2 Intern
•
130 Posts
0
45
November 21st, 2024 18:10
SRM Load-Balancer on Collector Hosts Dropping Balancing Property Of Device(s)
Hi All.
SRM 5.0.0. HITACHI VSP ARRAY discovered with REST API via single collector instance(Note no multiple instances or multiple collectors involved - which are known to cause issues).
But Load Balancer on Collector Host(Hitachi Collect) is giving multiple 'Dropping' errors like:
AbstractDecisionTree::dispatchValue(): Missing required balancing property device for value HDS-REST-Collector74XXXXLUN00:02:BARequests... dropping!
WARNING -- -- AbstractDecisionTree::dispatchValue(): Missing required balancing property device for value HDS-REST-Collector74XXXXLUN00:02:BALatency... dropping!
WARNING -- -- AbstractDecisionTree::dispatchValue(): Missing required balancing property device for value HDS-REST-Collector74XXXXLUN00:02:BAReadLatency... dropping!
- There are no errors in Hitachi-device-manager-collector Logs. Totally clean logs.
- As a try (KB Article: 000030933)Performance Polling period in Solution Pack (Device Manager Collect) increased to 60 minutes. No luck. Still dropping.
- Thought about KB Article: 000199345, but no idea which file to modify. May be not applicable.
- Could it be: KB: 000173587: The Load Balancer Connector (LBC) fails to recover after it goes into failover mode. But how to confirm this.
- Found exactly matchig KB:000059671 - Performance data missing for Hitachi arrays. But that KB has LUN IDs seems changing. In our case the LUN00:02 remains same. Deos this means only one LUN is affected. Others will be fine.
- Also, this point has been taken care already: Cross-Reference Array Serial Number: Cross-reference the array serial number with data from capacity collection to add the necessary device property to the collected performance data. Avoid Separate Collection Instances or Hosts: We have Single Collector instance.
- When we restart the Load-Balancer Service, these 'Dropping' errors goes away. But then seen again after 15-20 minutes.
Please advise if anyone has some tips or pointers regarding this.
Thanks.
DELL-Sam L
Moderator
•
7.6K Posts
1
November 25th, 2024 22:15
Hello Bhuppi1,
It would be best to open a support case so that we can review the logs to see what is going on.
DELL-Sam L
Moderator
•
7.6K Posts
1
November 22nd, 2024 17:46
Hello Bhuppi1
Have you done the resolution steps that were in KB:000059671?
Bhuppi1
2 Intern
•
130 Posts
0
November 25th, 2024 14:16
@DELL-Sam L Hi. Yes. The KB talks about - Cross-Reference Array Serial Number: Cross-reference the array serial number with data from capacity collection to add the necessary device property to the collected performance data. Avoid Separate Collection Instances or Hosts. And we do have single Collector instance only. Also, Load Balancer on Collector Host is pointing to itself only(local host).
What is more interesting is that these errors are not seen always. Sometime they arrive and most of itme they disappear. Like today we only see this(which seems okay):
INFO -- [2024-11-25 04:13:07 COT] -- AccessorManager::fetchDataFromAccessors(): A total of 0 lines were written to PTF file 'serialnb-from-mgmtip.csv' in 31 ms.
INFO -- [2024-11-25 04:13:07 COT] -- AccessorManager::fetchDataFromAccessors(): A total of 553 lines were written to PTF file 'serialnb-from-fqdn.csv' in 57 ms.
INFO -- [2024-11-25 04:13:07 COT] -- AccessorManager::fetchDataFromAccessors(): A total of 551 lines were written to PTF file 'serialnb-from-hostname.csv' in 49 ms.
INFO -- [2024-11-25 04:13:07 COT] -- AccessorManager::fetchDataFromAccessors(): A total of 556 lines were written to PTF file 'devtype-from-serialnb.csv' in 48 ms.
INFO -- [2024-11-25 04:13:07 COT] -- PropertyTagger::reload(): Reloading configuration...
INFO -- [2024-11-25 04:13:07 COT] -- PropertyTagger::extractRefresh(): Will check configuration for change every 6 HOURS.