It seems that accordingly to the number of Management servers, the requests
with StatsCollector are increasing the same way
It seems that StatsCollector are not load-balanced across the # of
Management servers

Do you have a solution to avoid this charge ?

Regards, Benoit

Le lun. 11 avr. 2022 à 12:40, benoit lair <[email protected]> a écrit :

> Hello,
>
> Nobody has ideas why ACS is scanning so often my Datastores ?
>
> Le ven. 8 avr. 2022 à 16:10, benoit lair <[email protected]> a écrit :
>
>> Hello Folks,
>>
>> I am facing to a strange issue on my xcp-ng cluster with acs 4.16
>>
>> I have 4 ACS Mgmt servers participating to my Cloud installation
>>
>> All of them are contacting every time and very (too) regularly my xcp-ng
>> Pool master, generation some load avg and some Iops
>>
>> From ACS logs i have these entries which occurs very regularly :
>>
>> 2022-04-08 16:03:47,368 DEBUG [c.c.h.o.r.Ovm3HypervisorGuru]
>> (StatsCollector-6:ctx-e75f548d) (logid:09ef2f54) getCommandHostDelegation:
>> class com.cloud.agent.api.GetStorageStatsCommand
>> 2022-04-08 16:03:47,368 DEBUG [c.c.h.XenServerGuru]
>> (StatsCollector-6:ctx-e75f548d) (logid:09ef2f54) We are returning the
>> default host to execute commands because the command is not of Copy type.
>> 2022-04-08 16:03:47,372 DEBUG [c.c.a.m.ClusteredAgentAttache]
>> (StatsCollector-6:ctx-e75f548d) (logid:09ef2f54) Seq 1-5337328508387459309:
>> Forwarding null to 77026952423534
>> 2022-04-08 16:03:50,588 DEBUG [c.c.a.t.Request]
>> (StatsCollector-6:ctx-e75f548d) (logid:09ef2f54) Seq 1-5337328508387459309:
>> Received:  { Ans: , MgmtId: 2955451650215, via: 1(xcp-cluster1-node3), Ver:
>> v1, Flags: 10, { GetStorageStatsAnswer } }
>> 2022-04-08 16:03:50,597 DEBUG [c.c.h.o.r.Ovm3HypervisorGuru]
>> (StatsCollector-6:ctx-e75f548d) (logid:09ef2f54) getCommandHostDelegation:
>> class com.cloud.agent.api.GetStorageStatsCommand
>> 2022-04-08 16:03:50,597 DEBUG [c.c.h.XenServerGuru]
>> (StatsCollector-6:ctx-e75f548d) (logid:09ef2f54) We are returning the
>> default host to execute commands because the command is not of Copy type.
>> 2022-04-08 16:03:50,601 DEBUG [c.c.a.m.ClusteredAgentAttache]
>> (StatsCollector-6:ctx-e75f548d) (logid:09ef2f54) Seq 1-5337328508387459310:
>> Forwarding null to 77026952423534
>>
>> From Xcp-ng, this is launching :
>>
>> "/usr/bin/python /opt/xensource/sm/LVMoISCSISR
>> <methodCall><methodName>sr_scan</methodName><params><param><value>....."
>>
>> I have my /var/log/SMLog which is growing very fastly
>>
>> Do you know why ACS is scanning so fastly my xcp-ng storage ?
>>
>>
>> Thanks a lot for your help and ideas
>>
>> Best regards
>> Benoit Lair
>>
>

Reply via email to