Azure CVO中的磁盘延迟高、IOPS低
- Views:
- 1
- Visibility:
- Public
- Votes:
- 0
- Category:
- cloud-volumes-ontap-cvo
- Specialty:
- perf<a>2009年888483</a>
- Last Updated:
适用场景
- ONTAP 9
- CVO
- Azure
问题描述
- 此延迟是通过QoS统计信息中的磁盘延迟来衡量的。
Cluster::> qos statistics volume latency show -vserver vs0 -volume vol0
Workload ID Latency Network Cluster Data Disk QoS NVRAM Cloud
--------------- ------ ---------- ---------- ---------- ---------- ---------- ---------- ---------- ----------
vol0 6762 137.73ms 1.91ms 1.00us 22.93ms 112.89ms 0ms 0ms 0ms
vol0 6762 133.29ms 1.82ms 0ms 2.84ms 128.62ms 0ms 0ms 0ms
vol0 6762 78.48ms 1.96ms 0ms 7.69ms 68.83ms 0ms 0ms 0ms
vol0 6762 50.51ms 239.00us 0ms 995.00us 49.28ms 0ms 0ms 0ms
- statut命令指示在增加磁盘利用率的同时丢弃IOPS。
disk ut% xfers ureads--chain-usecs writes--chain-usecs cpreads-chain-usecs greads--chain-usecs gwrites-chain-usecs
/data_aggr1/plex0/rg0:
0a.00.18 100 280.18 280.18 64.00 67 0.00 .... . 0.00 .... . 0.00 .... . 0.00 .... .
0a.00.10 100 280.18 280.18 64.00 64 0.00 .... . 0.00 .... . 0.00 .... . 0.00 .... .
0a.00.19 100 280.18 280.18 64.00 71 0.00 .... . 0.00 .... . 0.00 .... . 0.00 .... .
0a.00.12 100 280.18 280.18 64.00 69 0.00 .... . 0.00 .... . 0.00 .... . 0.00 .... .
0a.00.17 100 280.18 280.18 64.00 75 0.00 .... . 0.00 .... . 0.00 .... . 0.00 .... .
0a.00.16 100 280.18 280.18 64.00 77 0.00 .... . 0.00 .... . 0.00 .... . 0.00 .... .
- 磁盘IOPS从>3K减少到~200、磁盘利用率增加到100%、同时具有200_个待定IOPS和排队IOPS。
- EMS日志显示连接错误、如下所示
Tue Jul 16 14:47:05 +1000 [azpcvoause101c-01: pha_main000: scsi.cmd.floatTimeout:notice]: "Disk" device "0d.3L0": The page blob I/O operation timeout increased to "20" seconds: HA status "0x4": OSC Error "2": retry count "1": elapsed time "9712" milliseconds: cdb 0x88:00000000000272728820:0009. Disk 0d.3 S/N [sa82p4kn1zbcontainerotqmyqedc7iw] UID [sa82p4kn1zbcontainerotqmyqedc7iw] Target Address