Interface drops on VSF links

[Pk21]

Hi,

We have 2 HP J9850A Switch 5406Rzl2 switches which are linked via VSF. The VSF consist of 2 x 10Gbps links.

On 1 of the 2 links there are lots of Tx drops so I want to move this link to other switch ports to see if that solves the issues or disable the link completely.

Is this possible without downtime?

Our VSF configuration:

vsf
enable domain 911
member 1
type "J9850A" mac-address 548028-xxxxxx
priority 64
link 1 1/B8,1/C8
link 1 name "I-Link1_1"
exit
member 2
type "J9850A" mac-address 548028-xxxxxx
priority 128
link 1 2/B8,2/C8
link 1 name "I-Link2_1"
exit
oobm-mad
port-speed 10g

#

sh vsf

VSF Domain ID : 911
MAC Address : 548028-xxxxxx
VSF Topology : Chain
VSF Status : Active
Uptime : 153d 4h 44m
VSF MAD : OOBM
VSF Port Speed : 10G
Software Version : KB.16.09.0003

Mbr
ID MAC Address Model Pri Status
--- ----------------- ------------------------------------- --- ---------------
1 548028-xxxxxx HP J9850A Switch 5406Rzl2 64 Standby
2 548028-xxxxxx HP J9850A Switch 5406Rzl2 128 Commander

Thanks in advance,

Peter

[parnassus]

---

@Pk21 wrote: Is this possible without downtime?

---

Hello Peter! Yes, it is.

I've done that when a rearranging and addition of VSF Links was needed on our 2x5406R zl2 VSF...I recabled moving from a 2x10 to a 4x10 (with rearrange), clearly at least one VSF physical link was always seated on its original position during the procedure (it's the very minimum requirement to not split the VSF).

Read about it on the thread VSF links: can some links be rearranged without issues? discussed on Aruba Airheads. Tested an it worked flawlessly.

[Pk21]

Thanks Parnassus.

I just disconnected the link by physically removing the cable after business hours. But by disconnecting the 1 VSF link with drops the problem was not fixed, when I checked the next day there were drops on the other active link.  I will try to reach out to the HPE support team.

Status and Counters - Port Counters for VSF port 1/C8
Utilization Rx : 03.13 % Utilization Tx : 00.81 %
Name :
MAC Address : 548028-80f6b8
Link Status : Up
Port Enabled : Yes
Totals (Since boot or last clear) :
Bytes Rx : 2,119,428,380 Bytes Tx : 699,379,294
Packets Rx : 223714 Packets Tx : 1,600,815,929
Errors (Since boot or last clear) :
FCS Rx : 0 Drops Tx : 81,617,902
Alignment Rx : 0 Collisions Tx : 0
Runts Rx : 0 Late Colln Tx : 0
Giants Rx : 0 Excessive Colln : 0
Total Rx Errors : 0 Deferred Tx : 0

Others (Since boot or last clear) :
Discard Rx : 0 Out Queue Len : 0
Unknown Protos : 0
Rates (5 minute weighted average) :
Total Rx(Kbps) : 81,224 Total Tx(Kbps) : 312,944
Packets Rx (Pkts/sec) : 116,082 Packets Tx (Pkts/sec) : 125,276
Utilization Rx : 00.81 % Utilization Tx : 03.12 %

Status and Counters - Port Counters for VSF port 2/C8

Name :
MAC Address : 548028-7310b8
Link Status : Up
Port Enabled : Yes
Totals (Since boot or last clear) :
Bytes Rx : 2,213,545,195 Bytes Tx : 924,309,904
Packets Rx : 223714 Packets Tx : 1,156,041,646
Errors (Since boot or last clear) :
FCS Rx : 0 Drops Tx : 10,684,929
Alignment Rx : 0 Collisions Tx : 0
Runts Rx : 0 Late Colln Tx : 0
Giants Rx : 0 Excessive Colln : 0
Total Rx Errors : 0 Deferred Tx : 0

Others (Since boot or last clear) :
Discard Rx : 189,418 Out Queue Len : 0
Unknown Protos : 0
Rates (5 minute weighted average) :
Total Rx(Kbps) : 321,384 Total Tx(Kbps) : 81,360
Packets Rx (Pkts/sec) : 125,760 Packets Tx (Pkts/sec) : 116,096
Utilization Rx : 03.21 % Utilization Tx : 00.81 %

[parnassus]

Hi Peter, that's pretty strange. I guess you're probably using J9546A or J9538A as 10G eight ports modules on your VSF pair...if you are using J9538A modules you are probably using DAC (J9281D, as example) between 1/C8 and 2/C8 as your VSF Link, isn't it? Then...were ethernet statistics properly cleared on involved ports? ...in other terms are you sure that transmitted packets drops (Drops Tx) refers to current status of the remaining single active VSF Link 1/C8 <--> 2/C8? alternatively are you using 10GBase-T patch cable on J9546A? if so...used Cat. 6 patch is OK between above ports?

Edit: on our VSF (more than 650 days of uptime) VSF links all show 0 drops (cabling rearrangement was done by shutting down involved ports first IIRC).

[parnassus]

Hello Peter, do you have any news? please keep us updated, VSF related threads are, IMHO, hot topics.