ankara-re0 rpd[3147]: %DAEMON-4: bgp_event: peer 192.168.1.8 (Internal AS 64512) old state Established event HoldTime new state Idle
< ... >
Aug 28 16:44:13 ankara-re0 rpd[3147]: %DAEMON-4: bgp_event: peer 192.168.1.8 (Internal AS 64512) old state OpenConfirm event RecvKeepAlive new state Established

--
BR, Georg


kanagaraj at aims

Aug 28, 2006, 11:10 PM

Post #5 of 5 (696 views)
Permalink
Juniper log settings [In reply to]

Hi,
Thanks for the prompt response. How about the physical interface
itself? At the moment the log only records physical interface down status
only and not when its up.

Regards,
Kana

----- Original Message -----
From: "Jee Kay" <jeekay [at] gmail>
To: "Kanagaraj Krishna" <kanagaraj at aims.com.my>
Cc: <juniper-nsp at puck.nether.net>
Sent: Monday, August 28, 2006 7:09 PM
Subject: Re: [j-nsp] Juniper log settings


> On 28/08/06, Kanagaraj Krishna <kanagaraj at aims.com.my> wrote:
> > file messages {
> > any notice;
> > authorization info;
> > }
>
> If memory serves, JunOS logs BGP neighbour down messages at NOTICE,
> but logs BGP neighbour up messages at INFO. For the same reason, our
> management software only sees things going down via syslog but has to
> SNMP for things going up :)
>
> If you change that to 'any info' you should be good to go.
>

nsp juniper RSS feed   Index | Next | Previous | View Threaded
 
 


Interested in having your list archived? Contact Gossamer Threads
 
  Web Applications & Managed Hosting Powered by Gossamer Threads Inc.