Synchronization failing , Event ID 809 , Underlying Device

Software-based VM-centric and flash-friendly VM storage + free version

Moderators: anton (staff), art (staff), Max (staff), Anatoly (staff)

Post Reply
hmadra
Posts: 2
Joined: Fri Nov 23, 2018 4:32 pm

Fri Nov 23, 2018 4:38 pm

Hello

I just installed a 2 Node Cluster
2 x HP G8 with 2x 2.4GHz 8Core CPU , 64GB Memory
2 x 10GB Sync Channels , P2P Connected in Team
RAID 1 - 6 1TB Samsung EVO 860 SSD Each on HP 420i with 1GB Cache

I Created a 2.5 TB Standard Device on Samsung SSDs. However , the Synchronization between the Nodes is failing after 2-3 Minutes with the error
"HA Device iqn ....... : command "Write". Underlying device response time is longer than expected. Response time is 10 Sec."
I have restarted SYNC few time manually, but it keeps failing.

Any Ideas.

thanks

HS
Oleg(staff)
Staff
Posts: 568
Joined: Fri Nov 24, 2017 7:52 am

Mon Nov 26, 2018 11:49 am

First of all, we do not recommend any form of NIC teaming for ISCSI and synchronization traffic.
Also, do you have Jumbo frames enabled? If yes, please try to disable Jumbo frames.
hmadra
Posts: 2
Joined: Fri Nov 23, 2018 4:32 pm

Mon Nov 26, 2018 6:21 pm

So.. No Jumbo Frames on the SYNC Link or Any Links , including iSCSI ?
Cause StarWind KB Says otherwise

https://knowledgebase.starwindsoftware. ... important/
Oleg(staff)
Staff
Posts: 568
Joined: Fri Nov 24, 2017 7:52 am

Tue Nov 27, 2018 11:48 am

Some Broadcom based NICs have packet loss problems with Jumbo frames enabled.
As I already told, please try to disable Jumbo frames on both ISCSI and synchronization links.
Post Reply