Windows 2016 std failover cluster with HPE MSA2052 SAS

I

Ivan_KPD

Hello,
Concerning 2-node failover cluster windows server 2016 std
SAN: HPE MSA2052-SAS, 2 Hostserver: ML350Gen10 HBA:E208e-P

Everything works perfect but as soon as a MSA LUN gets part of the cluster , (Host)servers getting stuck in a restart/shutdown process.
This already happens if a MSA disk is added as “Available storage” in the Cluster manager.
If we take the disk offline the host proceed with its shutdown or restart process.

We did some extra tests and noticed that we don’t have the problem if we don’t’ use MPIO.
Of course, for this test we only configured one link to a LUN/server, in production we need multiple links(redundancy)-> also MPIO
For MPIO settings/policies we follow the HPE MSA2050 best practices, and tried also other options.

We are building Clusters this way (MSAxxxx fiberchannel/SAS) since Windows server 2008R2.
The most recent working combination is with windows server 2016std and a MSA2040 fiber channel and two ML350Gen9. This is the first config where it goes wrong and don’t find a solution ☹

So, we have the feeling that maybe a Windows or driver bug is causing this?

Thanks in advance for any suggestions,

Continue reading...
 

Similar threads

H
Replies
0
Views
55
Hinterleitner Andreas
H
P
Replies
0
Views
93
Pradeep Goonetillake
P
M
Replies
0
Views
53
Mladen Zecevic
M
Back
Top Bottom