WHEA-Logger flood - Event ID 17

[Keltec]

Hi all,

Thanks to this thread I was able to do more digging and find the problem driver (well for us at least but this post may be relevant to others on this thread).  The problem is not fixed but a workaround in place.  

We were getting 2x WHEA PCIe codes - the 0x8086:0xA116 one (suggesting Intel) which was very irregular, and this one:

Vendor ID: Device ID: 0x10EC:0x8168

Now initial serches showed more on the former, not the latter (hence finding this thread), so we focused on the issue being HP and Intel, but the 0x10EC Vendor ID is Realtek.  Looking at the reports of what people are sending on this thread and with a developing spreadsheet (to determine commonality), it looks like Realtek is coming up trumps.  

So we disabled the Realtek GBE Family Ethernet NIC in our customers HP Pav AIO 23".  They are not using the NIC but the WiFi Adapter instead (Broadcom adapter), so it was safe to disable  and the problem went away.  Not only did disabling the Realtek NIC stop erroring its head off, but it stopped the Intel error aswell.  I can't find the Intel A116 code in the PCIe database either, which suggests that this ID might be anomalous and a PCI bus resetting or something as the result of the realtek?  I'm not expert here, just making educated guesses now... 🙂

We found the realtek NIC (lo and behold) was updated on 03/15/2016 - at the same time the customer reported the problems.  The driver details of this install are:

Realtek 8168

Driver version 10.3.723.2015

Driver date: 23/07/2015

First Install date: 15/03/2016

So we thought OK lets update the Realtek NIC and see if a newer driver fixes the problem:

Realtek 8168 (reports as Realtek PCIe GBE Family Controller

Driver version 10.7.107.2016

Driver date: 01/07/2016

First Install date:  19/03/2016.

Nope.. No difference.  Starts off OK after reboot, then 2 mins in, it can generate errors every ~5secs

What's worse, is the Rollback Driver option is no longer available and it's late so I'm not firing off a system restore just yet.  If I do a system restore though, I'll have to disable Windows update, software assistant etc., and that's based on the proviso that this is a software issue.  

At least we have a workaround.

Tech details - the error:

A corrected hardware error has occurred.

Component: PCI Express Endpoint
Error Source: Advanced Error Reporting (PCI Express)

Bus: Device:Function: 0x4:0x0:0x0
Vendor ID: Device ID: 0x10EC:0x8168
Class Code: 0x10000

The details view of this entry contains further information.

PC Type:

HP Pavillion 23" All in One (Product ID:  23-q103a).

Windows 10, i5-6400T, Realtek card reader, Realtek GBE (per above), 

Intel Graphics 530 + AMD R7 A360 GPU?

WiFi:  Broadcom BCM43142 Wireless-N adapter

Questions for HP on this (OT): 

Do you want to add a few more vendors to the hardware list for a single unit next time?  This looks like a shotty PC I would have built back in the 90's.  Is there really a cost benefit in MFR to retail by using SOOOO many vendors and presuming they will all work harmoniously, for marginal financial benefit? Seems to me the more vendors you use in a single product the higher the likelihood of compatibility issues and problems at the customers expense.

Realtek - what happened to your QA?  Judging by the looks of this issue, it's not just HP and the common thread in the last few pages is that this is a realtek problem - regardless of whether its networking, card reader etc., it's starting to look like the pitch forks are coming your way.

To both:  I don't know the source of the update (HP Assistant or Windows Update) so Realtek is more likely to blame here than HP (in my circumstance).

As my google searches showed up nothing for the codes above, I've kept them here for completeness in the hope that this will help others (and your mileage WILL vary :)).

Jase.

[Rien72]

Have disabled all 3 realtek devices.

Realtek PCIE CardReader

Realtek PCIe FE Family Controller

Realtek RTL8723BE 802.11 b/g/n Wi-fi Adapter

And indeed the flood stops.

After enabling PCIE CardReader :  Still nothing in eventviewer

After enabling PCIe FE Family Controller : Still nothing in eventviewer

After enabling RTL8723BE 802.11 b/g/n Wi-Fi adapter : Flooding eventviewer.

Seems here the wifi card is to blame.

[Bipinvsnl]

@REO51ST 

Realtek driver version : 10.3.723.2015

[evans89lfc]

@Rien72, i have an intel wifi card will disable that and see if the error messages stop

[Keltec]

So after 2 days of testing, it appears that for us the errors have stopped but the problem persists.  We tried a system restore last night but that did not resolve the problem.

Next steps will be to run a hardware diag, reset BIOS to defaults and then log a case with HP.

Will let you know how I go.

[Keltec]

OK so no good.  Eventlogs definitely date the issue back to 15/03, but what's interesting is that it wasn't the network card drivers but the PCIe Card Reader Drivers.  Seems that this is the most common of responses to date.

Have done a heap of changes today including 

* BIOS reset to default and update

* System Restore on the weekend

Issue still persists.  Since those changes around 5 hours ago, we also noticed the card reader reporting WHEL errors now.  We're now thinking its either a hardware issue or Realtek hardware failure. 

The other Intel issues also point to the PCIe Root Complex where 0xA116 and 0xA118 were reported within seconds of a Realtek NIC failure (but since it has been disabled, no reports here).  

We won't know until tomorrow if disabling the Realtek Card Reader has improved the issue or resolved it (as it's hard to discern if the applications are ok or if our Internet connection is laggy, making remote support a trial in itself), but will report back.

At this stage, the customer could have purchased 2x new AIO's for the amount of time we have spent on this, despite the warranty is with HP (and in warranty) and we're unlikely to get any form of compensation for it either.

I'm logging a case with HP now and will let you know if there's any outcome there other than replacing the system board.

Jase.

[evans89lfc]

at the moment my case is with HP in regards to my laptop that has the fault, and waiting to hear back in regards to compensation

[TheOldTech]

I have an ASUS N552VX that is having the  WHEA Logger errors. Just by disabling the Realtek NIC in device manager

it stopped the errors. Realtek has just released a new PCIE network card driver and it seems to have stopped the errors

with the NIC re enabled, hope it continues. Here is a link to the driver.

http://www.realtek.com.tw/downloads/downloadsView.aspx?Langid=1&PNid=13&PFid=5&Level=5&Conn=4&DownTy...

Cheers TheOldTech.

[Rien72]

Just to report back.

 This morning I checked the device manager, clicked the wifi adapter and did check for updated drivers.

And it found the one of 26.02.2016.

Driver version : 2023.28.115.2016

The flood has stopped in eventviewer about WHEA. Only with an restart I get around 400 entry's in eventviewer.

After clearing it, no new entry's appear.

[TheOldTech]

Sorry for the false hopes. My problem returned with the new driver loaded.

Guess we will have to wait for Intel, or a bios upgrade.

Cheers TheOldtech.