#volume dissociation :)
1 messages · Page 1 of 1 (latest)
@ionic pollen Could you share metric query? Are there any changes to this volume label during this time range?
what could make a label change?
also i have this on several volumes
volume_avg_latency{datacenter=~"$Datacenter",cluster=~"$Cluster",node=~"$Node"}
and
topk($TopResources, avg_over_time(volume_avg_latency{datacenter=~"$Datacenter",cluster=~"$Cluster",node=~"$Node"}[3h] @ end()))
what is funny is that grafana see that it is the same volume
Okay. Were there any aggr move operation for this Volume?
Could you share Harvest version?
From your screenshot, it appears that we have two time series for this volume at the same timestamp. Could you please check in Prometheus by running a query to see how many records are returned for the relevant time range? Let's try a range of one week.
volume_avg_latency{volume="VOLNAME"}
no aggr move
Okay let’s check the Prometheus query. I suspect if we have multiple pollers polling same cluster.
funny 🙂
are you able to read?
the red line has an attribute "style=flexvol"
the blue doesn't
Interesting. Could you share Harvest version?
Yes upgrade is fine but I don't see any change in this area which will fix this issue. But you'll get other improvements since then.
Could you share Harvest logs here https://upload.nabox.org/meyu-wuho-vety
ok i will do
Thanks
FYI i'm not in nabox
That should be fine.
Log collection steps are available here
https://netapp.github.io/harvest/24.11.1/help/log-collection/
I have added a possible fix here in PR https://github.com/NetApp/harvest/pull/3361/files
We'll know more from logs.
Issue: https://github.com/NetApp/harvest/issues/3364
Thanks for the logs @ionic pollen . I have asked more details via email.
I have opened 2 issues to track these other issues.
https://github.com/NetApp/harvest/issues/3363
https://github.com/NetApp/harvest/issues/3362
Based on timeouts in logs, There are high chances that PR https://github.com/NetApp/harvest/pull/3361 will fix volume style issue.