E010A0401 EID_IP_ERRORS_EXCESSIVE

[testmado662]

Hello,

I found old posts (2018) about that topic but no real solution in old posts. Some replace their 1Gbps Ethernet adapter by 10Gbps, others spoke about a version change with an impact about that problem.
I don't find anything clear. Some new tracks about that problem and possible causes ?

I've identified 6 lefthand frames on 27 we currently host, with Excessive packet loss.
If 21 frames don't get that problem I don't find a clear discriminant for the ones wit problem. Problem appears on P4000 or VSA, in ALB or in Actice/Passive.
The only common point I currently found, all 6 frames get 12.7 or 12.8 lefthand version, 5 in 12.7 and 1 in 12.8.

We currently don't use HPE or CISCO swiths but Giga industrial switchs. We planned to change Storage switchs but all frames are currently connected to industrial switch.
I saw posts with users using HPE or CISCO and they got the problem.
We don't use STP and tne problem is not persistant, someday we will have warning massages and for one week no Network problem appears.
I verified Bandwidth and the all get 40MB/sec
We don't configure MTU and Speed duplex on lefthand, we are using what defined on switch (MTU=1500, 1000FD, no Jumbo Frame. Does it's necessary to define MTU, Speed and Duplex ?

If somebody get some solutions or tracks to explore, in order to find the origin of that trouble?

Thanks for your help and feedback,

[danevald]

Just to throw my 2 cents in on this...

We've had VSAs running for 5+ years on multiple locations, all running same server hardware, same brand of switches(HP/Aruba) and we've only seen the issue on 1 location. This location is larger network wise, with many more switches, devices and users. Also core switch is 1st gen 5400 series and on other locations we're running newer revision of the same model, so when we recently upgraded the storage part of our network to StoreFabric models with 10/25/100G links and separated the traffic away from the core switch, we we're sure we'd seen the last of the problems. But we are still seeing the issue, and like you say - some weeks go by with no errors and then sometimes we have multiple error during a week.

In the upgrade we've switched to use Nimble arrays, removed almost all the load away from the VSA and the Nimbles are never reporting any network issues.

So I must say, I've come to the conclusion that something is bothering the software on the VSAs and in our case, we are not going to throw any more time at it...

[testmado662]

Hi Daneval,

Thanks for your feedback.

We get the same sensation, it's hard to know why some systems get that incident and others no. We can replace P4000 or VSA to fix the problem but the only which get the solution is HPE.

I would like to know now, what HPE think about that, if they identified the problem and if a patch is planned.

Now, I'm going to work with RCA (Root Cause Analysis) rules and try to identify something common in systems with incident.

Regards

[testmado662]

Hello,

Does someone get another feedback to share ?

Thanks