Vmanage RESTAPI statcycletime & entry_time

[AD32111]

I see a difference of 13 mins between statcycletime & entry_time. Sometimes it is 23 mins. 

How can we know the min/max difference between entry_time & statscycletime. What can cause this difference? 

statistics/approute API Response:  

{

"remote_color": "mpls",

"fec_re": 0,

"vqoe_score": 5,

"device_model": "vedge-cloud",

"latency": 58,

"tx_octets": 67526,

"dst_ip": "172.16.2.14",

"local_color": "mpls",

"src_ip": "172.16.2.10",

"sla_class_names": "Bulk-Data",

"loss": 6,

"total": 600,

"tx_pkts": 417,

"fec_tx": 0,

"rx_octets": 45284118,

"statcycletime": 1610021400086,

"state": "Up",

"local_system_ip": "1.1.7.101",

"tenant": "default",

"entry_time": 1610020052910,

"loss_percentage": 1,

"rx_pkts": 363090,

"vmanage_system_ip": "1.1.7.101",

"fec_rx": 0,

"src_port": 12366,

"jitter": 32,

"remote_system_ip": "1.1.8.101",

"vdevice_name": "1.1.7.101",

"proto": "IPSEC",

"vip_idx": 24254,

"dst_port": 12346,

"name": "1.1.7.101:mpls-1.1.8.101:mpls",

"sla_class_list": "0,1",

"tunnel_color": "mpls:mpls",

"host_name": "BRF-VE-1",

"id": "AXbcwO4y7xjbcbVX8PFy"

}

There is a difference of 23 mins 

Epoch date  Human-readable date (GMT) 

1610021400	2021-01-07 12:10:00
1610020052	2021-01-07 11:47:32

[msuchand]

Hi , 

'entry_time' is the time at which statistics entry was created by edge router and 'statcycletime' is the time at which vManage has processed that entry after getting it from edge router.  one of the possible reason for the difference is the statistics collection interval because vmanage collects statistics at certain specified interval.  

for consuming the statistics the 3rd party applications should be looking at only 'entry_time' as that indicates when exactly that particular statistics entry was created on edge router.

Thanks,
Sai

[serene365org86747]

Thank you

[AD32111]

Hi Sai,

I notice that the entry_time is different on different vEdges. Also set the stat collection interval to 10 mins on vManage same as the device collection time which is 10 mins.

Why is vManage poll at 11:50 not getting the device snapshots from 11:37 and 11:47. 

For the below device/tunnel, current time is 12:13, latest snapshot from device is 11:47, so there is a delay of 26 mins. 

If i have 500 devices, this will vary for each device. I would want to use bulk API instead of realtime APIs for getting aggregated throughput per app/tunnel. 

What should be my start/end entry_times for the query for bulk api's so that i can aggregate octets to get throughput stats per app/tunnel. 

Bulk API : Current Time 12:13. Time slot : 11:40 to 12:13
stat cycle time entry_time stat cycle time (UTC) entry_time (UTC)
1610021400087 1610020052910 07/01/21 12:10 07/01/21 11:47
1610020800084 1610019452913 07/01/21 12:00 07/01/21 11:37
1610020200089 1610018852914 07/01/21 11:50 07/01/21 11:27

Monitoring API :  Current Time 12:13. Time slot : 11:40 to 12:13
1610021400087 1610020052910 07/01/21 12:10 07/01/21 11:47
1610020800084 1610019452913 07/01/21 12:00 07/01/21 11:37
1610020200089 1610018852914 07/01/21 11:50 07/01/21 11:27