Ping delays between HP & Dell Servers

[MikeL_4]

We have a group of 20 Dell and 10 HP servers, all running Red Hat 5.5 x84_64 bit OS on them, where the client is reporting issues with there application aborting at times due to network loss..
The client did notice is that ping's showed delays consistently in the time of the application issue between Dell and HP boxes but not between Dell and Dell boxes or between HP and HP boxes. The majority of the delays appear only when they go between two different machines types...

All of the servers are running 1Gb NIC interfaces and are all on the same subnet by themselves...

Can anyone point me into a direction to try and resolve this problem, the /var/log/messages shows NO network errors at all, and the only thing the client can say is that the response between servers goes up when they experience there problem...

Thanks

[root@DellServer ~]# ethtool -i eth0
driver: bnx2
version: 1.9.3
firmware-version: 5.0.4 UMP 1.1.9
bus-info: 0000:03:00.0

[root@DellServer ~]# ethtool -i eth3
driver: e1000e
version: 1.0.2-k2
firmware-version: 5.6-2
bus-info: 0000:08:00.1

[root@DellServer ~]# ethtool eth0
Settings for eth0:
Supported ports: [ TP ]
Supported link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Full
Supports auto-negotiation: Yes
Advertised link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Full
Advertised auto-negotiation: Yes
Speed: 1000Mb/s
Duplex: Full
Port: Twisted Pair
PHYAD: 1
Transceiver: internal
Auto-negotiation: on
Supports Wake-on: g
Wake-on: d
Link detected: yes

[root@DellServer ~]# ethtool eth3
Settings for eth3:
Supported ports: [ TP ]
Supported link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Full
Supports auto-negotiation: Yes
Advertised link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Full
Advertised auto-negotiation: Yes
Speed: 1000Mb/s
Duplex: Full
Port: Twisted Pair
PHYAD: 1
Transceiver: internal
Auto-negotiation: on
Supports Wake-on: d
Wake-on: d
Current message level: 0x00000001 (1)
Link detected: yes

[root@HPServer ~]# ethtool -i eth0
driver: tg3
version: 3.96-1
firmware-version: 5704-v3.27b, ASFIPMIc v2.36
bus-info: 0000:03:01.0

[root@HPServer ~]# ethtool -i eth1
driver: tg3
version: 3.96-1
firmware-version: 5704-v3.27b
bus-info: 0000:03:01.1

[root@HPServer ~]# ethtool eth0
Settings for eth0:
Supported ports: [ TP ]
Supported link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Half 1000baseT/Full
Supports auto-negotiation: Yes
Advertised link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Half 1000baseT/Full
Advertised auto-negotiation: Yes
Speed: 1000Mb/s
Duplex: Full
Port: Twisted Pair
PHYAD: 1
Transceiver: internal
Auto-negotiation: on
Supports Wake-on: g
Wake-on: g
Current message level: 0x000000ff (255)
Link detected: yes

[root@HPServer ~]# ethtool eth1
Settings for eth1:
Supported ports: [ TP ]
Supported link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Half 1000baseT/Full
Supports auto-negotiation: Yes
Advertised link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Half 1000baseT/Full
Advertised auto-negotiation: Yes
Speed: 1000Mb/s
Duplex: Full
Port: Twisted Pair
PHYAD: 1
Transceiver: internal
Auto-negotiation: on
Supports Wake-on: g
Wake-on: g
Current message level: 0x000000ff (255)
Link detected: yes

[Vitaly Karasik_1 · Accepted solution]

I suggest:
- check "ifconfig" and "netstat -s" error statistics
- check your switch errors
- test connectivity between *all* servers - your first impression about "Dell vs. HP" may be right, but may be not

[nightwich]

Hi Mikel


First step is to do a netstat -in and see if you have network error's


Best regards.

[MikeL_4]

This is what is so much of a puzzle, I've checked the statistics as mentioned and I see no errors being recorded on the LAN's...

and, the LAN's are not going down at all, it's just that at times the response times are really high causing the applications the cliets are running to fail...

[Steven E. Protter]

Shalom,

1) Look beyond the servers. Have the network infrastructure checked, not for errors but for current versions of router OS software.

2) Go to Dell and HP and get the latest drivers for RHEL for these NIC cards.

3) If Dell and HP are not helpful, go to the card manufacturers site.

4) Replace network cabling.

5) Do a test between two servers without network infrastructure. Run a cable between NIC ports of a Dell and an HP server, run a private network and see if there are still delays. Transfer some large files and see how fast they move. They should move at at least 20 Mb/s.

6) Consider using network bonding, or changing the bonding mode if you are already using it.

I'm not dealing directly with the issue. What I'm doing is providing you some tests to eliminate possible problem areas and focus your investigation.

SEP