"no instances, entering standby Poller=aces collector=Zapi:SnapMirror task=data"" | NetApp | Page 1

violet radish Jan 29, 2024, 7:00 PM

#

Can't figure this one out. (What else is new?) My passwords are good, and the curl test gets the version, so it isn't the password or the account.

#

"no instances, entering standby Poller=aces collector=Zapi:SnapMirror task=data""

haughty kite Jan 29, 2024, 7:03 PM

#

hi @violet radish can you screenshot the actual log line or send the log files to ng-harvest-files@netapp.com

no instances typically means there are not instances of that object on the cluster. In this case, no SnapMirrors. It should be an INFO message and not an error for that reason

#

does the cluster in question have SnapMirrors configured on it?

violet radish Jan 29, 2024, 7:05 PM

#

I have lots if snapmirrors, both volume and SVM

#

This is not the destination

#

They just rebooted my box, I can't see what the destination is saying, that is a positive message, but I still don't see any SMs. I'll get it in a few

haughty kite Jan 29, 2024, 7:08 PM

#

looking at the log files you shared earlier, I see the same in those logs.
INF collector/collector.go:342 > no instances, entering standby Poller=aces collector=Zapi:SnapMirror task=data - that Zapi is returning nada. Let me find the right curl for you

#

try this, replacing user,pass, and $ip
curl --user user:pass -snk 'https://$ip/api/private/cli/snapmirror'

violet radish Jan 29, 2024, 7:31 PM

#

Getting a user is not authorized on both source and destination

haughty kite Jan 29, 2024, 7:32 PM

#

do you get similar with the following? Same as before, replace user, pass, and ip with your values
curl --user user:pass -snk --data-ascii '<?xml version="1.0" encoding="UTF-8"?> <netapp xmlns="http://www.netapp.com/filer/admin" version="1.30"> <snapmirror-get-iter/> </netapp>' -H "Content-Type: text/xml" -X POST 'https://$ip/servlets/netapp.servlets.admin.XMLrequest_filer'

violet radish Jan 29, 2024, 7:37 PM

#

Should I assume the curl command is trying to get to netapp.com? I live in a dark site

haughty kite Jan 29, 2024, 7:40 PM

#

no, the curl command is talking to your ontap cluster. Make sure you replace $ip with your cluster admin ip, same as what you specifiy in your harvest.yml file

violet radish Jan 29, 2024, 7:45 PM

#

I think I typed it erroneously, no output; let me try again

haughty kite Jan 29, 2024, 7:47 PM

#

sorry, that's rough to type out. Was hoping your could copy/paste
i tried locally and it also works without the XML namespace so you can use this too
curl --user user:pass -snk --data-ascii '<?xml version="1.0" encoding="UTF-8"?> <netapp version="1.30"> <snapmirror-get-iter/> </netapp>' -H "Content-Type: text/xml" -X POST 'https://$ip/servlets/netapp.servlets.admin.XMLrequest_filer'

violet radish Jan 29, 2024, 8:00 PM

#

I think I finally got it correct: here is what it said on the SM source:

#

<?xml version'='1.0' encoding='UTF-8' ?>

#

<!DOCTYPE netapp SYSTEM 'file:/etc/netapp_gx.dtd'>

#

#

haughty kite Jan 29, 2024, 8:03 PM

#

thanks. that matches the log message we started this conversation with. <num-records>0</num-records> means there are no instances

violet radish Jan 29, 2024, 8:04 PM

#

Should I try on the destination?

haughty kite Jan 29, 2024, 8:06 PM

#

the log message you pasted at the top is about talking to cluster aces. Is that the cluster you used in the curl commands above? If so, no need to look further. The log message is correct, Harvest sends that ZAPI to ONTAP asking for all snapmirror instances and ONTAP tells Harvest there are none. We print that message so we (and you) understand there won't be any metrics for that object

violet radish Jan 29, 2024, 8:07 PM

#

Got a lot of output on the on acesd which is the SM destination

#

So how do I collect data?

haughty kite Jan 29, 2024, 8:10 PM

#

Can you explain what you think should be happening that is not happening? I thought you question was about the log message you pasted, which hopefully I've explained. And yes, acesd has 263 instances

violet radish Jan 29, 2024, 8:14 PM

#

When I open the SnapMirror dashboard, I get 0's and 'No data'. I see the Datacenter as my Secondary, but SourceClusteer is ALL and I cannot see individual source clusters. The same for Nodes, and SVM's, Volumes and Source Volumes

haughty kite Jan 29, 2024, 8:15 PM

#

ok, that's the first mention of dashboard, got it

violet radish Jan 29, 2024, 8:15 PM

#

I'm sorry, sometimes I ass-u-me. 😉

haughty kite Jan 29, 2024, 8:16 PM

#

no worries! easy to do when you've been troubleshooting

#

what version of Harvest are you on?

