Bad response of vscan server causes high volume latency
Applies to
ONTAP 9
Issue
- High read/write latency due to bad response from vscan server.
scan_request_timedout
can be seen by following the below steps.- ASUP ->
VSCAN-EVENTS.XML
orvserver vscan show-events
command via advanced mode may report:scan-internal-error
scanner-busy
scan-timedout
- ASUP ->
VSCAN-CONNECTION-STATUS-ALL.XML
may report that the vscan servers configured on the CIFS vserver aredisconnected
. - Start collecting the statistics by running the below command and then reproduce the slowness issue:
cluster1::> set -privilege diag
cluster1::*> statistics start -object offbox_vscan_server -sample-id vscan_test1
- Stop the statistics once the issue is reproduced:
cluster1::*> statistics stop -object offbox_vscan_server -sample-id vscan_test1
- To view the statistics, use the command below to check for the total requests dispatched and the number of requests
timedout
. The below example shows all 11 requests dispatched weretimedout
which happens due to a bad vserver response:
cluster1::*> statistics samples show -sample-id vscan_test1
Object: offbox_vscan_server
Instance: cluster_name:<node IP>:Node_name
Start-time: 3/6/2024 12:09:18
End-time: 3/6/2024 12:15:59
Elapsed-time: 401s
Scope: scope_name
Number of Constituents: 1 (complete_aggregation)
Counter Value
-------------------------------- --------------------------------
scan_request_dispatched 11
scan_request_dispatched_rate 0
scan_request_pending 1
scan_request_requeued 0
scan_request_timedout 11