Re: IPMP enabled serious Network Problems

On Jun 20, 4:00 pm, James Carlson <james.d.carl...@xxxxxxx> wrote:
chbueh...@xxxxxxxxx writes:
This means several Servers in the same subnet are no longer available.
If we disable the IPMP, by shutting down one port of one Cisco Switch
we have no problems at all.

Any log messages? Any information in "ifconfig -a"? What exactly is
root # ifconfig -a
lo0: flags=2001000849<UP,LOOPBACK,RUNNING,MULTICAST,IPv4,VIRTUAL> mtu
8232 index
1
inet 127.0.0.1 netmask ff000000
bge0:
flags=9040843<UP,BROADCAST,RUNNING,MULTICAST,DEPRECATED,IPv4,NOFAILOVER>
m
tu 1500 index 2
inet 10.xx.xx.49 netmask ffffff00 broadcast xxxxxxx.255
groupname ipmp0
ether 0:3:ba:6:d9:99
bge0:1: flags=1000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500
index 2
inet 10.xx.xx.50 netmask ffffff00 broadcast xxxxxxx.255
bge1:
flags=69040843<UP,BROADCAST,RUNNING,MULTICAST,DEPRECATED,IPv4,NOFAILOVER,S
TANDBY,INACTIVE> mtu 1500 index 3
inet 10.xx.xx.51 netmask ffffff00 broadcast xxxxxxx.255
groupname ipmp0
ether 0:3:ba:6:d9:9a
root #
your configuration -- /etc/hostname*
root # ls host*
hostname.bge0 hostname.bge1 hosts hosts.ori
root #

and /etc/default/mpathd? What

root # more mpathd
#
#pragma ident "@(#)mpathd.dfl 1.2 00/07/17 SMI"
#
# Time taken by mpathd to detect a NIC failure in ms. The minimum time
# that can be specified is 100 ms.
#
FAILURE_DETECTION_TIME=10000
#
# Failback is enabled by default. To disable failback turn off this
option
#
FAILBACK=yes
#
# By default only interfaces configured as part of multipathing groups
# are tracked. Turn off this option to track all network interfaces
# on the system
#
TRACK_INTERFACES_ONLY_WITH_GROUPS=yes
root #

drivers do you have? What Solaris version?

We are using Solaris 10

What sort of Cisco
We have 2 Locations connect via gigabit fibre on both Locations
there are 2 4506 Catalysts on each Location. OSPF between.


switch? How is your network configured? Do you use VRRP or HSRP?

We are using HSRP

Do you have any unusual (or ill-advised) driver configuration, such as
forced duplex settings?
We treid all possible configurations

What debugging have you done? What have you tried? Are there any
particular events or changes that seem related to the problem? Are
the switches themselves reporting any sort of errors?

We are seeing a lot of Output Drops on many Interfaces on the
Catalyst, as soons
as we start with the backups:

GigabitEthernet3/5 is up, line protocol is up (connected)
Hardware is Gigabit Ethernet Port, address is 0012.7fb6.df84 (bia
0012.7fb6.df84)
MTU 1500 bytes, BW 1000000 Kbit, DLY 10 usec,
reliability 255/255, txload 1/255, rxload 1/255
Encapsulation ARPA, loopback not set
Keepalive set (10 sec)
Full-duplex, 100Mb/s, link type is auto, media type is 10/100/1000-
TX
input flow-control is off, output flow-control is off
ARP type: ARPA, ARP Timeout 04:00:00
Last input 00:00:42, output never, output hang never
Last clearing of "show interface" counters 3w0d
Input queue: 0/2000/0/0 (size/max/drops/flushes); Total output
drops: 3919485296
Queueing strategy: fifo
Output queue: 0/40 (size/max)
5 minute input rate 1000 bits/sec, 1 packets/sec
5 minute output rate 8000 bits/sec, 10 packets/sec
14204975 packets input, 2769721882 bytes, 0 no buffer
Received 325431 broadcasts (325280 multicasts)
0 runts, 0 giants, 0 throttles
0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored
0 input packets with dribble condition detected
2123430443 packets output, 2854855264867 bytes, 0 underruns
0 output errors, 0 collisions, 0 interface resets
0 babbles, 0 late collision, 0 deferred
0 lost carrier, 0 no carrier
0 output buffer failures, 0 output buffers swapped out





If someone faced the same problem and was able to solve it, a hint
would be very appreciated.

I don't know if I (personally) can help, but I doubt that anyone can
help much until you provide some details. Simply saying that the
servers become unreachable doesn't provide much to go on.

It seems like the network gets flooded with something, but even in
sniffer-traces
we were not able to see something special or unexpected.



--
James Carlson, Solaris Networking <james.d.carl...@xxxxxxx>
Sun Microsystems / 1 Network Drive 71.232W Vox +1 781 442 2084
MS UBUR02-212 / Burlington MA 01803-2757 42.496N Fax +1 781 442 1677


.

Relevant Pages

  • IPMP with ipfilter problems - still not solved - ipfilter not responsible for problems
    ... test IP addresses in their own IPMP group, ... reconfiguring the interfaces, I had to wait only a few minutes before the ... inet 202.51.161.138 netmask ffffff00 broadcast 202.51.161.255 ...
    (SunManagers)
  • Re: [PATCH 4/5] ieee802154: add documentation about our stack
    ... +Currently only IEEE 802.15.4 layer is implemented. ... +Most of IEEE 802.15.4 MLME interfaces are directly mapped on netlink commands. ... +2) 'SoftMAC' or just radio. ... are you sending IP packets over this ARPHRD_IEEE802154 network devices ...
    (Linux-Kernel)
  • Re: ISR CBAC prolem
    ... When I apply CBAC onto inside interface without any ACL's ... Have you determined if packets are arriving out of order? ... I'm running CEF and netflow on both outside and inside interfaces... ... > increase the inspection timouts. ...
    (comp.dcom.sys.cisco)
  • Re: Help Broadcasting a UDP packet on the LAN:URGENT
    ... We use all-ones packets well ... > network interfacethey should be using to do this. ... > interfaces because you have a per-network broadcast address if you want ... That way you get "for free" to control which interfaces should send ...
    (freebsd-net)
  • Bridging interfaces
    ... I seem to be having some trouble bridging interfaces in FreeBSD 6.2-STABLE. ... packets transmitted, 2 packets received, 0% packet loss ... inet 192.168.1.2 netmask 0xffffff00 broadcast 192.168.1.255 ...
    (freebsd-questions)