#CIFS Auth Latency
1 messages · Page 1 of 1 (latest)
hi @strange monolith I'm not aware of any counters that would help troubleshoot LDAP. @terse sail do you know of anything?
Maybe these could help? https://kb.netapp.com/Advice_and_Troubleshooting/Data_Storage_Software/ONTAP_OS/How_to_troubleshoot_LDAP_issues_in_Windows_Active_Directory
https://www.netapp.com/media/19423-tr-4835.pdf?v=10-6-2020-9-13-16-PM
Hi Chris, Ralf in the AMS support center uploaded our perf archive to an internal grafana box and it had a nice auth latency display
sweet, that implies there are counters we could consume
It would be fantastic if we could get that built into the CIFS latency views
if you know which counters are being used, please share. Otherwise, I'll connect with Rolf later and try to find them. Are you working with Rolf D, Rolf H, or Rolf J? 😃
Do you have a case number?
@strange monolith ?
It could be from SMB2 view in PAS, which would be the SMB2 object.
Sorry mate, let me find it, 2009463576
Yeah it was from the SMB2 object I remember that much
Thanks @terse sail @strange monolith
yes smb2 object is available in Ontap Zapi side. However this counter is not present in Rest side.
Is that easy to add to restapi? is there any equivalent? Just something to check for AD latency session setup etc would be ideal
ONTAP Rest performance counter API are available from 9.12 onwards. One need to request ONTAP to add any missing/new object as needed. For now, this object is available in Zapi so it should work till Zapi are available in ONTAP.
Would it be Possible to include a range of CIFS auth latency counters into the next release on the dashboards?
is your requirement to add similar smb2 view dashboard from PAS to harvest?
Ideally mate, Just want to check auth latency is good, if that is SMB2 view that Ralf and the support guys use with PAS that will do
Sure We'll add it to roadmap.
Added a feature request https://github.com/NetApp/harvest/issues/1696
Thank you Rahul
@strange monolith Would you mind filling up this SharePoint form to address the SMB 2 REST gaps. The REST team follows this process to track incoming requests for gaps https://forms.office.com/Pages/ResponsePage.aspx?id=oBEJS5uSFUeUS8A3RRZbOtlEKM3rNwBHjLH8dubcgOVURVM2UzIzTkQzSzdTU0pQRVFFRENZWlAxNi4u
Sure
We are running 9.10P11 and collecting via ZAPI (need to enable restapi for the service account) . Is the dashboard available to use now?
hi @strange monolith yes these changes were added a couple of weeks ago for ZapiPerf and are available in nightly. Screenshots https://github.com/NetApp/harvest/pull/1754
the smb2.yaml template is disable by default in conf/zapiperf/default.yaml
Ah theres no data on the dashboard being collected
Ah now enabled! I will report back tomorrow with my findings
Still showing as no data 😦 Our service account is configured as per https://nabox.org/documentation/configuration/ and dashboard, and the smb2yaml is enabled
let's check your poller logs - can you ssh into nabox and run dc logs nabox-harvest2 | grep SMB2 or if you'd rather, grab the logs and email them to us and we'll take a look. ng-harvest-files@netapp.com https://nabox.org/documentation/troubleshooting/#collecting-logs
If nothing is returned that means the SMB2 collector was not started. can you ssh into nabox and check that SMB2 is enabled in /opt/harvest2-conf/conf/zapiperf/default.yaml thanks for the screenshots. That does show that smb2 is not there
did you download a nightly build of Harvest and add it to nabox?
something's wrong since you screenshot does not match https://github.com/NetApp/harvest/blob/main/conf/zapiperf/default.yaml
oh! scroll down further and see if SMB2 is down near the bottom 😄
can you grab the poller logs for one of your pollers that should be using SMB2 and we'll take a look, https://github.com/NetApp/harvest/wiki/FAQ#how-do-i-share-sensitive-log-files-with-netapp
dc logs nabox-harvest2 > nabox-harvest2.log zip if it's too big
Next stupid question, where would that log file be?
Just sent them over to you
got it
thank you
there are some REST permission issues we can work on later - the won't affect this issue. from your ssh shell can you run dc down. This will bounce the pollers
i see that pollers were restarted version="harvest version 23.03.24-nightly (commit cea86732) (build date 2023-03-24T00:25:15-0400) linux/amd64 but there is no mention of trying to start SMB2. This is the list of ZapiPerf collectors started from your logs
there are 632 instances of API request rejected => Insufficient privileges: user 'grafana' does not have read access to this resource This means the ONTAP user named grafana does not have ZAPI read permissions. Details https://netapp.github.io/harvest/23.02/prepare-cdot-clusters/#ontap-cli
and that's because of a new template that was added so not related to SMB2
Zapi:QosPolicyAdaptive and Zapi:QosPolicyFixed
for the recently added qos policy templates these additional permissions are required
security login role create -role harvest2-role -access readonly -cmddirname "qos adaptive-policy-group" security login role create -role harvest2-role -access readonly -cmddirname "qos policy-group"