NAbox V4 migration: "gap" in collected data | NetApp | Page 1

vivid scroll Nov 26, 2024, 3:17 AM

#

Tried our second migration (first one was in a lab environment with few data)
NAbox V3 has been upgraded according to documentation
Migration took around 12 hours to complete (data from 2023-07)
The systems are ok in NAbox V4, but the 12 hours it took for the migration to complete are "missing" in V4 dashboard.
Tried to redo the migration but didn't succeed as the systems are already present in V4.
Any chance to fill this gap ?

Also regarding "best practices" for migration is it best to migrate all systems together or perform a migration system per system?

any advice ?
Any way to add some VM worker to speed up the migration process or to change the "min time" for migration, guess we could loose data from last year...

faint walrus Nov 26, 2024, 5:19 AM

#

@crisp rose

crisp rose Nov 27, 2024, 12:52 PM

#

Sorry I'm not ignoring this. So in theory, when a poller is added by the migrate tool, it is immediatly started, do you remember if that was the case for the period you have a gap ? If you zoom inside the gap do you see some data ?

vivid scroll Nov 27, 2024, 2:43 PM

#

no data in the gap.
we had a permission issue (didn't pay attention to role changes between harvest v22 & v24)
so no data was collected during migration and until permissions were fixed...
will give another try to the migration this upcoming weekend.
Any parameter to pass to migration to filter older data?

crisp rose Nov 27, 2024, 10:07 PM

#

If you run the manual steps for the migration, you can refer to this documentation https://docs.victoriametrics.com/vmctl/#migrating-data-from-prometheus and use --prom-filter-time-start and --prom-filter-time-end

vmctl

Documentation for VictoriaMetrics, VictoriaLogs, Operator, Managed VictoriaMetrics and vmanomaly

#NAbox V4 migration: "gap" in collected data