Active IQ Unified Manager Discussions
Active IQ Unified Manager Discussions
Hello,
We are using harvest to get performance on cDOT 8.2 and 8.3.
Now we found a bug burt969339.(http://mysupport.netapp.com/NOW/cgi-bin/bol?Type=Detail&Display=969339)
So we want to stop executing "cifs-session-get-iter" from harvst. (we want to continue to use others)
Which part of code shuold we modify?
Regards,
Shuhei
Solved! See The Solution
Hi @ShuheiEndo,
Harvest uses "perf-object-get-instances" API call to get CIFS related stats. It accesses the counter manager subsystem which keeps track of stats pushed to it by other subsystems. I asked engineering if the bug could be caused via an API call to CIFS counters in counter manager and receved back:
> The counters are safe to query any time.
So I can confirm that Harvest cannot cause this bug.
Cheers,
Chris Madden
Storage Architect, NetApp EMEA (and author of Harvest)
Blog: It all begins with data
If this post resolved your issue, please help others by selecting ACCEPT AS SOLUTION or adding a KUDO
Hi @ShuheiEndo,
Harvest does not call this API so no edits are needed to avoid this bug. If you think Harvest is somehow causing you to hit this bug please send me a private message with more info including the support case #.
Cheers,
Chris Madden
Storage Architect, NetApp EMEA (and author of Harvest)
Blog: It all begins with data
If this post resolved your issue, please help others by selecting ACCEPT AS SOLUTION or adding a KUDO
As an aside, cDot 8.3.2P1 is now available which contains a fix for this bug. This patch level was released on the support site on April 15th. I had to install it Saturday to deal with another of the bug fixes it addressed - stable in our operation so far. I can't comment on how well it addresses the CIFS session issue as we've never hit that bug in practice.
Hope this helps.
Bob Greenwald
Lead Storage Engineer | Consilio
NCIE SAN cDot, Data Protection
Kudos and accepted solutions are always appreciated
Thank you, Bob.
I'll consider to upgrade.
Hi @ShuheiEndo,
Harvest uses "perf-object-get-instances" API call to get CIFS related stats. It accesses the counter manager subsystem which keeps track of stats pushed to it by other subsystems. I asked engineering if the bug could be caused via an API call to CIFS counters in counter manager and receved back:
> The counters are safe to query any time.
So I can confirm that Harvest cannot cause this bug.
Cheers,
Chris Madden
Storage Architect, NetApp EMEA (and author of Harvest)
Blog: It all begins with data
If this post resolved your issue, please help others by selecting ACCEPT AS SOLUTION or adding a KUDO
Thank you so much.
We restarted our harvest again !
(yes, harvest uses only numbers, not client's computer names)