Hi Everyone,
I configured cluster of 4 Supermicro X10DRIservers to test storage spaces direct. Supermicro server configuration:
- 2 x Intel Xeon E5-2640v3 CPU
- 10 x 32GB 2133MHz DDR4 RAM
- 2 x SSD 400GB SATA Intel S3610
- 2 x HDD 2TB SATA Seagate ST2000NX0253
- 6 x SSD 800GB SATA Intel S3610
- Intel X520-DA2 E10G42BTDA Network adapter
All was fine, cluster and S2D were working without any problems. But after a month, I replaced Intel X520-DA2 E10G42BTDA Network adapter on Mellanox ConnectX-3 Pro Ethernet Adapter. After that, I started receiving Warnings in Cluster Events log:
"Cluster Shared Volume 'VD1' ('Cluster Virtual Disk (VD1)') has entered a paused state because of 'STATUS_IO_TIMEOUT(c00000b5)'. All I/O will temporarily be queued until a path to the volume is reestablished."
I checked health of physical disks - they fine, also I checked network and found I had about 500K Received Discarded Packets per server. Itlookspretty bad, but I'm not sure. Is it a problem? Orshould I continue my investigation?