Frequent iScsiPrt 20 errors

Question

Storage array is SCv2000, connected to 2 PC6224 switches. 2 R710 servers running Server 2019 in a HyperV Failover Cluster. Each server has 2 BCM5709C iSCSI dual port cards.   Two ports are dedicated to the OS, two are dedicated to HyperV Virtual Switches. I am having problems with a Server 2019 virtual machine receiving frequent iScsiPrt errors in the event log.  Whenever there is any activity directed at the iSCSI targets, I get iScsiPrt event 20 followed immediately by iScsiPrt Event 34. Event 20: Connection to the target was lost. The initiator will attempt to retry the connection. Event 34: A connection to the target was lost, but Initiator successfully reconnected to the target. Dump data contains the target name. Corresponding errors in the Storage Manager log:  2019-08-23T15:26:58.616515-04:00 Warning SCSITarg Error IORPTracekey[2043d000004fd] MZ[ Removed by Moderator 114] Sub[Virtualization] LUN[1] [Write_10] Cdb[2a00005e418800000800000000000000] Top Controller 2019-08-23T15:26:58.616553-04:00 Warning SCSITarg Error IORPTracekey[2043d000004fb] MZ[ Removed by Moderator 114] Sub[Virtualization] LUN[1] [Write_10] Cdb[2a0001d7715000020000000000000000] Top Controller 2019-08-23T15:26:26.540394-04:00 Warning QRQCM QRQCMConnection CA Activate Failed: ControllerId=0x00034380 ObjId=0 Top Controller   I do not get these errors on the hosts, just the VM.  I also do not get these errors on another VM using Server 2012R2. I have rebuilt the VM from the ground up twice, once making sure to do all available Windows Updates, and again without doing any updates.  The behavior is the same. The behavior persists no matter which host is currently hosting the VM. Does anyone have any suggestions on how to get these errors to clear?  They seem to be having a performance impact on the SQL database attached to this VM.

DELL-Sam L · Answer

Hello InboundTech, What you will need to call into support so that we can remote into your SCv2000 & trace why that error is happing on your host.  Please let us know if you have any other questions.

subale · Answer

Did you find any think on this issue, i am same issue on my VM's on Hyper-V, but no errors on Host

axel · Answer

Hello,

If this is still actual, we've managed to solve this issue by disabling Recv Segment Coalescing on guest's Hyper-V Network Adapter designated for iSCSI traffic. This is Windows Server 2019/2022 feature. Guests with 2016 Server work fine.

Tomas Kalabis · Answer

Same problem here.  HPE DL360 g10 with dedicated 2 x 10 GbE for iSCSI , S4112 for networking and storage array is SC3020 with Failover Cluster Hyperv 2019.

DiegoLopez · Answer

Hello @Tomas Kalabis,

As previously suggested by my colleague @DELL-Sam L, you may need to call into support so we can connect remotely to your SC3020, export a log and check the error.

Please, let us know if you have more questions.

Regards.

mpaszek · Answer

Hi DELL Support.

On server Dell Storage SCv3020 I got some wornigs:

QRRCMConnection CA Activete Failed: ControllerOD-0x000D24C2 ObjId=0

Could I ask for some feedback and some help info.

Many thanks in advance

mpaszek · Answer

Thanks for feedback I didn't opened ticket, I started form forum. Please let me know the best way how to open ticket on support.

R1Pilot76 · Answer

Did you ever get a fix for this? I am having the same issue and Dell have looked at it and couldn't find anything wrong.

DELL-Joey C · Answer

Hi,

Apologies that the support could not find the root cause. Probably best to provide more information about your configuration, storage model, OS and error messages, maybe the public users may have some solution for the issue.

MTK_User · Answer

Hi, This solution have worked for us. Just disabled VMQ and IPSec for the Virtual NICs destinated to iSCSI traffic in the VM config. Regards Guille

Compellent

Was this post helpful?