violet radish Jan 29, 2024, 8:17 PM

#

Thank you... You are a gentleman and a scholar

haughty kite Jan 29, 2024, 8:18 PM

#

and sounds like when you click the SourceCluster dropdown you only see All and not clusters like this?

violet radish Jan 29, 2024, 8:21 PM

#

23.11.0-1

#

I only see my Secondary Dattacenter, which is the SM/DR cluster

#

But no omdividual cluster, just ALL

haughty kite Jan 29, 2024, 8:25 PM

#

the datacenter variable gets its values from this query

#

and the sourcecluster variable gets its values from this query

#

which is a join of snapmirror_labels and volume_labels so either the SnapMirror collector or the Volume collector must have errors if you aren't see the data you expect

#

can you send your log files to ng-harvest-files@netapp.com and/or check for any ERR in them?

violet radish Jan 29, 2024, 8:29 PM

#

How did I create a Primary and Secondary Datacenter?

haughty kite Jan 29, 2024, 8:30 PM

#

datacenter is taking directly from your harvest.yml like so

violet radish Jan 29, 2024, 8:31 PM

#

I didn't do it via yml, I am pretty sure it was a configuration item. Oh well, yes I can send logs, but my partner is on vacation and won't be in until Wed. I'll send them then. Thank you

#

Duh, found it

haughty kite Jan 29, 2024, 8:33 PM

#

the only place we read datacenter is from harvest.yml

#

where was it?

violet radish Jan 29, 2024, 8:33 PM

#

When you add the array, there ia a GROUP field.

#

NABOX/Systems/Storage Systems

haughty kite Jan 29, 2024, 8:37 PM

#

Nabox is a front-end to Harvest which means Nabox added datacenter: what-you-typed-for-group in your harvest.yml

violet radish Jan 29, 2024, 8:38 PM

#

Yes. Part of the config instructions, I believe

haughty kite Jan 29, 2024, 8:38 PM

#

glad you got it sorted

violet radish Jan 29, 2024, 8:38 PM

#

Now I just need to get it working

haughty kite Jan 29, 2024, 8:38 PM

#

I've got to run to a meeting. I'll catch you later

violet radish Jan 29, 2024, 8:38 PM

#

ok

#

Thank you

violet radish Jan 30, 2024, 12:23 PM

#

@zealous marsh Yann, have you seen any issues with the SnapMirror dashboard not reporting any data in NABOX 3.3, Harvest 23.11.0-1?

haughty kite Jan 30, 2024, 2:32 PM

#

hi @violet radish how is the snapmirror dashboard looking today?

violet radish Jan 30, 2024, 7:15 PM

#

Same sir. 0's and "No data" ; I haven't had a chance to look in logs

haughty kite Jan 30, 2024, 7:16 PM

#

did you fix the datacenter problem we dicussed yesterday?

zealous marsh Jan 31, 2024, 4:50 PM

#

violet radish <@323825973718155265> Yann, have you seen any issues with the SnapMirror dashboa...

No I think it's the first time I see this

narrow swan Feb 1, 2024, 7:52 AM

#

@violet radish As you have seen this info panel for SnapMirror dashboard that both the source and destination cluster must be monitored by Harevst and export the metrics in same Prometheus instance. Just to make sure, Could you please share the query result for this in your NABox prometheus instance.
You can access the prometheus here, https://[nabox ip]/prometheus/ replace [nabox ip] with your nabox address.

Run this query in promQL
group by (datacenter, cluster)(snapmirror_labels)

violet radish Feb 5, 2024, 5:09 PM

#

narrow swan <@734108538015842375> As you have seen this info panel for SnapMirror dashboard ...

I presume this is the prometheus screen; I executed the above, the only response was, "{cluster="acesd",datacenter="Secondary"}

haughty kite Feb 5, 2024, 5:14 PM

#

hi @violet radish yes, the PromQL Hardik shared is meant to be run in Prometheus
your response means there is only one cluster being monitored that is exporting snapmirror metrics. Is that what you expected?

haughty kite Feb 6, 2024, 12:55 PM

#

hi @violet radish can you send you log files so we can check if there are any issues? email them to ng-harvest-files@netapp.com

#"no instances, entering standby Poller=aces collector=Zapi:SnapMirror task=data""