RX packets drop at bonding interface

monu_1 — Tue, 09 Mar 2010 13:30:41 GMT

Hi All,

we are using bonding (mode=1) which creates bond2 interface enslaving eth4/eth5 interfaces.

We have seen everytime packets are dropping on bond2 for few hours and when I ran selftest offline, its rsolve the problem i.e pkt seen on this interface. The command to selftest on eth4 and eth5 as below:
root@payload3:~# ethtool -t eth4

The test result is PASS

The test extra info:

Register test (offline) 0

Eeprom test (offline) 0

Interrupt test (offline) 0

Loopback test (offline) 0

Link test (on/offline) 0

root@payload3:~# ethtool -t eth5

The test result is FAIL

The test extra info:

Register test (offline) 0

Eeprom test (offline) 0

Interrupt test (offline) 0

Loopback test (offline) 0

Link test (on/offline) 1

The tcpdump data on bond2 is captured at the time issue happened, but no packet is captured at this point of time.

root@payload1:~# ifconfig –a

bond2 Link encap:Ethernet HWaddr 00:80:42:1D:64:2C
inet6 addr: fe80::200:ff:fe00:0/64 Scope:Link
UP BROADCAST RUNNING PROMISC MASTER MULTICAST MTU:1500 Metric:1
RX packets:11381494 errors:0 dropped:3877327 overruns:0 frame:0
TX packets:19144 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:943464682 (899.7 MiB) TX bytes:1302504 (1.2 MiB)

eth4 Link encap:Ethernet HWaddr 00:80:42:1D:64:2C
inet6 addr: fe80::280:42ff:fe1d:642c/64 Scope:Link
UP BROADCAST RUNNING SLAVE MULTICAST MTU:1500 Metric:1
RX packets:11381494 errors:0 dropped:3877327 overruns:0 frame:0
TX packets:19144 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:943464682 (899.7 MiB) TX bytes:1302504 (1.2 MiB)
Base address:0x3000 Memory:d8400000-d8420000

eth5 Link encap:Ethernet HWaddr 00:80:42:1D:64:2C
UP BROADCAST SLAVE MULTICAST MTU:1500 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:0 (0.0 b) TX bytes:0 (0.0 b)
Base address:0x3040 Memory:d8420000-d8440000

root@payload1:~# cat /proc/net/bonding/bond2
Ethernet Channel Bonding Driver: v2.6.3 (June 8, 2005)

Bonding Mode: fault-tolerance (active-backup)
Primary Slave: None
Currently Active Slave: eth4
MII Status: up
MII Polling Interval (ms): 100
Up Delay (ms): 200
Down Delay (ms): 200

Slave Interface: eth4
MII Status: up
Link Failure Count: 4
Permanent HW addr: 00:80:42:1d:64:2c

Slave Interface: eth5
MII Status: down
Link Failure Count: 0
Permanent HW addr: 00:80:42:1d:64:2d

root@payload1:~# netstat -i
Kernel Interface table
Iface MTU Met RX-OK RX-ERR RX-DRP RX-OVR TX-OK TX-ERR TX-DRP TX-OVR Flg
bond0 1500 0 22431248 0 0 024034375 0 0 0 BMmRU
bond1 1500 0 20027975 0 0 015809489 0 0 0 BMmRU
bond2 1500 0 11381494 0 3877632 0 19144 0 0 0 BMPmRU
eth0 1500 0 22435724 0 0 024034360 0 0 0 BMRU
eth1 1500 0 20027969 0 0 015809488 0 0 0 BMRU
eth3 1500 0 0 0 0 0 0 0 0 0 BMU
eth4 1500 0 11381494 0 3877632 0 19144 0 0 0 BMsRU
eth5 1500 0 0 0 0 0 0 0 0 0 BMsU
eth0. 1500 0 18878772 0 0 020481895 0 0 0 BMsRU
eth0. 1500 0 3552476 0 0 0 3552480 0 0 0 BMsRU
eth1. 1500 0 16473477 0 0 012257008 0 0 0 BMsRU
eth1. 1500 0 3554498 0 0 0 3552481 0 0 0 BMsRU
lo 16436 0 11069821 0 0 011069821 0 0 0 LRU

We would like to know what is wrong with the interface as why it is not getting any packets and how the running of this command resolved the issue for us everytime.

Thanks,
MKS

Re: RX packets drop at bonding interface

Steven E. Protter — Tue, 09 Mar 2010 13:53:32 GMT

Shalom MKS,

Packets are going to drop once in a while whether the interface is bonded or not.

If it is significant, or performance is impacted, then look into the issue.

Same as with anything else:
1) Cable could be bad.
2) NIC port could be bad.
3) Switch may not handle the bonding well.
4) Switch might need a firmware or configuration update.
5) You may wish to try a different bonding mode in modprobe.conf

SEP

Re: RX packets drop at bonding interface

monu_1 — Wed, 10 Mar 2010 06:16:43 GMT

Thanks SEP.

Why packets seen when we ran ethtool selftest in offline mode on eth4 during problem time? As per my understanding this will reset the interface at this particular point of time and porblem is going away. Isn't this ruled out any hardware problem in this interface?

Also, will this issue be cause if we use other than mode 1 bonding on far end?

Your comments is highly appriciate.

Thanks in advance.

MKS.

topic Re: RX packets drop at bonding interface in Operating System - Linux

RX packets drop at bonding interface

Re: RX packets drop at bonding interface

Re: RX packets drop at bonding interface